TECHNIC.41- REPORT iXTICIi/TR-30/G32

66 Q)

T!= 1988 ANTHROPOMETRIC SURVEY CV N OF US ARMY PERSONNEL: Q CORRELATION COEFFICIENTS AND 14 REGRESSION EQUATIONS <r:

PART 1 STATISTICAL TECHNIQUES, LANDMARK, AND

MEASUREMENT DEFINITIONS BY

JAMES CHEVERUD CLAIRE C. GORDON* ROBERT A. WALKER* CASHELL JACQUISH

LUCI KGHN ALLEN MOORE

NYUTA YAMASHITA

7

MAY 1990

FINAL REPORT APRIL 1989 - MARCH 1990

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED

UNITED STATES ARMY NATICK RESEARCH, DEVELOPMENT AND ENGlNEERING CENTER

NATICK, MASSACHUSETTS 01 760-5000 NORTHWESTERN UNIVERSITY, EVANSTO", IL LINGIS 60201

'SOLDIER SCIENCE DIRECTORATE

DISCLAIMERS

The findings contained in this report are not to be construed as an official Department of the Aimy position unless so designated by other authorized documents. Citation of trade names in this report does not constitute an official endorsement or approvzl of the use of such items.

DESTRUCTION NOTICE

For Classified Documents: Follow the procedures in DoD 5200.22-M, Industrial Security Manual, Section 11-19 or DoD 5200.1 -R, Information Security Program - -

Regulation, Chapter 1X.

Destroy by any method that prevents disclosure of contents or reconstruction of the document.

Uni lassif ied ECL;.Tf CL;SSI~:CATION OF THIS PAGE

REPORT DOCUMENTATlON PACE Farm A p p w r d OMB NO. 0 7 w o r m

la. REPORT SECURITY CLASSIFICATION 1 b. RESTRICTIVE MARKINGS Unclassified

!a. SECURITY CUSSIFICATION AUTHORITY 3 . DISTRIBUTIONIAVAIUBILITY OF REPORT Approved for public release;

!5 . DECLASSIFIUTIONIDOWNGUDING SCMDULE Distribution unlimited

I PERFORMING ORGANIZATION REPORr NUMBER(S) 5. MONITORING ORGANIZATION REPORT NUMBER!S)

NATICK/TR-~O/O~~ 1

ia. NAME GF PERFORMING ORGANIZATION 6b. OFFICE SYMBOL 79. NAME OF MONITORING ORGANIZATION (If applicrbk)

Eorthwestern University U.S. Army Natick RD&E Center

ic ADDRESS (City, State, and ZIP Cadc) 7b. ADDRESS (City, State. a d ZIPCodcJ

Evanston, Illinois 60201 I Natick, Massachusetts 01760-5000

I I-

h. NAME OF FUNDING 1 SPONSORING 8b. OFFICE SYMaOL 9. PROC'JREMENT Ih LTRUMENT IDENTIFICATION NUMBER ORGANIZATION {If applicable)

STRNC-Y BF Contract DAAKbO-89-C-1006 I

ic ADDRESS (Cicy, State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERS

FROGPAM TASK WORK UNIT NO. ACCESSION NO.

I I I I . TITLE (lncludc Securrly Ua~l f rcahon)

1'388 ANTHRCPOLYETRIC SURVEY OF U.S. ARMY PERSONNEL: CORRELATION COEFFICIENTS AND REGRESSION EQUATIONS Part 1 Statistical Techniques, Landmarhand Measurement ~efinitions

12 PERSONAiAUTHOR(S) James Lheverua, Claire C. Gz+rdon*, Robert A. Walker.*, Cashell Jacquish, 1 . ~ 1 c i Kohn, Allen Hoore, and iUvuta Yamashita .-

13r. TYPE OF REPORT I i3b . Ti&% COVERED 114 DATE OF REPORT (Year, Month, Day) 11 5. PAGE COUNT ,, I F ~ C M 1 Apr 8910 31 Mar 94 1990 May I Final ! 6 SUPPLEMENTARY NOTATICN

:':nffil-iated Gith-soldier Science Directorate, U.S. Army Natick Research, Development, and Engineering Center

17 COSATl CODES I 3. SualECT TERMS (Cononue on raven* t f MCCW~). dnd M n t r t y by block number)

FIELD 1 GROUP ( SUbGROUP ANTHROPOMETRY HUMAN 8 0 3 Y SIZE PEN I I ANTHROPOMETRIC SURVEY MEASUREE!EIZS WOMEN I L

I t ANTHROPOHETRIC DATA DIMENSIONS ARMY 19. A 6 S i U C l (Contrnua on revers* r f n w e s w y and rdentrty by bloc& number) In this five-part series of reports ';re#tabulate$ the statistical relationships among anthropometric' measurements so that designers of Army materiel systems will be able to utilize the data in improving the man-materiel interface. The kinds of statistical relationships tabulated include simple correlations, partial correlations, bivariate regressions and multiple regressions. The simple correlations, partial correlations, and bivariate regressions are among all pairs of characters. Four sets of partial correlations are inciuded, partialling out: stature only; weight only; stature and weight simultaneously; and stature, weight, end age simultaneously. Standard multiple regressions include specific anthropometric dimensions regressed on standard pairs of measurements. Stepuise multiple regressions were also calculated for each anthropometric dimension using the first five independent variables selected in the stepwise procedure. All analyses were performed separately for males and females and are reported separately in the table~.(Continued)

-.. 121. ABSTRACT fECURtTY CUSSlFICATlON

Box 19-.Abstract (Continued)

1988 Anr-hro~ornetric Survev of US Armv Personnel: Replression Equations and C~rr~lation Coefficients appears in five parts, as follows. part 1 Statistical Techniaues. Landmark. a& Heasurernent Definitions ( p p 1--51), NATICK/TR-Y0/032; Part 2 Simple and Partial Correlation Tables--Male ( p p . 52--312), NATICK/TR-90/033; Part 3 Simole and Parcial Correlation Tables--Female (pp 313--572), NkTICK/TR-90/034; part 4 Bivarlate Re~i-ession Tables (pp. 573--853), NATICK/TR-90/035; Part 5 Ste~wise and Stendard Mu'ltivle Rezression Tables (pp. 854--984), NATICK/TR-90/036.

The report, L988 Anthro~ometric Survev of US Armv Personnel: Bivariate Frequency Tables, part of the same contract, is complete in one volume, NATICK/TR-90/031.

