Race, Gender, and Workplace Power

Power defined as ldquocontrol over resourcespeople and thingsrdquo (Wolf and Fligstein

1979) is an essential aspect of social stratifi-cation (Bendix 1956 Braverman 1974Dahrendorf 1959) In this study we focus on onedimension of power authority and control overothers in the workplace Weber ([1914] 1968)conceptualized this dimension of power as aform of ldquolegitimate authorityrdquo because it derivesfrom organizational positions that people occu-py rather than from the people themselves Weconcentrate specifically on how race ethnicityand gender affect the likelihood of attainingsuccessively higher levels of such authority

This issue merits attention for several reasonsDespite significant progress in the socioeco-nomic status of working women and minoritiesdata analyses at the city state national andcross-national levels continue to document raceand gender inequalities in workplace power (seeSmith 2002 for a recent review) Moreover fed-eral reports indicate that discrimination claimspertaining to promotion have risen steadily inrecent years while discrimination claims per-

taining to hiring have declined (Myerson 1997)These patterns suggest that workplace powerconstitutes a central battleground in struggles forequalizing opportunities in coming yearsmdashapossibility fueled by rising expectations ofunprecedented numbers of women and minori-ties now employed in the formal economy

Another reason for investigating racial andgender inequalities in workplace power is thatdespite the popularity of the ldquoglass ceilingrdquometaphor surprisingly little direct researchexists on related assumptions For examplenumerous quantitative studies document ascrip-tive inequalities in workplace power but fewexamine whether these inequalities increase athigher levels of power and none adequatelyaccount for factors that ethnographic researchcontends are important namely the ascriptivecontexts in which positions of power are embed-ded and the networks that supposedly play keyroles in the allocation of these positions to par-ticular individuals Conversely ethnographicresearch is insightful but it only comes from afew select case studies (eg Kanter 1977)

RRaaccee GGeennddeerr aanndd WWoorrkkppllaaccee PPoowweerr

James R Elliott Ryan A SmithTulane University The City University of New York

Survey data support hypotheses regarding differential access to workplace power among

women and minorities relative to white men Specific findings indicate that relative to

white men all groups encounter increasing inequality at higher levels of power but only

black women seem to experience this form of inequality as a result of direct

discrimination Further analysis indicates that network assistance is more a response to

this form of discrimination than an indirect cause Finally analysis shows that most

groups attain power through homosocial reproduction but what differs is the opportunity

to engage in such reproduction wherein white men excel These findings imply that while

women and minorities face lower odds than white men of achieving higher levels of

workplace power the reasons for this disadvantage vary among respective groups and

thus will likely require different remedies

AAMMEERRIICCAANN SSOOCCIIOOLLOOGGIICCAALL RREEVVIIEEWW 22000044 VVOOLL 6699 ((JJuunnee336655ndashndash338866))

1600-ASR 693 filename69303-elliott

To view additional data andor results that sup-plement this article please see the ASR Web site(httpwwwasanetorgjournalsasr2004toc039html)

Direct correspondence to James Elliott TulaneUniversity Sociology Department 220 NewcombHall New Orleans LA 70118 (jretulaneedu) Theauthors thank Charlie Brody Beth Rubin SandersKorenman Neil Bennett and anonymous reviewersfor their helpful comments on prior drafts

Delivered by Ingenta to University Of Oregon Library

Sat 30 Sep 2006 155556

Consequently it remains uncertain just howgeneralizable the claims from these studies areand the selection biases they might reflect

Finally prior quantitative studies on ascrip-tive inequalities in workplace power focus eitheron race or gender but not both (for notableexceptions see McGuire and Reskin 1993Tomaskovic-Devey 1993) This conventionalapproach is problematic because it reinforces theerroneous assumption that racial stratificationand gender stratification are mutually exclu-sive systems that we can somehow sum tounderstand differences among non-white-menBy contrast we view racial ethnic and genderstratifications as having fundamentally similarcauses and how these causes overlap to producedifferent outcomes for different groups remainsa decidedly empirical question In taking thisapproach it is important to move beyond exclu-sively black-white comparisons to considerLatinos not only because Latinos constitutethe largest and fastest growing panethnic groupin US society but also because recent researchindicates that employers of all ethnoracial back-grounds tend to prefer Latinos over blacks whenfilling positions in their organizations (Moss andTilly 2001 Wilson 1996)

In the present study we address these short-comings in prior research through an examina-tion of authority attainment among white blackand Latino men and women Our objectives aretwofold (1) assess the extent to which inequal-ity in workplace power increases among womenand minorities relative to white men at higherlevels of power and (2) examine the mecha-nisms of allocation responsible for this form ofinequality for each group In pursuing theseobjectives we focus specifically on power posi-tions under the supervision of others One rea-son for this focus is that workplace power is notlimited to the upper echelon of Fortune 500 cor-porations it is found in all places of employmentWhile popular reporting on the glass ceilingoften obscures and even trivializes the phenom-enon it is critical to general understandingbecause even seemingly mundane jobs are not leftto run themselves as the lack of autonomy inmost low-skill positions attests Instead peopleare selected to fill positions of power throughoutall levels of the workforce And mid-level posi-tions not only constitute the most common formof legitimate authority but they also representpositions wherein vertical and horizontal inter-

group competition is likely to be greatest amongincumbents of roughly equal credentials

BBAACCKKGGRROOUUNNDD

A popular explanation for ascriptive inequali-ties in workplace power invokes the metaphorof a glass ceiling which was first popularizedin a 1986 Wall Street Journal article thatdescribed barriers women often face as theyclimb corporate ladders In 1995 the UnitedStates Department of Labor issued an officialreport on this subject noting that during theinterim decade observers had extended the termtwo ways first to include racial and ethnicminorities in addition to women and second torefer to all management and decision-makingpositions not just to top-level positions at largecorporations (Federal Glass CeilingCommission 1995iii) A core idea invoked bythe metaphor is that while employers might letwomen and minorities into low positions ofauthority they are much less likely to let theminto high positions that involve greater control

While we do not explicitly test for corporateglass ceilings in this research we do examinewhether women and minorities have an increas-ingly difficult time relative to white menaccessing jobs with greater organizationalpower Recent research along these lines hastaken one of two general approaches The morerestrictive approach views this type of inequal-ity in terms of an absolute barrier that blockswomen and minorities from higher positionsof workplace power because they are womenand minorities (Jacobs 1992 Morrison andGlinow 1990 Reskin and McBrier 2000 Reskinand Ross 1992) Taken literally this perspectiveimplies an invisible barrier below which womenand minorities attain a modest degree of work-place power (eg supervisory authority) andabove which they do not (eg managerial con-trol) A less restrictive approach views increas-ing inequality as a form of disadvantage facingwomen and minorities relative to white menwhich intensifies at higher levels of workplacepower

This second approach to conceptualizingincreasing inequality has two important impli-cations for empirical assessment First inequal-ity is presumed to occur not in a single absolutestep but rather over several steps of increasingmagnitude relative to white men Second and

336666mdashmdashndashndashAAMMEERRIICCAANN SSOOCCIIOOLLOOGGIICCAALL RREEVVIIEEWW



Sat 30 Sep 2006 155556

as such a declining share of women andorminorities in positions of higher power offersnecessary but insufficient evidence of increas-ing inequality Instead sufficient evidencerequires decreasing probabilities of advance-ment relative to white men at higher levels ofpower To illustrate Table 1 depicts nonsup-portive and supportive evidence for increasinginequality in a simple three-level hierarchy(worker supervisor manager) In this table theldquowhite-black ratiordquo is the key statistic becauseit summarizes black menrsquos probability ofadvancement relative to white men at two suc-cessive levels of power first from worker tosupervisor and second from supervisor to man-ager In the panel labeled ldquolack of support forincreasing inequalityrdquo the relative ratio ofadvancement for black men decreases from 25to 20 with movement up the hierarchy By con-trast in the panel labeled ldquosupport for increas-ing inequalityrdquo the relative ratio for black menincreases from 25 to 40 This difference insupportive and nonsupportive evidence forincreasing inequality relative to white menoccurs despite the fact that the share of blackmen decreases with movement up the hierarchyin both panels (see columns labeled ldquo blackmenrdquo)

In the present research we focus on this odds-based criterion for assessing inequality andoperationalize this focus via the followinghypothesis

Hypothesis 1 Women and minoritiesrsquo odds ofadvancement decrease relative to whitemen at higher levels of power

In the most direct test of this hypothesis to dateBaxter and Wright (2000) use a six-level indexof workplace power to examine cross-national

differences in gender inequalities and find noevidence in the United States that womenrsquos like-lihood of advancement relative to men declinesat successively higher levels of power (see alsoYamagata et al 1997) Their small sample sizeshowever prevent the authors from showingincreasing inequality for racial minorities athigher levels of power particularly women ofcolor On this subject studies by Cotter et al(2001) and Morgan (1998) are instructive albeitindirect Both studies use longitudinal data toexamine wage inequality in individual careersbut they reach different conclusions about thepresence of increasing racial and gender disad-vantage relative to white men at later stagesFor example Cotter et alrsquos analysis of data fromthe Panel Survey of Income Dynamics revealsincreasing wage inequality for white and blackwomen relative to white men but not for blackmen From this evidence the authors concludethat increasing inequality in wages in individ-ual careers is more reflective of gender thanracial stratification Morgan (1998) by con-trast uses a single-cohort longitudinal designwith data from the Survey of Natural and SocialScientists and Engineers and a multi-cohortcross-sectional design with data from the 1992Survey of Women and Men Engineers Fromthese analyses Morgan concludes that increasedpay gaps among men and women at later careerstages are more reflective of cohort-replace-ment dynamicsmdashthat is past inequalities work-ing themselves through the systemmdashthanincreasing gender inequality among men andwomen as they progress in their careers1

RRAACCEE GGEENNDDEERR AANNDD WWOORRKKPPLLAACCEE PPOOWWEERRmdashmdashndashndash336677


Table 1 Inequality among Men at Higher Levels of Workplace Power

Men in Power Level Odds of Advancement to Next Level of Power

Power Level White (n) Black (n) Black White Black White-Black Ratio

Lack of Support for Increasing InequalitymdashManager 10 02 17 mdash mdash mdashmdashSupervisor 20 08 29 050 25 21mdashWorker 30 30 50 067 27 251Support for Increasing InequalitymdashManager 20 02 09 mdash mdash mdashmdashSupervisor 20 08 29 100 25 41mdashWorker 30 30 50 067 27 251

Note Data adapted from Baxter and Wright 2000

1 For criticism of Morganrsquos analysis and conclu-sions see Alessio and Andrzejewski (2000) withreply by Morgan (2000)


Sat 30 Sep 2006 155556

Together these two studies provide keyinsights into race gender and employmentinequalities in the United States but they alsoleave us unclear about what to expect in analy-ses of workplace power Some of this uncer-tainty stems from their mixed results and someof it stems from the fact that they analyze wagesnot workplace power The last point is impor-tant because racial and gender differences inworkplace power are an important source ofwage inequality (Halaby 1979 Kluegel 1979Robinson and Kelley 1979 Smith 1997) Thusto clarify this issue it is important not only toexamine these differences in workplace powerdirectly relative to white men but also to exam-ine the mechanisms of allocation commonlypresumed responsible for these differences athigher levels of legitimate authority

MECHANISMS OF ALLOCATION Observers com-monly point to three broad mechanisms respon-sible for increasing ascriptive inequalities athigher levels of workplace power One mecha-nism is direct discrimination which can take twodistinct forms ldquotaste discriminationrdquo in theform of old-fashioned racism and sexism basedon out-group prejudice and antipathy and ldquosta-tistical discriminationrdquo in which employers userace and sex as proxies for assessing potentialproductivity in candidates when they lack otherinformation about the candidates For exam-ple if women generally are less likely to putwork demands above family demands thenemployers might use this easy-to-observe trait(sex) to screen and evaluate managerial candi-dates in favor of men regardless of the (unob-served) work commitment of individual maleand female candidates under reviewResearchers typically identify these forms ofdiscrimination using a residual (or ldquonet gaprdquo)strategy in which disadvantage relative to(white) men is evaluated after statistically con-trolling for job-relevant factors such as educa-tion experience and employment context (egWright Baxter and Birkelund 1995) If relativedisadvantage for the minority andor femalegroup in question increases with movement upthe power hierarchy increasing disadvantage ispresumed to exist and to be the result of directdiscrimination although the precise form(ldquotasterdquo or ldquostatisticalrdquo) is difficult to differen-tiate

Another broad mechanism presumed to gen-erate increasing inequalities in workplace powerinvolves a more indirect process namely theexclusion of women and minorities relative towhite men from networks that regulate accessto information opportunities and resourcesneeded to advance in the workplace Researchon this subject generally shows that work-relat-ed networks help workers gain skills acquirelegitimacy and climb promotional ladders(Bridges and Villemez 1986 Campbell andRosenfeld 1985 Podolny and Baron 1997) andthat these resources are important because mostemployeesrsquo job training and career develop-ment come from informal instruction ratherthan continuing education and explicit on-the-job training (United States Bureau of LaborStatistics 1996)

Additionally recent research by McGuire(2002) indicates that exclusion of women andminorities can occur within job-related net-works as well as outside such networks In herstudy of over a thousand financial-servicesemployees McGuire found that even whenblack and white women held jobs in which theyhad personal ties to the same types of higher-level employees as white men they receivedsignificantly less work-related help from theseties than similarly situated white men McGuireconcludes that this discrepancy arises becausenetwork members are less likely to invest inwomen than (white) men as a result of culturalbeliefs that rank women below men The impli-cation is that workers not just employers userace and gender to rank network members andthis ranking influences the type and amount ofassistance available to members of differentgroups Although McGuire focused primarily onnetwork assistance within organizations suchassistance can also be useful in gaining positionsof power across organizations by providinginformal training and contacts necessary toopen doors to other employers On the basis ofthis rich body of research we advance the fol-lowing general hypothesis

Hypothesis 2 Women and minoritiesrsquo odds ofnetwork assistance decrease relative towhite men at higher levels of power

A corollary to this hypothesis is that women andminorities often rely more on education andexperience relative to white men to ldquobreakintordquo higher levels of power often having to




Sat 30 Sep 2006 155556

ldquoout-credentialrdquo white-male counterparts tocompensate for their relative lack of networkassistance Researchers sometimes characterizethese distinct modes of advancement as ldquospon-soredrdquo and ldquocontestrdquo mobility regimes respec-tivelymdashthe first pertaining to network utilizationamong white men the second to skills-basedcompetition among women and minorities (egMueller Parcel and Tanaka 1989) While theldquocontestrdquo regime might seem fair in its empha-sis on objective skills-based traits it can leadto relative disadvantage for women and minori-ties for a couple reasons

First given practical limits to educationalattainment and experience it becomes increas-ingly difficult to ldquoout-credentialrdquo other workerswith movement up workplace power hierar-chies leaving network assistance still a key fac-tor in determining who will advance and whowill not Second as women and minorities moveup organizational chains of command their out-group or ldquootherrdquo status often becomes moreevident leaving them more susceptible to infor-mal processes of exclusion and assessment assymbols of an ldquootherrdquo category rather than asindividuals We examine this corollary set ofassumptions via the following hypothesis

Hypothesis 3 Women and minoritiesrsquo relianceon education and experience intensifiesrelative to white men at higher levels ofpower

Support for this hypothesis would mean thatregression coefficients for education and expe-rience in equations predicting workplace powerwould be statistically significant and increas-ingly large for women and minorities relativeto white men with upward movement in work-place power

A third mechanism presumed to generateincreasing inequality in workplace power isascriptive dissimilarity with superiors who over-see higher positions of power In perhaps the bestknown discussion of these dynamics Kanter(1977) contends that with movement up orga-nizational hierarchies power positions becomecharacterized by increasing uncertainty inter-dependence and necessity for rapid accuratecommunication about murky matters such asrelations between organizational means andends and criteria for performance evaluationThese job characteristics in turn place a pre-mium on discretion and trust among workers

selected to advance up the power hierarchy Oneway that higher-level managers try to maximizethese traits and impose greater predictability onan otherwise uncertain environment is to main-tain relative social homogeneity among indi-viduals they select to f ill positions oforganizational power beneath them The under-lying idea is that communication discretionand trust are facilitated by social similarityHigher-level managers prefer this type of rela-tionship over the strain of dealing with peoplewho are different when higher degrees of legit-imate authority are at stake2

Kanter refers to this process generally asldquohomosocial reproductionrdquo because it tends toreproduce the social characteristics of organi-zational power structures over successive gen-erations of workersmdashan idea that traces back toWilbert Moorersquos concept of ldquobureaucratic kin-ship systemsrdquo (Moore 1962) Because whitemen have historically held the reins of power inUS workplaces they benefit most from theseuniversal tendencies for in-group favoritism asthey move up organizational hierarchies creat-ing increasing inequality for out-group mem-bers We examine this mechanism of allocationvia the following hypothesis

Hypothesis 4 White menrsquos odds of having self-similar superiors increase relative towomen and minorities at higher levels ofpower

This hypothesis assumes that in most organi-zational contexts homosocial reproductionoperates in a vertical fashion with superiorsselecting individuals like themselves to fillpower positions below them rather than in ahorizontal fashion with superiors selecting indi-viduals like themselves to fill power positionsalongside them To illustrate consider a simple



2 This argument is consistent with recent researchin ldquoorganizational demographyrdquo and ldquonew econom-ic sociologyrdquo that claims the following (1) peopletend to make sense of their social worlds by catego-rizing others into in-groups and out-groups and(2) this normal information processing occurs large-ly outside conscious control biasing treatment ofothers because of race gender and other discernibletraits of group membership (see Pfeffer 1983 Reskin2002) See Tsui and OrsquoReilly (1989) for specifics onimportance of ascriptive similarity for performanceevaluation in superior-subordinate dyads


Sat 30 Sep 2006 155556

three-level firm in which managers are Assupervisors are Bs and workers are CsHypothesis 4 implies the following conditions(1) As will tend to fill openings for B with indi-viduals like themselves (2) likewise Bs willtend to fill openings for C with individuals likethemselves (3) the first tendency will be greaterthan the second tendency because more poweris at stake and (4) this process benefits whitemen more than other groups because white menare the group best positioned to benefit from in-group favoritism at higher levels of power

ADDITIONAL CONSIDERATIONS Our hypothe-ses reflect common assumptions about increas-ing ascriptive inequality at higher levels ofworkplace power and the mechanisms that pro-duce it However these assumptions remainopen to several criticisms First thus far empir-ical support for the idea that white men unique-ly benefit from network assistance comes fromindirect assessments (eg Mueller Parcel andTanaka 1989 Wilson 1997) These assessmentstypically use regression analysis to show thatjob-relevant factors such as education and expe-rience are more predictive of authority attain-ment by women and minorities than by whitemen leaving the latter with greater unexplainedvariance Researchers then interpret this greaterunexplained variance for white men as evidenceof the relative importance of unobserved mech-anisms including network assistance for whitemenrsquos power attainment This interpretation isa logical but assailable inference in the absenceof direct measurement of network assistance

Second researchers commonly assume thatwhite men exert control over most if not allpositions of workplace power particularly thosewith greater legitimate authority While thisassumption might be true in relative termswhite menrsquos control over US workplaces is notabsolute Third and relatedly researchers havenever demonstrated empirically tendenciestoward homosocial reproduction across groupsand organizational contexts Thus it remainsuncertain if the process Kanter describes is uni-versal or if white men are more likely to engagein this homosocial reproduction than womenand minorities Both points are important polit-ically as well as sociologically because theyspeak to how women and minorities behavewhen they rise to positions of power andwhether once in these positions their behavior

is likely to open doors to minority and womenrsquospower attainment in the future

Together these additional considerations raisethe possibility that how members of differentrace and gender groups advance up workplacepower hierarchies depends not just on their ownrace and gender but also on the race and gen-der of those overseeing the power positions inquestionmdashascriptive similarity with superiorsmight not be an additive factor but one thatconditions how other factors operate This pos-sibility suggests that a key difference betweenwhite men and other groups is that white menby virtue of being dominant nearly always riseto power under ldquosimilar othersrdquo whereas womenand minorities generally take two tracks theyadvance under white men or they advance undersimilar others Which track women and minori-ties take in turn might influence which factor(network assistance or human capital) is likelyto be most important for advancementExtrapolating from white menrsquos experiencesunder similar others we might reasonablyexpect network assistance to be more importantfor advancement among women and minoritieswho work under ascriptively similar superiorsthan those who work under ascriptively dis-similar (eg white-male) superiors whereas theopposite would be true of education and expe-riencemdashit would be less important for advance-ment under ascriptively similar superiors thanunder ascriptively dissimilar superiorsAdditional consideration of these potential con-ditional effects of relative superior ascriptionmotivates our final hypothesis

Hypothesis 5 Women and minorities underself-similar superiors rely more on networkassistance and less on human capital toattain higher positions of power thanwomen and minorities under self-dissimilarsuperiors