References for the series appear at the end of Part 5 .

CONTENTS

Part 1 Statistical Techniques, Landmark, and Measurement Definitions 1

STATISTICAL E C H N I Q W

General Statistical Assumptions Simple Correlation Coefficients Partial Correlation Coefficients Bivariate Linear Regression Multiple Linear Regression

LANDMARK DEFINITIONS AND ][LLUSTRATIONS 13

MEASUREMENT DEFINITIONS AND ILLUSTRATIONS 25

Standard Measurement Definitisns Visual Index INDEX

Part 2 Simple and Partial Correlation Tables-Male

Table 1: Male Simple Correlations Table 2: Male Partial Correlations-Stature Table 3: Male Partial Correlations--Weight Table 4: Male Partial Correlations-stature Sr Weight Table 5: Male Partial Coirelations-stature, Weight & Age

Part 3 Simple and Partial Correlation Tables-Female

Table 6: Female Simple Corrziations Table 7: Female Partial Correlations-6:ature Table 8: Female Partial Correlations-Weight Table 9: Female Partial Correlations-Gtature & Wei ht Table 10: Female Partial Correlations-Gtature, ~ e i & t & Age 5il

Part 4 Bivarlate Regression Tables 573

Table 11: Male Bivariate Re essions Table 12: Female Bivariate egressions INDEX

18 Part 5 Stepwise and Standard Multiple Regresslon Tables

Table 13: Male Ste wise Multi le Re reuions 9 P a R 855 Table 14: Female tepwise Mu tlple egressions 90 1

Table 15: Male Standard Multi le Re essions Table 16: Female Standard M 3. t~ple 18 egressions

REFERENCES

IIWEX

PREFACE

This report was prepared for use by Army and other personriel in designing human-materiel interfaces. The work was begun in the Department of CpD Biology & Anatomy, Northwestern University, Evanston, IL and completed in the Department of .hatomy & Neurobiology at the Washington University School of Medicine, St. Louis, MO. We wish to thank the contract aklministrators at both institutions and at the U. S. Army Natick Research, Development, and Engineering Center for their help in facilitating the work, especially in regard to moving the project in midstream.

This report was prepared b James M. Cheverud and colleagues at Northwestern University and Was 6 ington University under Army contract DAAK6G89-C-1006 during the period April 1989 through March 1990. Dr. Claire C. Gordon was the roject director of the U. S. Army 1988 Anthropometric Survey, and Dr. Robert A. 4alker was the project officcr for the contract. Dr. Gordon and Dr. Walker are affiliated with the AnthropoEogy Group, Materiel Systems Human Factors Branch, Behavioral Sciences Divisicm, Soldier Science Directorate.

Part 1 Statistical Techniques, Landmark, rnd Measurement Definitions

CHAPTER I

INTRODUCTION

The Army hxs the responsibility to provide clothing, personal protective equipment, and workspaces that accommoda:e the full rangz of body size variation present in its user population. This mission can be accomplished only with the aid of an anthrcpometric data base that accurately reflects the body size and shape distribtitions of soldiers. Rece~tly (1987-1988), the Army updated its 20-year-old data ba;e by conducting a full-scale anthropometric survey that consisted of ove 180 body and head dimensions taken on a sample of more thaq 9 0 soldiers. i Previous anthropometric surveys were conducted in 1966 for males- and in 1377 for females? The demographic composition of the Army has chnnsed dramatically since these ea lier surveys in the proportional representation of various age, sex and racial groupsJ Thus it was apparent in the mid-80s that the Army's anthropometric data base needed to be generated anew in order to account for demographic changes and secular trends in body size and proportion.

The survey featured a sampling strategy in which demographic minority groups were intentionally "oversampled" in order to accommodate anticipated demographic shifts i n the Army population and in order to support basic research goals. This methodology will allow researchers to use the data coilected to anticipate future changes in the composition of Army's staff. Survey participants were selected at random within their age/gender/race strata at 11 Army posts in the continental United States.

At the close of the survey, a working data base representing the male and female components was created by stratified random sampling of the total data base, such that the age/race distributions of the working subsets exactly match those of thc June 1088 Active Duty Army. The working sample contains 1774 male individuals and 2208 females. Demographic details of the working data base sample, alonq with basic statistics for the measurements taken, can be found in Cordon et al.

I n this report we tabuiate the statistical relationships m o n g anthropometric measurements so that designers of Army materiel systems will be able to utilize the data to improve the human-materiel inlcrface. The kinds of statistical relationships tabulated ir~clude simple correlations, partial corre!atims, hivariate regressions, and multiple regressions. 'The simple correlations, partial correlations, and bivariate regressions are among all pairs of characters. Four sets of partial correlations are included, partialling out: stature only; weight only; staiure and weight simultaneously; and statute, weight, and age simultaneously. Standard multiple regressions include specific anthropometric dimensions regressed on standard pairs of measurements. Stepwise multiple regressions were also calcu!ated for each anihropornetric dimension using the first five independent variables selected in the stepwise procedure. All analyses were performed separately for males and females and are reported separately in the tables below.

A total of 180 an:hropornetric measurements (see Chapter IV) are included In the analysis. Fortyeight of these are detailed head measurements derived fro !! three-dimensional coordinates measured with an automated headboard device. The remamind 132 measurements were standard anthropometric dimensions representing the size of parts from all over the body6 Definitions of each of the measurements analyzed are provided in Chapter IV.

The next three chapters provide a technical description of the statistical analyses performed and the use of the reported statistics (Chapter 11), the definitions and il1ustra:ions of the various landmarks used in anthropometric measurement (Chapter IJI), and the defhitions and illustrations of the anthropometric measurements themselves (Chapter IV). Following this preliminary descriptive information the various statistical tables are presented in four chapters, one for each type of analysis. Chapter V contains simple and various partial correlation tables for males and, separately, for females. Chapter VI contains listings of d l of the bivariate regressions of sufficient strength of association to allow prediction of anthropometric measurement values from others among the total set of measurements analyzed. For each anthropometric variable, Chapter VII includes the multiple regressions with one, two, three, four, and five independent variables that best predict the depedent variable's value. The final chapter, Chapter VIII, includes the multiple regression models for specific sets of dependent and independent variabies.