DDAATTAA

Data for our study come from the Multi-CitySurvey of Urban Inequality (MCSUI) which isa multistage stratified area-probability sampleof white and minority respondents in AtlantaBoston and Los Angeles conducted during1992ndash1994 a time of local and national eco-nomic expansion The survey was administeredthrough face-to-face interviews that lasted




Sat 30 Sep 2006 155556

approximately two hours Race and ethnicity ofrespondents and interviewers were matched tominimize well-known race-of-interviewereffects (see Johnson Oliver and Bobo 1994)3

For our purposes the advantages of theMCSUI are fourfold First its multiethnic sam-ple allows us to examine Latino men and womenin addition to whites and blacksmdashthis is a nov-elty in quantitative research on workplace powerin the United States4 Second the MCSUI pro-vides data on multiple successive levels ofworkplace power and on the race and sex ofimmediate superiors in addition to data abouthuman capital and employment context Thisinformation allows us to test for evidence ofincreasing inequality at higher levels of power(see Table 1) and the extent to which this evi-dence might differ according to the relative andabsolute characteristics of superiors involvedThird the MCSUI provides data about howworkers acquired their jobs (eg through for-mal searches or with network assistance) whichallows us to test assumptions about the impor-tance of such assistance for advancement direct-ly Finally the MCSUI draws from a diverse setof metro economies which while perhaps notrepresentative of the US labor force as a wholecollectively draws from a wide range of labormarket processes and contexts that can influencethe distribution of workplace power

While these features make the MCSUI thebest large-scale dataset available for the kind ofanalyses we wish to conduct it is not withoutweaknesses First our indicator of workplacepower is basic we measure broad differences

among workers supervisors and managersWhile these differences are important they donot allow us to identify individualsrsquo detailedpositions on the corporate latter as priorresearch on gender inequality has sought to do(eg Baxter and Wright 2000) Second infor-mation on network assistance is self-reportedand may reflect group differences in the likeli-hood of such reporting in addition to differencesin actual networking effectiveness Finally thereis no way to determine entirely whether evidenceof homosocial reproduction is a product of in-group preference or the result of sex and racesegregation that effectively limits candidates toin-group members only

With these limitations in mind we select onlycivilian labor force participants between theages of 21 and 64 who are not self-employedWe focus on non-self-employed workersbecause we are interested in power distributionamong paid employees not among individualswho made a qualitative shift from employee toemployer Second in tests of ldquohomosocial repro-ductionrdquo and other allocation processes speci-fied in Hypotheses 2ndash5 we restrict our sampleto workers who report having immediate super-visors which means that our results cannot begeneralized to the very tops of organizationalhierarchies where power holders have no supe-riors We believe that this focus is justified bythe fact that such mid-level positions comprisethe overwhelming majority of workplace powerpositions in the United States and representstrata where competition for legitimate author-ity among individuals of different races eth-nicities and genders is likely to be mostcommon5

VVAARRIIAABBLLEESS

In this section we discuss our operational def-initions for key variables starting with thedependent variable legitimate authority (ieworkplace power) All variables are listed inTable 2 with subsample means and standarddeviations



3 The MCSUI also includes data from Detroitwhich we omit because of a lack of information onkey labor market variables (eg level of workplacepower) For limitations of using cross-sectional datato study increasing inequality in workplace powerattainment see Baxter and Wright (2000)

4 We refer to men of Hispanic descent as ldquoLatinordquoand women of Hispanic descent as ldquoLatinardquo Thevast majority of these groups in our sample comesfrom Central America and the Caribbean Withinthese subsamples Mexicans comprise the largestgroup (39 percent) with most residing in LosAngeles Puerto Ricans and Dominicans comprise thenext largest groups (20 percent and 13 percent respec-tively) with most residing in Boston Sample limi-tations preclude us from analyzing these specificgroups directly

5 Exclusion of respondents with no immediatesuperior reduced our sample by 93 percent Furtherinvestigation indicates that excluded respondentstended on average to exhibit the same odds of powerattainment as respondents in our sample


Sat 30 Sep 2006 155556

WWOORRKKPPLLAACCEE PPOOWWEERR

Employed respondents in the MCSUI wereasked three closed-ended questions commonlyused in survey research on workplace power (a)Do you supervise another employee who isdirectly responsible to you (b) Do you influ-ence or set the rate of pay received by others(c) Do you have the authority to hire or fire oth-ers We use responses to these questions toclassify employees into one of three hierarchi-cal levels

0 = worker (ldquonordquo to a b and c) (1)

1 = supervisor (ldquoyesrdquo only to a) (2)

2 = manager (ldquoyesrdquo to a and ldquoyesrdquo to b or c) (3)

We combine questions b and c to help constructa single indicator of managerial status for sev-eral reasons first all respondents who answeredldquoyesrdquo to b or c also answered ldquoyesrdquo to a sec-ond the correlation between b and c in ourpooled sample is quite high (r = 543 p lt 0001)with 70 percent of respondents answering ldquoyesrdquoto b also answering ldquoyesrdquo to c and third b andc are conceptually similar in that they denotecontrol over the distribution of organizational

resources as well as people To test for increas-ing inequality we use multinomial regressionanalysis to compare odds of being a supervisorversus being a worker (1 versus 0) with odds ofbeing a manager versus being a supervisor (2versus 1) If the second set of odds is statisticallysignificant and larger than the first set of oddswe conclude the existence of increasing inequal-ity for the group in question

This operationalization of workplace poweris preferable to an occupationally based meas-ure because legitimate authority extends wellbeyond the boundaries of officially recognizedmanagerial occupations In the MCSUI forexample only five percent of employees withsupervisory or managerial status work in a man-agerial occupation (Census Occupation Codes23ndash42) Moreover supervisory status is asso-ciated with 193 distinct 3-digit CensusOccupation Codes and managerial status isassociated with 133 such codes To assess thevalidity of our three-level dependent variablewe computed means and t-tests for factors com-monly associated with movement up the powerhierarchy Results indicate monotonic and sta-tistically significant differences from one levelto the next along multiple dimensions of



Table 2 Variables Used in Analysis

Men Women

XX

Managerial status (01)Supervisory status (01)Ascriptively similar superior (01)Network assistance (01)Education years ofWork experience total yearsPrior job-specific experience (01)Years with employer Logn (number of workers mdashin establishment)Public sector (01)Logn (work hoursweek)Professionaltechnical mdashoccupation (01)Craftrepair occupation (01)Service occupation (01)Sales and clerical occupation (01)Married (01)Children in household (01)

Note Data shown as mean with standard deviation in parenthesesa ldquoLatinordquo and ldquoLatinardquo refer to men and women of Hispanic descent respectively

White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556

socioeconomic status compensation job com-plexity education experience and ascriptivejob context (see Table A1 on the ASR Web sitesupplement httpwwwasanetorgjournalsasr2004toc039html) These differ-ences combine with prior research using sim-ilar questions to support the empirical validityof our parsimonious measure of workplacepower

KKEEYY FFAACCTTOORRSS IINN AALLLLOOCCAATTIIOONN OOFF WWOORRKKPPLLAACCEE

PPOOWWEERR

For Hypotheses 2 and 5 we operationalizenetwork assistance as a dummy variable basedon the following question ldquoDid you find yourjob through friends or relatives other peoplenewspaper ads or some other wayrdquo If therespondent reported using a personal contactthe interviewer collected information aboutthe mode of assistance Using this informationwe define network assistance conservatively ascases in which a job contact talked to theemployer on the respondentrsquos behalf provid-ed a reference or hired the respondent Weexclude contacts who merely passed alonginformation about the job because this modeof network assistance is considered secondaryto workplace power distribution which empha-sizes processes of sponsorship over mereinformation flow Because the MCSUI askedabout the use of job contacts only amongrespondents who reported actively searchingfor jobs we also define workers who enterednew jobs without an active search as receivingnetwork assistance The logic here is that inthese cases job networks brought the employ-er to the respondent rather than vice versa andthat both scenarios constitute a ldquostrongrdquo formof network assistance (see Granovetter 1995)

For Hypotheses 1 3 and 5 we operational-ize four indicators of human capital We meas-ure education as the total number of years offormal schooling We also include three indi-cators of labor force experience We measuretotal work experience as the number of yearsthat a respondent was employed formally sincefirst leaving full-time school We measureprior job-specific experience as a simpledummy indicator (0 = no 1 = yes) based on thequestion ldquoDid you have any previous experi-ence in this type of job excluding schoolingbefore you were hiredrdquo Finally we measure

organizational tenure as the number of yearsthat the respondent reports being employedwith his or her respective employer

For Hypotheses 4 and 5 we operationalize adummy indicator that is set to 1 if the respon-dent works under an ascriptively similar supe-riormdashthat is someone of the sameraceethnicity and sexmdashand 0 if the respondentworks under an ascriptively dissimilar superi-or This indicator is based on three nested ques-tions ldquoDo you have an immediate supervisoron your job to whom you are directly respon-siblerdquo ldquoWhat is your immediate supervisorrsquosrace or ethnic originrdquo ldquoIs your immediatesupervisor a man or a womanrdquo We interpreta value of 1 for this variable as indicating a pat-tern consistent with homosocial reproductionOur reasoning is that even if a respondentrsquosimmediate superior did not have sole respon-sibility for filling the respondentrsquos currentposition he or she almost certainly providedmeaningful input In supplemental analyses wealso use information on the race and ethnici-ty of coworkers which we discuss later in thisarticle

CCOONNTTRROOLL VVAARRIIAABBLLEESS

To isolate hypothesized relationships weinclude several job-relevant factors identifiedin prior research as being important covariatesof workplace power One such factor is estab-lishment size which reflects the vertical andhorizontal complexity of the organization inquestion and the number of power positionslikely to be available to respective employeesWe operationalize this factor as the naturallog of the number of employees that the respon-dent reports working at his or her establish-ment We also include a dummy indicator forpublic sector (0 = private sector 1 = public sec-tor) because prior research indicates that therelative disadvantage that women and minori-ties face in advancing up workplace powerhierarchies tends to be lower in public than pri-vate settings owing to more egalitarian hiringpractices and bureaucratic protocols foradvancement in the former (Fernandez 1975Wilson 1997)

Another factor related to workplace poweris time spent at work We operationalize thisvariable as the natural log of the average num-ber of hours worked per week We log this




Sat 30 Sep 2006 155556

variable to compress higher values becausework hours that extend beyond normal full-time status are more likely to be the result ofbeing a manager than a determinant of becom-ing a manager We also include a four-catego-ry indicator of occupational location Thisindicator is based on 1990 Census OccupationCodes and includes the following categories(1) professional and technical occupationswhich include officially titled managers andsupervisors (2) craft and repair occupations(3) service occupations and (4) clerical andsales occupations (reference category)6

Consistent with prior research on powerattainment (Baxter and Wright 2000Rosenfeld van Buren and Kalleberg 1998Wright Baxter and Birkelund 1995) we usethis crude indicator of occupational location tominimize problems of circularity that wouldresult if we used more refined categories or asingle continuous measure of occupationalstatus such as the socioeconomic index Theproblem with these more refined measures ofoccupational location is that they would betoo closely tied conceptually and empiricallyto our dependent variable of workplace powerto be included as compositional controls

In supplemental analyses we also includetwo indicators of family status currently mar-ried (0 = no 1 = yes) and children in the house-hold (0 = no 1 = yes) Researchers oftenassume that marriage and parenthood correlatenegatively with power attainment amongwomen who have historically been more like-ly than men to sacrifice employment mobili-ty for domestic responsibilities

RREESSUULLTTSS

TTEESSTTIINNGG FFOORR IINNCCRREEAASSIINNGG DDIISSAADDVVAANNTTAAGGEE AANNDD

DDIIRREECCTT DDIISSCCRRIIMMIINNAATTIIOONN AA ldquoldquoNNEETT GGAAPPrdquordquoAAPPPPRROOAACCHH

Hypothesis 1 states that women and minoritiesfind it increasingly difficult to advance relativeto white men at higher levels of workplacepower To test this hypothesis we estimate sev-eral nested multinomial regression equationsthat predict employment at successive levels ofworkplace power focusing specifically onchanges between worker-versus-supervisor andsupervisor-versus-manager comparisons Weinterpret increasing inequality or disadvantagespecified by Hypothesis 1 as one in which agrouprsquos manager-versus-supervisor coefficientis negative statistically significant and largerin magnitude than its supervisor-versus-work-er coefficient This pattern would imply thatthe group in question finds it more difficult rel-ative to white men to advance from supervisorto manager than from worker to supervisorResults from these analyses appear in Table 3

Model 1 estimates the ldquogross gaprdquo in author-ity for each group with no statistical controlsComparisons of coefficients in rows 1a and 1boffer initial support for the increasing-inequal-ity hypothesis for every group except whitewomen at the 05-level (two-tailed test) a one-tailed test however would include whitewomen As an interpretative example consid-er black men Results from Model 1 indicate nostatistical difference between black men andwhite men with respect to being a supervisorversus a worker (exp[ndash0027] = 097) Howeverthe anti-log of ndash0792 indicates that black menare only 045 times or about half as likely aswhite men to be managers as supervisors Thuswithout statistical controls we conclude that apattern of increasing inequality exists for blackmen relative to white men based on the judg-ment that 045 differs significantly from 097

A more rigorous test for the presence ofincreasing inequality includes statistical controlsfor nondiscriminatory factors associated withworkplace power For this test we fit two addi-tional models In Model 2 we add human-cap-ital factors (years of education total workexperience prior job-specific experience andemployer tenure) Comparing results acrossModels 1 and 2 indicates that these factorsexplain most of the increasing-inequality effect



6 Concern that occupational location is an endoge-nous variable is minimal here because we use broadcategories and because most workplace power isachieved outside officially recognized ldquomanagerrdquoand ldquosupervisorrdquo occupations Still we reestimatedall our models using a 10-category industrial typol-ogy in place of our occupation controls Results werenearly identical in both cases except estimates of theeffects of educational attainment are slightly lower inmodels with occupation rather than industry con-trols Thus our estimates of educational attainmentwith occupation controls provide a comparativelyconservative estimate of this variablersquos effect onauthority attainment


Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556

among Latinos and white women but notamong black men black women and Latinas(ie coefficients for these groups in Row 2b arestill negative statistically significant and larg-er than coefficients in Row 2a) Next in addi-tion to these human-capital factors we addcontrols for employment context in Model 3(establishment size publicprivate sector occu-pational location and hours worked per week)Comparing results across Models 2 and 3 indi-cates that these factors explain most of theincreasing-inequality effect among black menand Latinas but not among black women

These findings indicate that although eachmajor race-sex group exhibits a pattern ofincreasing inequality relative to white menonly black women exhibit this pattern after con-trolling for variation in human capital andemployment context suggesting that they suf-fer more than other groups from direct dis-crimination Specific calculations from Model3 indicate that net of the full set of controlsblack women are just as likely as white men tobe supervisors as workers (exp[ndash0042] = 096)but they are only 043 times as likely as whitemen to be managers as supervisors(exp[ndash0843]) The implication for the remain-ing groups (black men Latinos white womenand Latinas) is not that they are free fromincreasing inequality (relative to white men)Instead the implication is that this inequality ismore attributable to indirect processes affectinghuman capital attainment and assignment todifferent employment contexts than it is to directdiscrimination It is worth noting however thatin Model 3 coefficients for all groups (exceptLatinos) are in the hypothesized direction

A potential criticism of these results is thatwomen voluntarily make themselves less avail-able for promotion to save time and energy forfulfilling traditional wife and motherhood rolesat home (ie increasing inequality for womenis attributable to self-removal from higher lev-els of power not discrimination) To explorethis self-removal issue we estimated a fourthmultinomial regression equation that includedthe full set of controls present in Model 3 plusmain-effect and group-specific interaction termsfor marriage (yesno) and presence of childrenin the household (yesno) If self-removal isoperating we would expect womenrsquos relativegap in power attainment to be larger in com-parisons among married parents than in com-

parisons among single nonparents Results ofour supplemental analysis lend little support tothis expectation (see Table A2 on the ASR Website supplement httpwwwasanetorgjournalsasr2004toc039html) a chi-squaredtest indicates no significant improvement inmodel fit over Model 3 none of the respectivefamily-status interaction terms are statisticallysignificant at the 05-level and appropriate cal-culations reveal that the strongest evidence ofincreasing inequality among black women rel-ative to white men occurs in comparisonsamong single nonparents (the family status withthe least traditional self-removal pressures) notnonmarried parents (the family status with themost traditional self-removal pressures)

These findings affirm support for our con-clusions regarding black women from Table 2Moreover they are consistent with the conclu-sion by Wright et al (1995) that family statusaccounts for little of the observed gender gapsin workplace power in the United States Thesefindings are also consistent with those ofCassirer and Reskin (2000) who found thatnet of job-relevant factors men and womenhave equal aspirations of promotion regard-less of family status

TTEESSTTIINNGG DDIIFFFFEERREENNCCEESS IINN AALLLLOOCCAATTIIOONN

PPRROOCCEESSSSEESS AANN ldquoldquoIINNTTEERRAACCTTIIOONNrdquordquo AAPPPPRROOAACCHH

To test Hypotheses 2ndash4 we take an ldquointeractionapproachrdquo The logic behind this approach runsas follows To test if specific factors operate dif-ferently for white men than other groups wefirst establish how these factors operate forwhite men by estimating a ldquomain effectsrdquo modelseparately for them Next to test the extent towhich specific factors vary in their effectsbetween white men and each of the ldquootherrdquogroups we pool each separate ldquootherrdquo groupwith white men and estimate a model withappropriate interaction terms When a coeffi-cient for an interaction term is determined to bestatistically significant at the 05 level we con-clude that the interaction effect under review isstatistically different from zero otherwise weconclude that the interaction effect occurred bychance In these analyses negative and statis-tically significant coefficients at higher levelsof workplace power for indicators of networkassistance and self-similar superiors wouldaffirm Hypotheses 2 and 4 respectively By




Sat 30 Sep 2006 155556

contrast positive and statistically significantcoefficients for human-capital factors at high-er levels of power would affirm Hypothesis 3

For these and remaining regression analy-ses we restrict our sample to respondents whochanged jobs within five years of the surveyeither within the same organization or througha change in employer We impose this restric-tion because the MCSUI collected data aboutnetwork assistance only from recent job chang-ers in order to maximize measurement reliabil-itymdasha common practice in studies of jobnetworking (see Granovetter 1995) As a resultof this restriction we minimize differences inpower attainment that linger from past person-nel practices and maximize differences result-ing from contemporary practices that isbehavior most likely to be still in operationtoday

Results of these regression analyses appearin Table 4 As a point of comparison we startwith the baseline equation for white men Hereresults indicate that net of background factorsthe chief variable distinguishing supervisorsfrom workers is employer tenure Specificallyresults imply that for every additional year withan employer a white manrsquos odds of movingfrom worker to supervisor increase an averageof 5 percent (exp[0054]) This factor howev-er exerts little additional effect on moving fromsupervisor to manager Instead the key factorhere appears to involve ascriptive similaritywith higher-level superiors Specifically resultsindicate that white men are twice as likely toadvance from supervisors to managers whenthese managerial positions are overseen by whitemen than when they are overseen by ascrip-tively dissimilar superiors (exp[0706] = 203)This finding suggests that as white men moveup workplace power hierarchies they benefitincreasingly from practices of homosocial repro-duction By contrast the nonsignificant coeffi-cients for network assistance do not imply thatsuch assistance is unimportant but rather thatit is equally common among white men at alllevels of the power hierarchymdasha constant can-not explain a variable outcome such as work-place power Similar nonsignificant findingsresult for education and experience

The remaining results in Table 4 test whetherthese processes of advancement identified forwhite men differ significantly from those expe-rienced by the other groups in our study For

these tests we report coefficients for interactionterms from the respective pooled-equations esti-mated with white men For example in thepooled equation for black men the coefficient0001 (p gt 05) for ldquoyears with employerrdquo refersto the interaction term ldquoyears with employer timesblack manrdquo (with white men as the comparisongroup) The fact that this coefficient is statisti-cally nonsignificant at the 005 level net ofother factors implies that there is no statisticaldifference between black men and white menalong this dimension of authority attainment netof other factors both groups rely approximate-ly equally on organizational tenure to advancefrom worker to supervisor Similar conclusionsobtain for all other measures in the modelNotably the statistically nonsignificant coeffi-cients for self-similar superiors suggest thatblack men and white men rely approximatelyequally on processes of in-group favoritism toadvance from supervisor to manager status(ndash0769 p = 22)7 In general results for blackmen in Table 4 imply that they move up work-place power hierarchies much the same waythat white men do using organizational tenureto advance from worker to supervisor and usingin-group favoritism to advance from supervisorto manager These findings offer no support forHypotheses 2ndash4 and instead imply a set ofldquoseparate but parallelrdquo processes of authorityattainment for black men relative to white men

Results for Latinos reveal much the samepattern with one exception Organizationaltenure plays an even stronger role amongLatinos in advancing from worker to supervi-sor than it does among white men Appropriatecalculations from the full set of coefficients(not shown) indicate that whereas white menreceive a 5-percent bonus for each additionalyear of organizational tenure Latinos receive an18-percent bonus No other factors differ sig-