STATISTICAL TECHNIQUES

General Statistical bsclmptions

When ~xarnining anthropometric data it u assumed that tne data are normally or bivariate normally distribute and are a randcm sample of the populaticn from which they were drawn. That is, we assumed that the individuals included were independently ascertained an6 are representative of all individuals in the Amy. The means by which the workiqg data base was generated assures compliance with the random sample assumption while the large sample utilized here assures that the sampling d~stribution of the parameters will be distributed in a statistically appnpriate fashion. With these asumptions, we are able to reliably ascertain the probability that any s!atistic estimated frov this sample is different from zero.

Simple Corielation Coeflicients

A simple correlation coefficient (r), sometimes referred to as a Pearson productmoment correlation, is a measure of the strength of the linear relationship between two variables. Its value ranges from -1.00 to + 1.00, with -1.00 representing a perfect negative linear re!ationship (as X increases, Y decreases), 0.00 representing a total lack of !inear selatio~hip, and + 1.0 representing a perfecr positive linear relixionship (u X increases, Y a h incr~ases) betweer, the variables. A linear relationship occurs wLen !he chmge In Y (aY) over a given change in values of X (AX) is constant over the total range ol X values under consideration. Deviations from a linear relationship result in a lowering of the sbsolute value of correlation coefficients.

A simple correlation coefficient between characters X and Y is defined as fi3llows:

where COV(X,Y) is the covariance of the two charac!crs and V represents their variance. A covariance measures the extent to which two variables varj in concert. It is positive when individuals with a larger than average value of X are also typically larger than average for character Y, negative when largcr than aqcrage values of X are typically associated with smaller than avzrage values of Y, and zcro wnen a

given value of X may be associated with any vdue of Y. It is caiculated as:

wberea,and arc the average values of X and Y respectively and N is the number of individuals used to caiculate the covariance. Another eqivaltnt equation often used in calculation is:

- where XY is the mean of the a ~ ~ - p i o d ~ c t of the two variables. A variance measures the &ent of differences among individuals for the character ir~ question. It is calculated as follows:

or, alternatively,

7 where (X ) is the average of the squared values of X Inspection of these equations indicates that a varisnce is thc covariance of G tmit with itsdf.

A simple correlation is thus a Coyariancc standardized by the level of variation exhibited by the characters. 'Zhis sitnnnntdization allows one to compare the strengths of linear relatiomlip across chmter sets of grossly different levels of variability. Highly variable measurements, su& as weight, will have high cowimces with dl other characters as compared to lowly variable measurements, such as heel breadth. Mwever, because correlations are standardized for levels of

variability, they can be directly compared across characters. Thus, inspection of the correlation tables will allow designers to quickly determine which dimensioris are associated in the Army populatior~ and therefore which character sets need to impact design criteria in an integrated fashion.

I t is also importmt to determine the probability that rl partic~lar ?%timated correlation statisr~i is significantly dit'fcrent from zero. In finite samples, it is pc!ssible that correlat;om other than zero rnay be estimated due to random chanc~ errors even when the true population correlation is zero. Thus, we need to determine which of the many correlation coefficients estimated actually indicate a linear association between the variables. In general, an observed statistic is considered significantly different from zero when the probability of obtaining ~ c h an observation in the actual population IS 5% or less. This test applies to a single statistical determination.

The standxd criterion implies that 1 in every 20 itldependent statistics may appear significantly different from zero by chance alone. Hdwever, with the 180 traits in the 1988 Working Data Base, the full matrix of correlations between all traits contains [(I80 ' 179)/2], or 16,110 different correlation coefficients. With the xceptance of the 5 % probability level, one would expect approximate!y 800 significant c:)rrelations bv chznce alone, even when there is no correlation in the pc;pulc :;r! GS a whole. Tnus, with multiple statistical comparisons, it cap be difficult to determine which cc;rrelations are actually not equal to zero and which only appear so. This prohlern is known as a multiple comparisons problem and is overcome by usrng the Bonferroni criterion, or Dunn's test, for determining the significance of individual coefficients. Using the Bonferroni criterion, the experimentwise (or collective) error rate for Type I errors (errors in which the null hypothesis of no correlation is frllsely rejected) is limiterl to no more that 576, so that there is only a 5% probability of an incorrect rejection of the null hypothesis for even a single coefficient out of ih:: entire collection. This probability limitation is accomplished by dividing the error rate per comparison (typically 0.05) by the total number of comparisons rnade (k, with k = 16,110 in these analyses). Then the probability which must be rnct for 3 s in~le coefficient to be considered significantly different from zero is (O.OS/k). The significance levels presented in this report represent the Bonferroni sigrlificance Icvcls. 'Thrashold correlations, marking the boundary above which individual correlations are significant, were calculated using the Ponferr ni significance level ( P = O.OS/lh.llO) and the standard error c f the r(2rrelstion.g The ihrrshoid correlntions are 0. i 13 in [he male s:~mplc and 0.085 in the fema!e sample. 'Ihe threshold correlations differ in the two sexes due to differences in sanl?lt: size.

Part id Correlation Coeflicients

A partial corre1;llion coefficient measures the linear relationship between two v a ~ iahles after other, qxcificd variables have been held constant. The relationship between the two variables of interest is considered independent from

their common relationship to a third cr further variables. Thus, the partial correlation rV+* is the correlation between traits X and Y with trait Z held constant. When only one controlling variable is inchded, the correlation is referred to as a &st order partial correlation. It is calculated from the i.ppropriate simple ccrrelation coefficients as follows:

A second order partial correlation controls for two variables while a third order partial correlation controls for &ee variables. The second order panial correlation between X and Y contro!ling for b o h W and Z is calculated as:

from the first order partial correlations. A third ord:r partial correlation can be calculated from a set of second order partial correlations in a similar fxshion. Significance testing for partial correlations is essentially the ramc a for simple correlations. With samples as large as those included in the working data base, the critical partial correlation values for statistid significance using the Bonfcrroni critsrion arz the same as for the simple correlations, at least to the third decimal place.

Partial forrelations are calculated when one wishes to focus on that specific linear relationship between two variables vhich is independent of other factors. In anthropometric analyses, the variables mcst often 'partialled out' are size variables. For example, size is controlled in order to determine whether individuals of the same size with longer than average anns dso tend to have longer than average legs. 'This analysis is necessary since the simple relationship between arm and leg length may be due merely to the observation that larger people have both longer arms and longer legs rather than being due to some special relationship between arm and leg length. 'Rle partial correlations between all trait pairs have been calculated when adjusting for the effects of stature and weight separately, the combined effects of stzture and weight, and the combined effects of stature, weight, and age. Thus, these partial conelationr can be used to quantify and interpret body shapes.