7 For example to estimate the effect of having aself-similar superior among black men we would sumthe coefficient for having a self-similar superior withthe coefficient for being a black man (as opposed toa white man) with the coefficient for the interactionof these two factors This calculation yields a valueof ndash465 (757 + ndash457 + ndash765) This value comparesto a value of 757 for white men The interactioncoefficient of ndash765 (SE of 623) indicates that thisdifference is statistically insignificant at the 05 level


Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556

nificantly from white men including reliance onascriptively similar superiors to advance fromsupervisor to manager status Thus as withblack men we find little support for Hypotheses2ndash4 among Latinos

Results for women depict different scenariosFor white women f indings indicate thatadvancement from worker to supervisor occursmuch the same way as it does for white menHowever advancement from supervisor to man-ager occurs much less often under ascriptivelysimilar superiors Calculations from the full setof coefficients (not shown) indicate that whitewomen are three times more likely than whitemen to break into managerial positions underascriptively dissimilar superiors 86 percent ofwhom in our sample are white men This highprevalence of out-group status relative to (most-ly white-male) superiors may help to explainwhy in the face of the weakest statistical evi-dence for increasing inequality of any group inour study many observers still insist that suchinequality exists for white women if whitewomen are increasingly likely to work underwhite men as they advance up workplace powerhierarchies there may be both real and per-ceived obstacles to further advancement to unsu-pervised top-level positions These resultsaffirm Hypothesis 4 for white women but notHypotheses 2 and 3

Results for Latinas are similar to those forwhite women except ascriptive similarity withsuperiors appears even less effective relative towhite men for assuming higher positions ofpower In fact the likelihood of Latina super-visors and managers having ascriptively simi-lar superiors is so low that the effects of thisvariable on power attainment cannot be reli-ably calculated (In our sample 0 of the 20Latinas with manager status report an ascrip-tively similar superior and only 7 of the 51Latinas with supervisory status report an ascrip-tively similar superior) The implication is thatLatinas almost always break into power posi-tions under dissimilar superiors which likelylimits their odds of further advancement ifpractices of homosocial reproduction are oper-ating As with white women these results affirmHypothesis 4 but not Hypotheses 2 and 3

The final and perhaps most surprising set ofresults occur among black women Contrary toHypothesis 2 results indicate that network assis-tance is increasingly effective among black

women relative to white men for moving intohigher positions of power Appropriate calcu-lations from the full set of coefficients (notshown) indicate that whereas white men rely onnetwork assistance almost equally at all levelsof workplace power the odds of black womenadvancing from workers to supervisors increase39 percent when they receive network assis-tance and the odds of black women advancingfrom supervisors to managers increase 500 per-cent when they receive network assistanceThese findings suggest that instrumental net-work assistance can be an important responseto discrimination rather than simply an indirectcause Further investigation of the data revealthat black women most often rely on black mento assist them in attaining managerial positionsThree-quarters of the time these men are friendsor relatives and nearly two-thirds of the timethey also work for the employer in question Inour sample such assistance occurs most com-monly among registered nurses sales repre-sentatives and secretaries in predominantlyblack work settings

Overall then results in Table 4 offer no sup-port for Hypotheses 2 and 3 and occasionalsupport for Hypothesis 4 regarding patterns ofhomosocial reproduction Notably this supportfor Hypothesis 4 is countered by the finding thatnet of other factors black men Latinos andblack women receive roughly the same relativebenef it from homosocial reproduction inadvancing up the workplace power hierarchy aswhite men

AA CCLLOOSSEERR LLOOOOKK AATT HHOOMMOOSSOOCCIIAALL

RREEPPRROODDUUCCTTIIOONN

A potential criticism of our test of homosocialreproduction (Hypothesis 4) in Table 4 is thatit relies on a measure of ascriptive similaritywith superiors rather than a measure of ascrip-tive similarity with coworkers8 This criticismbuilds on an alternative interpretation ofhomosocial reproduction that understandspower holders as reserving power positionsalongside rather than under themselves forin-group members To test this alternative inter-pretation we constructed a dummy indicator for



8 We thank an anonymous reviewer for raising thispoint


Sat 30 Sep 2006 155556

working with rather than under ascriptivelysimilar coworkers We constructed this indica-tor from two sources of information First weidentified the racial majority if one existed ofcoworkers from the MCSUI question ldquoWhat isthe race and ethnicity of most of the employ-ees doing the kind of work you do at the placewhere you workrdquo Next lacking similar infor-mation about the gender of coworkers we usedmetropolitan-level data from the 1990 5Public Use Micro Samples (PUMS) to identi-fy the locally dominant sex for each of theroughly 500 detailed Census Occupation Codesin each metropolitan area We then appendedthis information to the MCSUI and created analternative indicator of homosocial reproduc-tion that takes a value of 1 if a respondentworks with mostly coethnic coworkers in anoccupation that locally consists of mostlysame-sex workers otherwise the indicatortakes a value of 0 We then reestimated theequations reported in Table 4 substituting thishorizontal indicator of homosocial reproductionfor our original vertical indicator of having anascriptively similar superior

Results of this supplemental analysis (seeTable A3 on the ASR Web site supplementhttpwwwasanetorgjournalsasr2004toc03html) indicate only one substantive changefrom findings reported in Table 4 Importantlythis change is that white men do not benefitfrom patterns of (horizontal) homosocial repro-duction as they move up workplace power hier-archies net of other factors this finding isconsistent with similar race-only analyses inprior research (Smith and Elliott 2002)Relatedly white and Latina women no longerdiffer significantly from white men along this(horizontal) dimension of homosocial repro-duction One implication of these findings isthat when generalizing about power attainmentacross a wide array of work settings homoso-cial reproduction is perhaps better conceptual-ized in terms of ascriptively similar superiorsregulating access to power positions beneaththem rather than in terms of ascriptively similarsuperiors regulating access to power positionsalongside them The opposite conceptualizationmight be more valid empirically when gener-alizing about top-level positions in large cor-porations of the type Kanter (1977) studied

Another consideration with respect tohomosocial reproduction is that it actually

derives from the product of two distinct rates(1) the relative opportunity to practice homoso-cial reproduction (ie how often group mem-bers are in positions to fill power positionsbeneath themselves) and (2) the rate ofhomosocial reproduction among group mem-bers given the opportunity (ie how often groupmembers select in-group members to fill thepower positions they oversee) To examine thesetwo rates and their product we use our origi-nal (vertical) indicator of homosocial repro-duction to examine how often different levelsof power are overseen by respective groups andhow often these groups appear to select otherin-group members to fill positions of powerimmediately below them Results appear inTable 5

Column 1 of Table 5 provides informationabout the relative opportunity to practicehomosocial reproduction Unsurprisinglyresults indicate that white men have the great-est opportunity to practice homosocial repro-duction and this opportunity increases at higherlevels of power For example results show that59 percent of manager positions in our sampleare overseen by white men compared with only41 percent of supervisor positions and 39 per-cent of worker positions Column 2 shows thatafter controlling for this opportunity structurewomen and minorities actually appear to prac-tice homosocial reproduction in positions ofpower at higher rates than white men Forexample results indicate that 28 percent ofwhite-male superiors select other white men tofill supervisor positions immediately belowthem This rate of homosocial reproductioncontrasts with the rate of 36 percent amongwhite women approximately 50 percent amongblack men Latinos and Latinas and 65 percentamong black women Rates of homosocialreproduction in manager positions convergeby comparison indicating that after controllingfor relative opportunity all groups are rough-ly the same in their tendency to select similarpeople to fill manager positions immediatelybeneath themselves

Overall these findings help to refine ourunderstanding of homosocial reproductionTable 5 indicates that patterns consistent withhomosocial reproduction are common amongall race-sex groups Yet only white men havesufficient opportunity to engage in these prac-tices with relative frequency and this frequency




Sat 30 Sep 2006 155556

increases with upward movement within work-place power hierarchies In other words in-group favoritism may be universal butopportunities to practice it are not As a resultcolumn 3 of Table 5 indicates that 27 percentof all manager positions in our sample werefilled in a manner consistent with white-malehomosocial reproductionmdashover three timesthe rate for white women and over eight timesthe rate for respective minority groups

AASSCCRRIIPPTTIIVVEE TTRRAAIITTSS OOFF SSUUPPEERRIIOORRSS AANNDD PPOOWWEERR

AATTTTAAIINNMMEENNTT OOFF MMIINNOORRIITTIIEESS AANNDD WWOOMMEENN

Hypothesis 5 asserts that how women andminorities advance up workplace power hier-

archies is conditioned by whether suchadvancement occurs under ascriptively similaror dissimilar superiors The underlying idea isthat homosocial reproduction influences therelative importance of network assistance andhuman capital in moving up organizationalchains of command To test this hypothesiswe estimate a multinomial regression equa-tion to predict the likelihood of employment atsuccessive levels of workplace power for eachgroup of non-white-men The independentvariables are the same as those in Model 3 ofTable 3 with the addition of interaction termsfor having an ascriptively similar superior(yesno) by network assistance and the fourindicators of human capital If Hypothesis 5 is



Table 5 Opportunity and Rate of Homosocial Reproduction Decomposed by Group and Level of Power

Gender and Race of Positions that Superiors Fill Superiors that Oversee Positions that Superiors Fill with In- through Homosocial Respondentsrsquo Work Respondentsa () Group Membersb () Reproductionc ()

Workers (n = 2480)mdashMenmdashmdashWhite 0394 229 (n = 978) 092mdashmdashBlack 0089 389 (n = 221) 035mdashLatino 0125 473 (n = 309) 059mdashWomenmdashmdashWhite 0240 303 (n = 596) 073mdashmdashBlack 0106 719 (n = 263) 076mdashmdashLatina 0046 611 (n = 113) 028mdashTotal 1000 (mdash 363Supervisors (n = 470)mdashMenmdashmdashWhite 0413 284 (n = 194) 117mdashmdashBlack 0094 477 (n = 44) 045mdashmdashLatino 0096 467 (n = 45) 045mdashWomenmdashmdashWhite 0264 363 (n = 124) 096mdashmdashBlack 0104 653 (n = 49) 068mdashmdashLatina 0030 500 (n = 14) 015mdashTotal 1001 (mdash 386Managers (n = 315)mdashMenmdashmdashWhite 0593 455 (n = 188) 270mdashmdashBlack 0082 385 (n = 26) 032mdashmdashLatino 0062 579 (n = 19) 036mdashWomenmdashmdashWhite 0189 450 (n = 60) 085mdashmdashBlack 0073 500 (n = 22) 037mdashmdashLatina 0000 NA (n = 0) 000mdashTotal 0999 (mdash 460