Bivariate Unear Regression

Bivariate regresion measures the hear relationship between a dependent (Y) and single independent (X) variab!e by estimating the number of unit

differences in trait Y which typically accompany a single unit of i~.crease or decrease in trait X. The model used for the linear relationship is:

where a is t ! x regression constant or intercept iden*ng the value of Y when X equals zero and by, is the regression coefficient or slope indicating the number of unit differences in Y given a single unit difference in X The regression coefficient i s calculated as:

where cov(X,Y) and V, art as defined above. The regrcssion constant is then calculatcd a:

where X and are the means of the independent and dependent variables, respectively. The values estimated for coetficients a and b,,, in this equation are those that minimize the squared deviations of obxwed Y values from the estimated regression line. This procedure is known LS least squares regression.

The use of this regression model, usually referred to as Model I regression, is based on three assumptions: ( 1 ) The illdependent variable, X. is measured without error; (2) the mean values of the dependent variable (Y) are a function of the values of the independent variable (X) and lie on a straight line described by equation H; and (3) io r any pnrticular value of the independent variahle (Xi), the dependent v;~riable's ( Y ) values are independently and normally distributed with constant variance acrocs various values of the independent variable. Assumption 2 was described above in the discussion of simple correlations while Assumption 3 was discussed in general terms at the beginning of the chapter. The first assumption will not generally hold for acthropometrrc variables, in that all measurements are taken with some degree of error. Even so, when prediction is the main purpose of regr ssion analyses, as i t is here, Model I regression is the most appropriate form to u5e. 5

The primary purpose for regression ctxificients in the context of tlis report

is to allow the prediction of the dependent variable given a specific value for the independent variable. This is accomplished by using equation 8 and treating a, and X as known and using them to estimate Y. The estimate of Y obtained in 1s fashion is the most likely, or average, value of Y for the Xi given in the equation. Fur example, we may predict the value of acromial height, sitting (ACRHTST, standard rneiisurernent 4) from acromial height (ACRHGHT, stanaard measurement 3) in males (see Measurement Definitions and Illustrations, Chapter 4). The appropi late regression equation is:

Estimated ACRHTST (mm) = 128.038 + 0.320 ' ACRHGHT (mm). (11)

Even though this point estimate is the most likely value of the dependent variable (ACRHTST) given the value of the independent variable (ACRSGHT), normal variation in the dependent value causes individual cases to deviate from the prediction. The standard error of the estimate measures the extent of variation in the dependent variable at any given value of the independent variable. It is the standard deviation of the residuals which, in turn, are the deviations of the observed values from those predicted by the regression model. Thus, we car. use the standard error of the estimate (g.x) to provide a likely range of dependent values given the independent value specified. G i v ~ n the very large sample sizes used here, the normal correction applied to the standard error of the estimate hen the independent variable deviates from its mean can hc dispensed with here? Thus a 95% confidence interval for estimated values of thc dependent variable can k calculated as:

Estimated Y 1 1.96 yXx. (12)

In our example estimating acromial height, sitting from acromial height, given an rtcromial neight of 1442 rnm and standard error of the estimate equal to 21.634 mm, the estimated acrornial height, sitting is 59% mm and its 95% confidence interval will Ix 5 W mrn f ( 1.90'2l.h34), rnnging from 556 mm to 640 mm. 'Ihis means that 95% of the time individt~iil\ w ~ t h an N'KfiGi l 'r of 1442 will have ACRIITST values between 5% mrn and MO mm. The intlcpcndent variahlc which best predicts a dependent variable is thc one with thc smallest standard error of the estimate.

Divariate regressian is very closely related to simple Sivariate correlation. The regression coefficient can he crrlculated as:

where si is the square root of the variance (see equation 4). Statistical significance tests are identical for both regression and correlation coefficients. Since the bivariate regression tables only contsin equa:ions for which the correliition is greater than 0.50, all regrtssion equations reported are significant ever. at the 0.901 level using the Ronferroni criteria.

Another parameier of interest for regression analysis is the coefficient of determination (R ). The coefficient of determination measures thz proportion of variation in the dependent variable which is associated witn the independent variable and thus measures the strength of associatior! represented by the regression. This coefficient is simply the square of the correlation coefficient between the two variables. Thus, it ranges from 0 to + 1. The higher the coefficient o\determination, the more tightly coupled the two variables considered are. The R values reported in these tables have been adjusted for the number of independerrt variables used in the analysis. This adjustment does not affect the bivariate coefficients of determination but does have a slight effect on the values reported for multiple regressions. In the regression of acromial height, sitting on acromial height, 46.6% of the variation is shared between the variables. Only regression equations with coefficients of determination above 0.25 are included in the tables.

The starrdard error of the regression coefficient is also provided in the tables. I t allows tests (if significance for estimated coefficients by allowing calculation of the 95% confidence interval for thc estimate. The 95% confidence interval for the regression coefficient is given by 2 1.96 times the standard error of the regression coefficient. For the regression of ACRHTST on ACRHGIJT, the standard error of the regression coefficient is 0.008, and the 95% confidence interval of the regression coefficient is (0.310, 0.342). Since the 95% confidence interval does not encompass 0.00, the slope of the regression is significantly different fronr 7ero at the 5 % level.

I t may also be desirable to predict the Y variahle in English units as well as in metric units. I n the condersion of a regression equation from metric to English units, the value of the slope remains unchanged unless the i~dependent variable is WEIGfiT, wnile the constant and the standard error of the estimate are multiplied by the factor 0.fUO. I n the regression of xrornial height, sitting on acromial height, thc constant in English un i t s i s 5 . 0 4 1 inches. and the equarion expressed in English u n i t \ is:

Ltimrrted ACKI { K T ( i n ) = 5.04 1 + 0.326 ' ACRf iGI1T (in), ( 14)

while the standard error of the c\timatc converted to I%glish units is 21.633 ' 0.040 or 0.852 inches. !n c;rw\ whcrc the indcpcntlenr variahlc is WI~l(;ll'I', the slope is

multiplied by :he factor 0,147 to convert it from millimeters per centigram to inches per pound. The constant and th!: standard error of the estimate are multiplied by the factor 0.040. The prediction of acromial height sitting (ACRHTST) in males from weight is expressed in these equivalent equations:

Estimated ACRHTST (mrn) = 491.407 + 0.136 * WEIGHT (cg) (15)

Estimated ACRHTST (in) = 19.347 t 0.020 WEIGHT (Ib). (16)

If one wishes to transform these equations to English units when WEIGHT is the dependent variable, the constant and standard error should be multiplied by 0.268 and the slope by 6.805. . .

Multiple Linear Regression

Multiple licear regression is useful in the prediction of a dependent variable from several independent variables simultaneously. The analyses tabulated here represent two classes of regression analysis, multiple regression and stepwise multip!e regression. 'The multiple regression equations represent the prediction of each dependent variable front a specified pair of independent variables; stature and chest circumference, sta!ure and shoulder circumference, waist circumference and crotch height, buttock circumference and crotch height, vertical trunk circumference and crotch height, and stature and weight. In contrast, stepwise multiple regression is performed in a systematic manner, such that it results in the best limited subset of predictor variables that accurately predict the dependent variable. The number of independent variables was limited to five for these analyses.

Multiple regression is a relatively straightforward extension of bivariate regression. Typically, the combination of two or more independent variables in a mu!tiple regression allows a more accurate prediction of a dependent variable than can be obtained from a single independent variable alone, as is done in bivariate regression. The general multiple regression model used is:

where the subscripts identify the first, recon4 third, up to kth independent variable. a is the regression constant or intercept, and f q is the partial regression coefficient. The partial regression coefficient is the number of unit changes in the dependent variable (Y) per ti1,it change in the indicated independent variable (Xi) when all other independent variables are held constant. it is the unstandardized version of the partial correlation ~ocfficicnt (cquatim 6) md can be calculated as follows when there are two independent variables (X and 2):

The standardizing term in equation 16 is the ratio of variance in Y not associated with Z to variance in Xi not associated with Z. This, it is seen &at equation 16 is similar in structue to equation 13 relating simple regression and correlation coefficients. Similar: more ccmplex, equations a n be derived for multiple regressions with further independent variable , although calculations are usually performed using matrix algebraic algorithms.' lie regression constant may be estimated as:

dthough once again the matrix formulation is preferred. The coefficients estimated by these algorithms describe the line for which the squared deviation of observed Y values from Y values predicted by the line of regression arc minimized.

In addition to the assumptions described above for bivariate regression, in multiple regression it is s s m e d that the correlations among the independent variables are not extreme. If correlations among independent variables are very high (greater than approximately 0.95), a multicollinearity problem will exist. This condition will lead to instability of regression coeffiaent estimates due to the limited amount of unique variance in specific independent variables available fcr partial correlation with the dependent variable. Howevcr, multicollinearity is unlikely to seriously affect prediction using the equations. Inspection of correlation coefficients among regression coefficient estimates indicates that multicollinearity is not a serious problem in the52 analyses.

All of the regressions presented are individually statistidly sigaificant at the 0.001 level and collectively at the 0.05 level csing the Bonfcnoni criterion. Confidence intervals for individual partial regression coefficients are provided in the tables and c m be used to derive 9570 corxidence limits as described in the section on bivariate regression coefficients. Likewise, multiple coefficients of determination are provided, whicb indicate the proportion of variance in the dependent variable associated with the set of independent variables.

Prediction of dependent variable values from specific sets of values for independent variables is the main purpose of the coefficients presented in this report. The procedure to be followed is the same as in bivanate regression with the simple addition of independent variables. For example, male abdominal extension depth, sitting (ABEXDPST) can be predicted from buttock circurnfmmce (BUITCIRC) and ~xotcb hei -bt (CRCHHGHT) using the following eauation:

Estimated ABEXDPST(nm) = -42.063 + 0.363 * BUlTCIRC(mm) -0.091 CRCHHGHT(mm). (20)

Note that all coefficients are statistica'lly significant although crotch height shows no significant bivariate relationship to abdominal extension depth, sitting. The inclusion of additional independent variables can alter the apparent relationship between the dependent and specific independent variables.

The standard error of the estimate provided in the multiple regression tables can be used, as above, to derive confidence intervals for dependent variable estimates. Due to the large sample sizes, and consequent small standard errors for the regression coefficients, inflation of the estimate's confidence intervals due to deviation of the independent variables from their meay is minor within the rar'ge of the data. If correction is desired, see Sakal and Rohlf (p. 637).

The stepwise multiple regressions were calculated using standard multiple regression procedures after stepwise selection of the appropriate variables. The stepwise procedure was followed in order to choose s set of five independent variables, which would best predict the dependent variable. While the best prediction for any single dependent variable will be obtained from the full set of other variables (179 in this report), such equations become unwieldy and unreliable when applied to samples other than the one used to estimate the equation. Thus. it is preferable to select some subset of independent variables that will produce nearly as accurate a prediction and be more reliably applied to new sam?les. No stepwise procedure is perfect and the inclusion or exclusion of specific variables from the selected set of independent variables should not be taken as indicating a speciai relationship with the depcndent variable. Potential independent variables showing high correlation with the dependent variable may be excluded from the final equation due to complex and multiple corinearities with independent variables already included in the favored variable set. Also, although stepwise procedures do not necessarily produce :he optimal five variable equation, they will typically produce an equation that approaches the optimal one in predictive power.

We used a stepwise forward selection procedure in which independent variables are added to the multiple regression equation sequentially. first variable entered is the one with the highest coefficient o f determination with the dependent variable. The second variable added is the one that causes the greatest increase in the multiple coefficient of determination. This is the variable with the highest partial correlation with the dependent variable, controlling for the first variable entered. Additional indepcndent variables are added, one at a time, using the same criterion. At each step the included variables are reevaluated so that any variable that initially fits the criterion but, with the new complement of independent variables, no longer appreciably increases the coefficient of determination, is removed from the selected set. This prccedure was followed until five independent variables were selected. These five variables were then used in a series of five regressions with the independent variabie, each further regresion including an additional selectcd independent variable. These regressions are then treated as any other multiple rcgrc4on. as dcscrihed ahwe.