Note Sample includes all eligible respondents not just workers entering new jobs within five years of the surveya Relative opportunity for homosocial reproductionb Rate of homosocial reproduction given the opportunityc The percent is an estimation Data in this column show aggregate rate of homosocial reproduction


Sat 30 Sep 2006 155556

correct we would expect coeff icients forldquoascriptively similar superior times network assis-tancerdquo to be positive and statistically signifi-cant whereas we would expect coefficientsfor corresponding interaction terms with edu-cation total work experience prior job-specificexperience and organizational tenure to benegative and statistically significant BecauseLatinas very rarely gain positions of powerunder ascriptively similar superiors we do notinclude them in this analysis

Results of this test offer little support forHypothesis 5 (see Table A4 on the ASR Website supplement httpwwwasanetorgjournalsasr2004toc039html) For blackmen Latinos and black women none of thecoefficients for respective interaction termsreach statistical significance at the 05 levelThese nonsignificant findings imply that net-work assistance and human capital are equal-ly predictive of authority attainment underascriptively similar superiors as under ascrip-tively dissimilar superiors Results for whitewomen by contrast produce several statisti-cally signif icant coeff icients for relevantinteraction terms The finding most consistentwith Hypothesis 5 indicates that for whitewomen organizational tenure matters less foradvancing from supervisor to manager underother white women than it does under ascrip-tively dissimilar superiors The other statisti-cally significant interaction term involvestotal work experience but its effects run con-trary to Hypothesis 5 To illustrate we solvethe equation for white women for increasingyears of work experience setting all otherfactors equal to subsample means for whitewomen We then plot the estimated odds ofemployment at successive levels of power inFigure 1

Results reveal countervailing effects of workexperience for white womenrsquos advancementFirst the top panel of Figure 1 indicates thatwhite women with relatively little work expe-rience are much more likely to advance fromworker to supervisor under ascriptively simi-lar superiors than under ascriptively dissimi-lar superiors This conditional differenceappears to last until white women gain between15 and 20 years of work experience at whichtime the relative odds of advancing from work-er to supervisor become roughly equal (butlow) under both types of superiors By contrast

and contrary to Hypothesis 5 the bottom panelof Figure 1 indicates that work experiencematters less for advancing from supervisor tomanager status under ascriptively dissimilarsuperiors than under ascriptively similar supe-riors

While unexpected these findings suggestthat white women tend to take one of twotracks up workplace power hierarchies (1)they enter into supervisory positions relative-ly early in their careers under other whitewomen and then as they accrue experienceslowly increase their odds of advancing fromsupervisor to manager or (2) they enter intomanagerial positions under white men rela-tively early in their careers Because whitemen oversee more managerial positions thanwhite women (see Table 5) the second routeis currently more common for white womenrsquosmanagerial attainment These findings sug-gest a very different picture than that impliedby Hypothesis 5 Instead of experience beingmore important for advancement from super-visor to manager status under white men expe-rience is less important

One possible explanation for this finding isthat white men tend to view experience amongwhite women less in terms of productive capac-ity and more in terms of fading desirabilityleading white men to favor younger less expe-rienced white women over older more expe-rienced white women for manager positions allelse equal Another possible explanation isthat times have changed in recent years suchthat white women entering the labor force nowdo not have to prove themselves to white-malesuperiors to the same extent as they had to inthe past in order to advance into managerialpositions Either way results yield little over-all support for Hypothesis 5 The main find-ings imply that human capital and networkassistance are equally determinant of advance-ment among black men black women andLatinos regardless of ascriptive similarity withsuperiors Among white women howeverascriptive similarity with superiors improvesodds of attaining supervisory status with littlework experience but this benefit does not thenopen doors to managerial power youth underwhite men does




Sat 30 Sep 2006 155556

CCOONNCCLLUUSSIIOONN

This research had two broad goals First wewanted to determine if there is empirical evi-dence of increasing inequality in workplacepower for a wider array of women and minori-ties than previously examined in sociologicalresearch Second we wished to study the mech-anisms that help create and sustain this form of

inequality among recent job entrantschangersIn this investigation we focused specificallyon hypothesized differences in human capitaland observable interpersonal mechanisms thatdifferentiate white men from other groupsdepending on their particular combination ofascriptive traits One of these mechanisms waseffective network assistance the other mecha-nism was superiorsrsquopreferences for similar oth-



Source Estimated multinomial regression equation predicting successive levels of workplace power among whitewomen (see ASR website Table A4) Control variables are set at subsample means for white women

Figure 1 White Womenrsquos Estimated Odds of Employment at Successive Levels of Power by Total Years of WorkExperience


Sat 30 Sep 2006 155556

ers Certainly these are not the only mecha-nisms that determine who attains power andthese mechanisms are surely mediated by orga-nizational policies and behavior that requirefurther investigation

Still our findings indicate that with reason-able statistical confidence men and women ofvarious races and ethnicities experience increas-ing inequality in workplace power relative towhite men but they experience it to differentdegrees and via different mechanisms AmongLatinos and white women increasing inequal-ity appears to result largely from human-capi-tal def iciencies relative to white men(specifically education among Latinos and workexperience among white women) Thereforeone policy recommendation might be to improvehuman capital among these groups in hopes thatgreater similarity in education and experiencewill bring greater similarity in workplace powerattainment However there are at least two rea-sons to be skeptical of this planrsquos success overthe long term

First white men currently do not need to dis-criminate against Latinos to ensure Latinosrsquononcompetitiveness because educational dif-ferences yield effectively the same result If thehuman-capital difference between white menand Latinos decline and competition intensifieswhite men might close ranks against Latinosthrough other mechanisms Second whitewomen appear to fare worse not better underwhite-male superiors as they gain work expe-riencemdashthe opposite of what we might expectBecause white men oversee the majority ofmanagerial positions in US workplaces thispattern can offset human-capital improvementsamong white women Both scenarios of courseultimately depend on organizational practicesthat convert human capital into equal opportu-nity and suggest that merely equalizing humancapital credentials will be insufficient by itselfto remove patterns of increasing disadvantagein tomorrowrsquos workplaces

With respect to networking the strongestempirical results run counter to expectationBlack women not white men appear most like-ly to rely on instrumental network assistance toattain positions of power This pattern couldreflect several dynamics First networking canserve as an important response as well as causeof direct discrimination as research on immi-grant adaptation and ethnic economies sug-

gests Second people who face multiple oppres-sions such as black women due to their race andgender might be uniquely conscious of net-work assistance when they receive it makingthem more likely than other groups includingwhite men to report better information on net-work assistance in surveys Third because net-working is typically more an intra-organizationalprocess than an extra-organizational processwhen positions of power are at stake our use ofa random sample of employees across manyworkplaces might understate the relative impor-tance of network assistance for white mensrsquoauthority attainment A larger random samplewithin organizations might yield results moreconsistent with traditional conceptualizations ofldquoold boyrdquo networks since this type of datawould permit better analysis of the differentialeffectiveness of networks within internal laborpools

Finally with respect to preferences for sim-ilar others there are strong findings to indicatethat most superiors regardless of their race andsex tend to fill power positions they overseewith ascriptively similar others that is theyappear to engage in what Kanter calledldquohomosocial reproductionrdquo Findings also showthat because there are more white men at high-er levels of workplace power than members ofother groups white men have greater opportu-nity to exercise this self-similar preference andin the process reproduce their advantage oversuccessive generations of employees Whatremains to be determined is the extent to whichthese patterns reflect a priori segregation ofraces and sexes across establishments and jobsas opposed to ldquoreal timerdquo preferences of supe-riors for similar others In extreme cases forcesof segregation are in operation long beforeemployers make hiring decisions regarding posi-tions of power This can leave superiors with fewnonsimilar candidates from which to chooserendering their ascriptive preferences moot Inother cases the circle of eligible candidatesmight be quite diverse rendering the prefer-ences of those in charge more salient for under-standing group inequalities in power attainment

This consideration points to another areawhere more and better information about intra-organizational dynamics could be useful indetermining the specific subprocesses at workin producing increasing ascriptive inequality inhigher positions of workplace power In addition




Sat 30 Sep 2006 155556

to this new (and costly) data collection strate-gy future research on this form of inequalitymight look to comparative ethnographies thatexamine how different groups perceive andadjust to the unique sets of obstacles they appearto face relative to white men as they move upworkplace power hierarchies As we await theseinsights we should resist the conclusion that allpeople who are not white men face the samehurdles to attaining higher levels of workplacepower Evidence here suggests that a one-size-fits-all explanation hides more than it revealsand that more research is still needed to pinpointthe precise mechanisms that convert differentcombinations of ascriptive characteristics intoinequalities in workplace power

James R Elliott is an Assistant Professor of Sociologyat Tulane University where he conducts research onurban development and ascriptive inequalities in thelabor market In addition to research on workplacepower he is currently examining the social and eco-nomic organization of immigrant dispersal from gate-way cities to new destinations throughout the USurban hierarchy

Ryan A Smith is an Associate Professor at the Schoolof Public Affairs City University of New York and arecent Scholar in Residence at the WEB DuBoisInstitute of Harvard University He has publishedinternationally recognized research on race and gen-der stratification in workplace authority and servesas an organizational change consultant to publicand private organizations

RREEFFEERREENNCCEESS

Alessio John C and Julie Andrzejewski 2000ldquoComment Unveiling the Hidden Glass CeilingAn Analysis of the Cohort Effect Claimrdquo AmericanSociological Review 65311ndash5

Baxter Janeen and Erik O Wright 2000 ldquoThe GlassCeiling Hypothesis A Comparative Study of theUnited States Sweden and Australiardquo Genderand Society 14275ndash94

Bendix Reinhard 1956 Work and Authority inIndustry New York Harper and Row

Braverman Harry 1974 Labor and MonopolyCapital New York Monthly Review Press

Bridges William P and Wayne J Villemez 1986ldquoInformal Hiring and Income in the Labor MarketrdquoAmerican Sociological Review 51574ndash82

Campbell Karen E and Rachel A Rosenfeld 1985ldquoJob Search and Job Mobility Sex and RaceDifferencesrdquo Research in the Sociology of Work3147ndash74

Cassirer Naomi and Barbara Reskin 2000 ldquoHighHopes Organizational Position Employment

Experiences and Womenrsquos and Menrsquos PromotionAspirationsrdquo Work and Occupations 27438ndash63

Cotter David A Joan M Hermsen Seth Ovadia andReeve Vanneman 2001 ldquoThe Glass CeilingEffectrdquo Social Forces 80655ndash82