CHAPTER 111

WNDMARK DEFINITIONS AND ILLUSTRATIONS

The dimensions are measured from one point on the body (or 3 fixed surface such as the floor) to another or, in the case of circumferences, arouqd the body at a specified level. To ensure that each dimension is measured accurately and consistently from one subject to the fiext, dimensions are defined in t e r n of body landmarks, ivhich serve as the origin, teminaticn, or level of measurement of a dimension.

Two men and two women were trained in locating many of these points by palpation or by sight, and placing actual drawn marks on the bodies of all subjects in this survey. Measurers were also trained to recognize other easily located landmarks such as dactylion 11 landmark, the tip of ;he index finger, for whicn marking was not necessary.

The landmarks used to define the measurements in the survey are briefly oescribed and illustrated cS:er the following pages. Detailed instructions for locating tilase landmarks can be found in the Measurer's andb book^

LANDMARKS

Abdominal point, anterior: The most protruding point of the relaxed abdomen of a seated subject.

Acromion, right and left: The p i n t of intersection of the lateral border of the acromial p r o w s and a e r u g d o the middle of the shoulder from the neck to the tip of the shoulder.

Arropodion: The tip of the Grst or / !,!,)\ secxmd toe of the 4&!:>-- - right foot, whichever is longer. t

Alarc, right and left: The lateral point on the flare o r wing of the nose.

Anterior superior iliac spine, right and left The anterior points of the right and left iliac crests.

pusteribr: right and Wt: The highest points of the right and left d a r y folds on the back.

Biceps point The highest point of the right fl&cd biceps ns viewed from the subject's right side.

Bustpoint, right and kfk The anterior poinu of the bra c"P"

LANDhIARKS (cont'd)

Buttock point, posterior: Point of maximum protrusion of the right buttock of a standing subject.

Calf: A point o n the side of the cdlf at the level of the maximum circumierence of the right calf.

Cheilion, r i b t and left T h e lateral point of the juncture of the fleshy tissue of the lips with the facial skin at the corner of the nouth .

Clavicle point, and kft: T h e superior point lateral ends of clavicle.

right

of the the

Buttock point, right lateral and left lateral. Points o n the thigh o r hip at the level of the

--

--

--

maximum protrusion of the right buttock.

Ce~calc: T n e superior palpable point of the spine of the seventh cervical vertebra.

Chin: The most protruding point on 1 . L - the bottom edge of the chin, along the

Uieion: The lowest :\

jawline. I

p i n t of the hairline \ . on the forchcar! in / the midsagittal plane.

I

a me tip . [b 1 f the right index

leibib point, right md left: The lateral mint of the right ieltoid muscle, and :he margin of the left deltoid muscle at the level of the right deltoid point.

Donal junctutt of the fooi and leg: The top of a skin crease between the right foot and the front of the ankle when the knees and ankles are flexed about 30 degrees.

Ear point: The lateral point (farthat from the head) of the right ear. F'

e middle fingei-. -

tonal junctun of thc ilf and thigh: The ~ncture between the ght calf and thigh ehind the knee of a ubject sitting with he knee flexed 90 legrcts,

right ear on its long axis. -+ b

Ear, top: The highest poht of the right ear on its long axis.

Emcanthus: T h e r\ :+ outside comer of the

\ right eye formed by the meeting of the upper and lower eyelids.

Eaoorbitale, right and left: The posterior p o h t an the frontal pro- of the zygomatic bone at the levci of the outer comer of the eye.

Elbow crease: T h e skin crease o n the inside of the right elbow joint when the elbow is flexed 90 degrees.

F i mctatanophakngcal protrusion: T?: lateral protrlsion of the right foot in the region af the fifth rnetatarsophalangeal joint.

First mctatanophalangeal protrusion: T h e medial protrusion of the right foot in the region of the first rnetatanophalangeal joint.

Glabek T h e anterior point o n the frontal bone midway between the bony browridga.

I I rightandlck The point of deepest indentation of thc temporal crest of the frontal bone above the browridges.

Glulcal furroar point: T h e lowwt point of the lowest f u m or crease at the juncture of the right buttock and the thigh

Gu~ion, ri$t and kft: ?he lateral point on the posterior angle of the mandib!~ (jawbone).

Iliocrirtale: Tnc highest palpable paint cf the right &ac crest of the pelvis, one half the distance bewcen the anterior iuperior iliac and posterior superior iliac spines.

Wrauhitalc, right and kfk The lowest point on the anterior border of the bony eye socket.

Inrwr thigh: A vcflical line hslfkay b e m . n the front and f l back of the right inner thigh, and' extending downward from the level of the gluteal fu.?ow.

f me inferior

Hatl point, latunl a d m a % & The lateral and medial points of the right heel located at or behind the most protruding point of the lateral malleolus (outside ankle bone).

@J Inferior b m t point:

the juncture lower of the breasts with 1

point of of the two the t o m .

Inh-athyroid: The infer i~r point in the midsagittal plane of the thyroid cartilage (Adam's ap?le).

Knee point, anterior: The most protruding point of the right kneecap of s scnted subject.

LatPsal femoral epicondylc, standing and sitting: Lateral point of thc right femoral epicondyle (knee pivot point).

T h e medial point of the right medial mal le~ius (inside ankle bme) .

1

Mebcarpale LI: T h e lateral point of the right metacarpo- phalangeal jomt LI (at the base of the index G n ~ e r on the outer d g e of the hand).

Midpatclh The anterior point halfway between the top and bottom of the right patciia ( the lcrieccap).

Lateral mallmlus: Thc lateral paint of the right lateral malieolus (outside ankle bone).

Muton: The inferior point of the mandible in the midsagiti;ll p i m e (bottom of the chin).

Metacapalc V: The medial point of the nght metacarpo- phalangeal joint V (at the base of the tittle finger on the outer cdgc of the hand)

Midshoulder: The phi on top of the right shoulder midway be tween the neck (right trnpezius point) and the tip of the shoulder (acromion, right).

LANDMARKS (cont'd)

ateral, and left a t a a k Anterior and Midspine: A line

down the center of the back

.., \d IA.-

ateral points at the , . m e of the n e c k

\

Olocranon, center: A point o n the center ot k? Olccranon, bottom ,

and rear. The lowest and rearmost points of the right elbow with the elbow flexed 30 degrees.

1 the CUW~:IJK of the , & rig!it olecranon process with the 'C Abow Ccxcd a b u t LJ

Otobasion superior. The anterior superior point of the juncture bctwccn the right car

Posterior superior iliac spine: T h e posterior point of the crcst of the right Ili?tm. A dimple normally over l ia this point. and the head. e

Pro& The p i n t cf the anterior projection of the tip of the nose.

Promcntoll: !lie antcrilx projccimn of thc soft t ~ u e of ttlc \----.I chin.

posterior point of the right hecl.

%,& 'b- L>*s

Radik R e highezt p i n t on the outside edge of [he right radius.

Scyc: Poinu on the upper arm and t o m asociatcd with the annholc of a garment.

Anterior scyc oa Lbc mm: A short h.. ' .,nzontal tine o n the 3r

lono originating at the !.. .-) a j x x of the right /?-\

Anterior rcye oo the upper urn: A short ho rmnta l ilne on the uppcr arm orlpnatmg at the apex of the nght arltcrior mllq fold

Pratcrior brimat.4 rcyc, right and kft A short hor~tunldi lmc on the back oriKlnatlng

,--,, I , i t thc npcx o f the In)\\cl r o r ,~x~Jlary fo!d.

Portcrior h g o d scyc, right uld kft A diagonal line connecting the apex of the pwtcrior axillary fo!d with the aucmha landmark o n the tip cf the 1 houlder.

M;Lcye, right and 161: A short h o r ~ w n t d imc hucctlng the PtrLcrior 4Fd ryc landmark.

srye Lcvcl at ma. iap i~~ A short horizontal lia: across the spine at the lcvcl of the Polnior hcwi.mlrl ccye landmarks.

Selliow The pint of the deepest depression of the nasal bones at the top of the nose.

Stlrfi M: TI:^ l~1.vest pcmt a i :it.: hottom ~f tkc right radius.

Suv-. Thc inferior pant ot the jug~lar notch o f the stcrnum (top of the breastbone).

south is closed. ' [ ~ -

Stomba: The point )f intersection of the ..per and 1-r lip kq& n the midsagittal ,lane when the i

Sub&-bular: The juncture, in the midsagittal plane, of the lower jaw (mandible) and the neck.

Suppattllr. The ~upcrior point o: the right patella (h-p).

Tcntb rib.: Thc inferior point of the right tenth rib (bottom of thc rib age).

llelion, right and left Center of the nipple (on males).

Thumbtip: The tip of the right thumb.

Thigh point, to?: The highest point of the top of the right thigh of a seated subject.

Top of head: The higbett point on the , jj ad

head when the head s / in the Frankfort plane.

Trapxius pirf right d kit: llic p i n t a t wh~ch the ~n lc r io r

- I / - . . '

1 ' i 5 -

--A\.

'rrtxhanlcr: A ;x)~r.t 1 t L;>Ls-----.\ J i .! at the center of the -l 1 1 1

1;iteral surface of thc , , 1 . , grcalcr trtxhantcr of I , \ I ! t t x r~p,ht fcmur o f J \ I / l

Trochantcrioa: Thc superior p i n t of4:hc grcatcr trtxhantcr of thc right fc~nur o f a standing subjcct.

Waist ( n a d indenta- tion): right aad lefg anterior and posterior: Level of the greatest indentation on the right side of the torso, or half

)°distance 10th rib and between IlioQistalc Waist (omphalion): right and left; anterior

if no single indentation and postexior: Level of is clear. the center of she navel.

Wrist, don& A line ac:oss the back of the right wist originating ar the stylioq landmark and perpendicular to

1 the long axis of the The lateral point on a m the zygomatic arch.

Zygohntalc, right and left: The lateral point of the frontal bone on i ts zygon?atic prows.

CHAPTER IY

MEASUREMENT DEFINITIONS AND ILLUSTIZATIONS

One hundred thirty-two directly measured :cithropmetric dimensions were obtained in the survey using standard instruments and ixethods. Where there was a choice of right or left sides, all measurements were taken on the right side of the body unless specified othenvise or, in rare cases, where an injury or anatomical abnormality made i t necessary to measure on the left side. 41 anthropometric dimensions were recorded to the nearest millimeter and all results reported here for these dimensions are in n~illimete:~. This unit allows standard errors and regression coefficients to be specified with a greater degree of precision than allowed on a centineter scale. Weight is reported to the nearest 0.1 kilogram. Detailed illustrated instructions for taking the measurements are included in Clauser et 31.' while detailed definitions are provided in Gordon et a1.l

In addition to general anthrnpometric measurements, a series of 48 special head and f x e dimensions were derived from the three dimensional coordiaates of 26 landmarks collected hy means of a special automated headboard device.) These additional measurements were collected because traditional anthropornetric measures do not provide enough detail in the hexi and face to maximally aid in the design of personal protective

es Irators. equipment, such as helmets, goggles, and r p'

The sample size analyzed here includes 1774 males and 2208 females. With only minor exceptions, all measuremcnt.. are available for iach individual. The landmark crinion cannot be located on bald subjects or those with receding hairlines. I t is also not recorded for individuals with hair transplants. This resulted in the 105s of two fcmale and tuerity-seven males for crinion - back of the head and crinion - top of the head. Intcrpupillary breadlhs were also unavailahle for three males and one female whose morphology lay outside the range that could be measured hy the device employed for this purpose. These small numbers of misslng values have no practical effect on the statistic. reported below.

For each standard mea.surement. :he listing provides the variablr: number used to idcntify vari;ibles in several of the tables below, the full measurerncnt rime, an eight charaster abbreviated name also defined in the data base, and a brief definition. The data ha\e vilriable nurnherc arc not consecutive bccar~se a series of anthropornetric rnc.;~snrernents dcrivcd trorn the original 132 standard anthroiwrncrric measurements arc not included in this report. A l w in the l'ables of this report variable number 1 corresponds to subject ID numner.

A visual indcx o f rneawrcmcnr illu\rrations is provided rrt the end of the chapter.

Standard Measurement Definitions

Note:

2

3

4

5

6

7

8

9

10

11

12

13

14

Ail measurements refer to a subject's right side unless otherwise specified.

ABDOMINAL EXTENSION DEPTH, Sl'I7'ING (ABEXDPST) -- horizontal distance between the anterior point of the abdomen and the back at the same level.

ACROMIAL HEIGHT (ACRHGHT) - vertical distance between the standing surface and the acromion landmark on the tip of the shoulder.

ACROMIAL HEIGHT, SI?TING (ACRHTST) -- vertical distance between the sitting surface and the acromion landmark on the tip of the shoulder.