Dahrendorf Ralf 1959 Class and Class Conflict inIndustrial Society Stanford CA StanfordUniversity Press

Federal Glass Ceiling Commission 1995 Good forBusiness Making Full Use of the Nationrsquos HumanCapital Washington DC US GovernmentPrinting Office

Fernandez John P 1975 Black Managers in WhiteCorporations New York John Wiley and Sons

Granovetter Mark 1995 ldquoAfterword 1994Reconsiderations and a New Agendardquo Pp 139ndash82in Getting a Job A Study of Contacts and Careers2d edition Chicago IL University of ChicagoPress

Halaby Charles N 1979 ldquoJob-Specif ic SexDifferences in Organizational Reward AttainmentWage Discrimination vs Rank SegregationrdquoSocial Forces 58108ndash27

Jacobs Jer ry 1992 ldquoWomenrsquos Entry intoManagement Trends in Earnings Authority andValues among Salaried Managersrdquo AdministrativeScience Quarterly 37 282ndash301

Johnson James H Melvin Oliver and LawrenceBobo 1994 ldquoUnraveling the Paradox ofDeepening Urban Inequality TheoreticalUnderpinnings and Research Design of a Multi-City Studyrdquo Urban Geography 1577ndash89

Kanter Rosabeth M 1977 Men and Women of theCorporation New York Basic Books

Kluegel James 1979 ldquoThe Causes and Cost ofRacial Exclusion from Job Authorityrdquo AmericanSociological Review 43285ndash301

McGuire Gail M 2002 ldquoGender Race and theShadow Structure A Study of Informal Networksand Inequality in a Work Organizationrdquo Gender ampSociety 16303ndash22

McGuire Gail M and Barbara F Reskin 1993ldquoAuthority Hierarchies at Work The Impacts ofRace and Sexrdquo Gender and Society 7487ndash506

Morgan Laurie A 1998 ldquoGlass-Ceiling Effect orCohort Effect A Longitudinal Study of the GenderEarnings Gap for Engineers 1982 to 1989rdquoAmerican Sociological Review 63479ndash93

mdashmdashmdash 2000 ldquoReply to Alessio and AndrzejewskiIs Engineering Hostile to Women An Analysis ofData From the 1993 National Survey of CollegeGraduatesrdquo American Sociological Review65316ndash21

Moore Wilbert 1962 The Conduct of theCorporation New York Random House Vintage

Morrison Ann M and Mary Ann Von Glinow 1990ldquoWomen and Minorities in ManagementrdquoAmerican Psychologist 45200ndash8

Moss Philip and Chris Tilly 2001 Stories Employers




Sat 30 Sep 2006 155556

Tell Race Skill and Hiring in America NewYork Russell Sage

Mueller Charles W Toby L Parcel and KazukoTanaka 1989 ldquoParticularism in AuthorityOutcomes of Black and White Supervisorsrdquo SocialScience Research 181ndash20

Myerson Allen R 1997 ldquoAs Federal Bias CasesDrop Workers Take Up the Fightrdquo New York TimesJanuary 12 Section 1 Column 3

Pfeffer Jeffrey 1983 ldquoOrganizational DemographyrdquoPp 299ndash357 in Research in OrganizationalBehavior edited by Larry L Cummings and BarryM Staw New York JAI

Podolny Joel M and James N Baron 1997ldquoResources and Relationships Social Networksand Mobility in the Workplacerdquo AmericanSociological Review 62673ndash93

Reskin Barbara F 2002 ldquoRethinking EmploymentDiscrimination and Its Remediesrdquo Pp 218ndash44 inThe New Economic Sociology Developments in anEmerging Field edited by Mauro F GuilleacutenRandall Collins Paula England and MarshallMeyer New York Russell Sage Foundation

Reskin Barbara F and Catherine Ross 1992ldquoAuthority and Earnings Among Managers TheContinuing Signif icance of Sexrdquo Work andOccupations 19342ndash65

Reskin Barbara F and Debra B McBrier 2000ldquoWhy Not Ascription OrganizationsrsquoEmploymentof Male and Female Managersrdquo AmericanSociological Review 65210ndash33

Robinson Robert V and Jonathan Kelley 1979ldquoClass As Conceived by Marx and DahrendorfrdquoAmerican Sociological Review 41209ndash34

Rosenfeld Rachel A Mark E Van Buren and ArneKalleberg 1998 ldquoGender Differences inSupervisory Authority Variation among AdvancedIndustrialized Democraciesrdquo Social ScienceResearch 2723ndash49

Smith Ryan A 1997 ldquoRace Job Authority andIncome A Cross-Temporal Study of Changes inthe Socioeconomic Status of Black and WhiteMen 1972ndash1994rdquo Social Problems 44701ndash19

mdashmdashmdash 2002 ldquoRace Gender and Authority in theWorkplace Theory and Researchrdquo Annual Reviewof Sociology 28509ndash42

Smith Ryan A and James R Elliott 2002 ldquoDoesEthnic Concentration Influence EmployeesrsquoAccessto Authority An Examination of ContemporaryUrban Labor Marketsrdquo Social Forces 81255ndash79

Tomaskovic-Devey Donald 1993 Gender andRacial Inequality at Work The Sources andConsequences of Job Segregation Ithaca NYILR Press

Tsui Anne and Charles A OrsquoReilly 1989 ldquoBeyondSimple Demographic Effects The Importance ofRelational Demography in Superior-SubordinateDyadsrdquo Academy of Management Journal32402ndash23

United States Bureau of Labor Statistics 1996 NewsRelease BLS Reports on the Amount of Formal andInformal Training Received by EmployeesWashington DC US Department of Labor

Weber Max [1914] 1968 Economy and SocietyAn Outline of Interpretative Sociology Translatedand edited by Guenther Roth and Claus Wittich3 volumes New York Bedminster

Wilson George 1997 ldquoPathways to Power RacialDifferences in the Determinants of Job AuthorityrdquoSocial Problems 4438ndash54

Wilson William Julius 1996 When WorkDisappears The World of the New Urban PoorNew York Knopf

Wolf Wendy C and Neil Fligstein 1979 ldquoSexualStratification Differences in Power in the WorkSettingrdquo Social Forces 5894ndash107

Wright Erik O Janeen Baxter and Gunn EBirkelund 1995 ldquoThe Gender Gap in WorkplaceAuthority A Cross-National Studyrdquo AmericanSociological Review 60407ndash35

Yamagata Hisashi Kuang SYeh Shelby Stewmanand Hiroko Dodge 1997 ldquoSex Segregation andGlass Ceilings A Comparative Static Model ofWomenrsquos Career Opportunities in the FederalGovernment over a Quarter Centuryrdquo AmericanJournal of Sociology 103566ndash632




Sat 30 Sep 2006 155556

Consequently it remains uncertain just howgeneralizable the claims from these studies areand the selection biases they might reflect

Finally prior quantitative studies on ascrip-tive inequalities in workplace power focus eitheron race or gender but not both (for notableexceptions see McGuire and Reskin 1993Tomaskovic-Devey 1993) This conventionalapproach is problematic because it reinforces theerroneous assumption that racial stratificationand gender stratification are mutually exclu-sive systems that we can somehow sum tounderstand differences among non-white-menBy contrast we view racial ethnic and genderstratifications as having fundamentally similarcauses and how these causes overlap to producedifferent outcomes for different groups remainsa decidedly empirical question In taking thisapproach it is important to move beyond exclu-sively black-white comparisons to considerLatinos not only because Latinos constitutethe largest and fastest growing panethnic groupin US society but also because recent researchindicates that employers of all ethnoracial back-grounds tend to prefer Latinos over blacks whenfilling positions in their organizations (Moss andTilly 2001 Wilson 1996)

In the present study we address these short-comings in prior research through an examina-tion of authority attainment among white blackand Latino men and women Our objectives aretwofold (1) assess the extent to which inequal-ity in workplace power increases among womenand minorities relative to white men at higherlevels of power and (2) examine the mecha-nisms of allocation responsible for this form ofinequality for each group In pursuing theseobjectives we focus specifically on power posi-tions under the supervision of others One rea-son for this focus is that workplace power is notlimited to the upper echelon of Fortune 500 cor-porations it is found in all places of employmentWhile popular reporting on the glass ceilingoften obscures and even trivializes the phenom-enon it is critical to general understandingbecause even seemingly mundane jobs are not leftto run themselves as the lack of autonomy inmost low-skill positions attests Instead peopleare selected to fill positions of power throughoutall levels of the workforce And mid-level posi-tions not only constitute the most common formof legitimate authority but they also representpositions wherein vertical and horizontal inter-

group competition is likely to be greatest amongincumbents of roughly equal credentials

BBAACCKKGGRROOUUNNDD

A popular explanation for ascriptive inequali-ties in workplace power invokes the metaphorof a glass ceiling which was first popularizedin a 1986 Wall Street Journal article thatdescribed barriers women often face as theyclimb corporate ladders In 1995 the UnitedStates Department of Labor issued an officialreport on this subject noting that during theinterim decade observers had extended the termtwo ways first to include racial and ethnicminorities in addition to women and second torefer to all management and decision-makingpositions not just to top-level positions at largecorporations (Federal Glass CeilingCommission 1995iii) A core idea invoked bythe metaphor is that while employers might letwomen and minorities into low positions ofauthority they are much less likely to let theminto high positions that involve greater control

While we do not explicitly test for corporateglass ceilings in this research we do examinewhether women and minorities have an increas-ingly difficult time relative to white menaccessing jobs with greater organizationalpower Recent research along these lines hastaken one of two general approaches The morerestrictive approach views this type of inequal-ity in terms of an absolute barrier that blockswomen and minorities from higher positionsof workplace power because they are womenand minorities (Jacobs 1992 Morrison andGlinow 1990 Reskin and McBrier 2000 Reskinand Ross 1992) Taken literally this perspectiveimplies an invisible barrier below which womenand minorities attain a modest degree of work-place power (eg supervisory authority) andabove which they do not (eg managerial con-trol) A less restrictive approach views increas-ing inequality as a form of disadvantage facingwomen and minorities relative to white menwhich intensifies at higher levels of workplacepower

This second approach to conceptualizingincreasing inequality has two important impli-cations for empirical assessment First inequal-ity is presumed to occur not in a single absolutestep but rather over several steps of increasingmagnitude relative to white men Second and




Sat 30 Sep 2006 155556















Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556







DDAATTAA





Sat 30 Sep 2006 155556






VVAARRIIAABBLLEESS








Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556



Myerson Allen R 1997 ldquoAs Federal Bias CasesDrop Workers Take Up the Fightrdquo New York TimesJanuary 12 Section 1 Page 1 Column 3























Sat 30 Sep 2006 155556















Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556







DDAATTAA





Sat 30 Sep 2006 155556






VVAARRIIAABBLLEESS








Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556







DDAATTAA





Sat 30 Sep 2006 155556






VVAARRIIAABBLLEESS








Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556







DDAATTAA





Sat 30 Sep 2006 155556






VVAARRIIAABBLLEESS








Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556







DDAATTAA





Sat 30 Sep 2006 155556






VVAARRIIAABBLLEESS








Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556






VVAARRIIAABBLLEESS








Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556












Men Women

XX



White(N = 566)

120 (325)

147 (354)

443 (497)

574 (495)139 (223)153 (101)581 (494)571 (630)435 (188)

184 (387)354 (409)431 (495)

075 (265)

133 (339)

355 (478)

476 (500)

410 (492)

Black(N = 885)

053 (224)

154 (361)

262 (440)

601 (490)132 (198)155 (108)421 (494)628 (737)461 (200)

266 (442)359 (310)270 (442)

097 (296)

238 (426)

388 (488)

212 (409)

490 (500)

Latinaa

(N = 535)

037 (190)

095 (294)

142 (349)

715 (452)105 (370)130 (106)394 (489)382 (471)394 (169)

136 (344)358 (340)127 (333)

370 (483)

254 (435)

248 (432)

344 (475)

703 (476)

White(N = 513)

203 (403)

163 (370)

712 (454)

596 (491)143 (244)170 (110)589 (493)686 (817)429 (197)

166 (372)372 (332)489 (500)

261 (439)

099 (300)

144 (351)

455 (498)

255 (436)

Black(N = 454)

101 (302)

181 (385)

258 (438)

646 (479)132 (218)171 (104)529 (499)603 (699)445 (200)

240 (428)366 (306)222 (416)

366 (482)

268 (433)

138 (346)

329 (471)

273 (446)

Latinoa

(N = 527)

102 (304)

142 (350)

338 (473)

726 (426)102 (374)156 (108)421 (494)422 (445)371 (166)

076 (265)367 (263)102 (303)

582 (493)

211 (408)

104 (306)

533 (499)

463 (499)


Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556



PPOOWWEERR











Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556




RREESSUULLTTSS










Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556



Tab

le 3

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Gro

ss a

nd N

et G

aps

in P

ower

Att

ainm

ent

Lev

els

of P

ower

Men

Wom

en

Bla

ck B

1L

atin

o B

2W

hite

B3

Bla

ck B

4L

atin

a B

5

Mod

el 1

Gro

ss G

apa

mdasha

Sup

ervi

sor

vers

us w

orke

r ndash

027

(

174)

ndash31

6(

176)

ndash25

6 (

172)

0ndash2

88

(15

4)0ndash

855

(19

2)mdash

b M

anag

er v

ersu

s su

perv

isor

ndash79

2

(23

5)ndash

542

(2

31)

ndash41

3 (

220)

ndash12

76

(

223)

ndash11

50

(

302)

mdashM

odel

2

= 1

324

(10

df)

Mod

el 2

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

alb

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

078

(

177)

ndash08

1(

193)

ndash19

8 (

173)

0ndash1

74

(

157)

0ndash4

72

(

205)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

649

(

238)

ndash08

0(

254)

ndash34

5 (

221)

ndash11

66

(

227)

0ndash7

44

(

316)

mdashM

odel

2

= 2

786

(18

df)

mdash

2 te

st o

f m

odel

2 v

ersu

s m

odel

1 =

146

2 (

8df)

Mod

el 3

Net

Gap

wit

h C

ontr

ols

for

Hum

an C

apit

al a

nd E

mpl

oym

ent C

onte

xtc

mdasha

Sup

ervi

sor

vers

us w

orke

rndash

119

(

181)

ndash12

1(

195)

ndash02

6 (

180)

0ndash0

42

(

164)

0ndash3

71

(

208)

mdashb

Man

ager

ver

sus

supe

rvis

orndash

311

(

262)

ndash14

5(

262)

ndash17

8 (

231)

0ndash8

43

(

238)

0ndash4

36

(

324)

mdashM

odel

2

448

3 (

30 d

f)mdash

2

test

of

mod

el 3

ver

sus

mod

el 2

= 1

697

(12

df)

Not

eD

ata

show

n fo

r sp

ecif

ic g

roup

ver

sus

whi

te m

en S

tand

ard

erro

rs a

ppea

r in

par

enth

eses

N =

34

80

alo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

blo

g[P

r(L

evel

n)P

r(L

evel

nndash1)

] =

a +

bi[r

acendash

sex i

] +

b2[

year

s of

edu

cati

on]

+ b

3[to

tal w

ork

expe

rien

ce]

+ b

4[pr

ior

jobndash

spec

ific

exp

erie

nce]

+ b

5[ye

ars

wit

h em

ploy

er]

bire

port

ed(w

hite

men

as

refe

renc

e gr

oup)

c

log[

Pr(

Lev

eln)

Pr(

Lev

elnndash

1)]

= a

+ b

i[rac

endashse

x i]

+ b

2[ye

ars

of e

duca

tion

] +

b3[

tota

l wor

k ex

peri

ence

] +

b4[

prio

r jo

bndashsp

ecif

ic e

xper

ienc

e] +

b5[

year

s w

ith

empl

oyer

] +

jbj[e

mpl

oym

ent c

onte

xtj]

indi

cato

rs o

f em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

hou

rs w

orke

d pe

r w

eek

(log

ged)

and

occu

pati

onal

loca

tion

(pr

ofes

sion

alt

echn

ical

cra

ftr

epai

r se

rvic

e c

leri

cal

sale

s [r

ef]

) b

ire

port

ed (

whi

te m

en a

s re

fere

nce

grou

p)

p

lt 0

5

p

lt 0

1

p

lt 0

01 (

two-

tail

ed te

st)


Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556












Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556



Tab

le 4

M

ulti

nom

ial R

egre

ssio

n C

oeff

icie

nts

for

Log

-odd

s of

Pow

er A

ttai

nmen

t and

Int

erac

tion

Coe

ffic

ient

s

Dif

fere

nces

fro

m W

hite

Men

Dif

fere

nces

am

ong

Whi

te M

enB

lack

Men

Lat

inos

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Key

Var

iabl

es a

nd M

odel

Sta

tist

ics

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Net

wor

k as

sist

ance

(0

1)0ndash

439

0(

282

)00

381

00

(35

2)00

715

0(

419

)0ndash

632

0(

578

)9

05

(47

8)ndash1

029

0(

610

)Y

ears

of

educ

atio

n0ndash

001

0(

061

)00

080

00

(07

9)00

131

0(

086

)00

108

0(

123

)0

39

(07

0)0ndash

039

0(

093

)To

tal w

ork

expe

rien

ce00

001

0(

015

)00

006

00

(01

8)00

013

0(

086

)0ndash

001

0(

030

)ndash

038

(02

3)0ndash

020

0(

030

)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

467

0(

282

)0ndash

197

00

(34

7)0ndash

023

0(

402

)0ndash

431

0(

538

)ndash

192

(40

8)0ndash

600

0(

529

)Y

ears

wit

h em

ploy

er00

054

0(

023

)00

022

00

(02

5)00

001

0(

033

)0ndash

013

0(

039

)1

13

(04

0)0ndash

044

0(

051

)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

0ndash1

74

0(2

83)

007

06

00(

345)

002

32

0(4

31)

0ndash7

65

0(6

23)

ndash26

4(

436)

0ndash8

52

0(6

01)

Con

stan

tndash5

608

(2

195

)ndash7

300

0(

312

8)ndash3

621

(1

631

)ndash6

247

(2

461

)ndash4

362

(1

744

)ndash8

200

(2

561

)M

odel

2

(df)

933

(24

)15

41

(38)

174

9 (3

8)N

442

829

888

Dif

fere

nces

fro

m W

hite

Men

Whi

te W

omen

Bla

ck W

omen

Lat

inas

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Sup

ervi

sor

vs

Man

ager

vs

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Wor

ker

Sup

ervi

sor

Yea

rs o

f ed

ucat

ion

000

28

0(0

82)

000

57

0(1

15)

001

30

0(0

77)

0ndash0

13

0(1

23)

018

0 (

076

)ndash

061

(11

6)To

tal w

ork

expe

rien

ce0ndash

006

0(

022

)00

001

0(

029

)00

003

0(

019

)0ndash

035

0(

030

)0

029

(02

2)ndash

067

(03

8)P

rior

job-

spec

ific

exp

erie

nce

(01

) 00

075

0(

410

)0ndash

003

0(

522

)00

060

0(

352

)00

045

0(

518

)ndash

521

(44

3)ndash

080

(70

0)Y

ears

wit

h em

ploy

er00

026

0(

036

)0ndash

028

0(

041

)0ndash

039

0(

028

)0ndash

021

0(

038

)ndash

002

(04

2)0

060

(05

9)A

scri

ptiv

ely

sim

ilar

sup

erio

r (0

1)

004

47

0(3

96)

ndash11

44

0(5

21)

001

12

0(3

76)

0ndash5

86

0(5

80)

NA

NA

Con

stan

tndash3

610

(1

514

)ndash9

246

(2

361

)ndash4

741

(1

559

)ndash8

519

(2

581

)ndash7

507

(1

847

)ndash5

814

(2

714

)M

odel

2

(df)

174

0 (3

8)25

70

(38)

227

5 (3

6)N

916

118

689

4

Not

e D

ata

show

n w

ith

stan

dard

err

ors

in p

aren

thes

es I

ndic

ator

s of

em

ploy

men

t con

text

incl

ude

num

ber

of w

orke

rs in

the

resp

onde

ntrsquos

est

abli

shm

ent (

logg

ed)

pub

lic

sect

or (

01)

ho

urs

wor

ked

per

wee

k (l

ogge

d) a

nd o

ccup

atio

nal l

ocat

ion

(pro

fess

iona

lte

chni

cal

craf

tre

pair

ser

vice

cle

rica

lsa

les

[ref

])

Sam

ples

incl

ude

only

wor

kers

ent

erin

g ne

w jo

bsw

ithi

n fi

ve y

ears

of

the

surv

ey

NA

= n

ot a

ppli

cabl

e to

o fe

w L

atin

a su

perv

isor

s (n

= 7

) an

d m

anag

ers

(n =

0)

repo

rt h

avin

g as

crip

tivel

y si

mil

ar s

uper

iors

wit

h w

hich

to c

ompa

re

p

lt 0

5 tw

o-ta

iled

test


Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556














Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556













Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556










Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556









Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556











Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556

































Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556


























Sat 30 Sep 2006 155556

Race, Gender, and Workplace Power

Documents

Transcript of Race, Gender, and Workplace Power