ACROMION-RADIALE LENGTH (ACRDLGTH) -- distance between the acromion landmark at the tip of the shoulder and the radiale landmark on the elbow.

ANKLE CIRCUMFERENCE (ANKK1WCT)I - minimum horizontal circumference of the ankle.

AXILLA HEIGHT (AXHGHT) -- vertical distance between the standing surface and the axillsry fold at the anterior scye landmark on the torso.

AXILLARY ARM CIRCUMFERENCE (AXARCIRC) -- circurnferencr of the upper arm perpendicular to its long axis at the level of the anterior scye landmark on the upper arm.

BALL OF FOOT CIRCUMFERENCE (BLFTCIRC) -- circumference of the foot at the first and fifth metatarsophalangeal protrusion landmarks on the ball of the foot.

BALL OF FOOT I.EN5TH (BLFTLCTH) - distance between the back of the heel and the landmark at the fi.-st metatarsophalangeal protrusioi~ on the ball of the foot.

BIACROMIAL BREADTH (BCRMBDTH) - posterior distance between the right and left acromion landmarks on the tips of the shoulders.

BICEPS C!KC1JMFERE34CE, FLEXED (RICIRCFL) -- circumference of the upper arm at the level of the ilcxed biceps point measured perpendicdar to the long axis of the arm.

BIDELTOID BREADTli (BIDL.BDTII) -- maximum horizontal dista ?cc between the lateral margins of the Lpper 2rms on the deltoid muscles.

BIMAI-I XOIAR HUEADTI{ (RIMI3DTtQ - horizontal distance between the maximum protrusions of the ankle. hones ( ~ ~ e d i a t and lateral malleoli). .

BlSPlNOUS BREADTH (BISBDTH) -- distance between the right and left anterior superior iliac spine landmarks.

BITRAGION CHIN ARC (BITCHARC) -- surface dic,,i?ce between the right and !eft tragion landmarks on the cartilaginous flaps in fror,. of the esrholss across the chin landmark.

BITRAGION CORONAL ARC (BI KOARC) -- surface Lrance between the right and left tragion landmarks on the cartilaginous flaps in front of the earholes across the top of the head in a corona! ?lane.

BI'IXAGION CRINION ARC (BITCKARC) -- surface distance between the right and left tragion landmarks on the cartilaginous Caps in front of the earholes across the top of the forehead at the crinion landmark.

BITRAGION FRONTAL ARC (3ITFRARC) -- surflce distance between the right and left tragion landmarks on the cartilaginous flaps in front of the earholes across the forehead just above the ridges of the eyebrows.

BITRAGION SUBMANDIBULAR ARC (BITSMARC) -- surface distacce between the right and left tragiotl landmarks on the cartilaginous flaps in front of the earholes across the submandibular landmark at the juncture of the jaw and the neck.

BITRAGION SUBNASALE ARCBITSNARC) -- surface distance between the right and left tragion landmarns on the cartilaginous flaps in front of the earholes across the subna\rtl iandmark.

BIZYGOMATIC BREADTIi (BIZI3DTH) -- maximum horizontal breadth of the face between the zygomatic arches.

BUSTPOINT/TIlEI_ION-B[;STPOlNTfliELION BREADTII (BSTPTBR) -- distance between the right and left bustpoints on women and the center of the nipples (thelion) on men.

UUTT'OCK CIKCIJMFIIKI3CE (BU-TTTCIRC) -- horizontal circumference of tho trunk at the levcl of the maximum protrusion of the right buttock.

BU'TKXK DEYTII ( I l U ' ~ 1 1 7 I ) -- horizontal depth of the torso at the level of the maximum protru\ion of the right huttock.

Il1JTI'OCK I i111Gl 1.1' (IlUTI'I iGi 1'1') -- vertical distance between the standing surftrce and the level of the maximuni protrusion of the right huttock.

BUTTOCK-KNEE I,EK.;GTI! (f3U7TKLT11) -- horizontal distance between the most posterior point on cither buttock and the front of the knee u measured in the sitting position with the knccs flcxcd 00 dcgrecs.

BU'lTOCK-POPLITEAL LENGTH (BUTTPLTH) - horizontal distance between the most posterior point on the buttock and the back of the knee as measured in the sitting position with the knees flexea 90 degrees.

CALF CIRCUMFERENCE (CALFCIRC) - maximum horizontal circumference of the calf.

CALF HEIGHT (CALFHGHT) -- vertical distance between the standing surface and the level of the maximum circumference of the calf.

CERVICALE HEIGHT (CERVHGHT) - vertical distance between the standing surface and the cervicale landmark at the back of the neck at the maximum point cjf

quiet respiration.

CERVICALE HEIGHT SITTING (CERVSlT) -- vertical distance between the sitting surface and the cemicale landmark on the back of the neck.

CHEST BREADTH (CHSTBDTH) - maximum horizontal breadth of chest at the level of the bustpoint/thelion.

CHEST CIRCUMFERENCE (CHSTCIRC) -- maxirnzrn horizorltal circumference of the chest ~t the level of the bustpoint on women and the nipple on nen.

CHEST CIRCUMFERENCE AT SCYE (CHSTCISC) - horizontal circumierence of the chest at the level of the sqe-at-midspinelandmark.

CHEST CIRCUMFERENCE BELOW BREAST (CHSTCB) - horizontal circumference of the chest at the level of the inferior juncture of the lowest breast with the rib ccrge.

CHEST DEPTII (CHSTDVII-I) -- horizontal distance between the chest at the level of the bustpoint on women and the nipple on men, and the back at the same level.

CHEST HEIGHT (CHSTHGHT) - vertical distance between the standing surface and the bustpoint on women and the nipple on men.

CROTCH HEIGI [T (CRCHHGHT) -- vertical distance between the standing surface and the crotch.

CROTCH LENGTH, NATURAL INDENTATION (CRCHLNI) -- distance through the crotch between the abdomen and back at the level of the natural indentation of the waist.

CROTCH LENGTH, OMPHALION (CRHLOM) -- distance through the cmtch between the navel and the same level on the back ai the waist.

CROTCH LEiNGTH, POSTERIOR NATURAL INDENTATION (CRLPNI) -- surface distance from the crotch at the inner thigh landmark to the back of the waist at the natural indentation of the waist.

CAOTCH LENGTH, POSTERIOR OhlPHALION (CRLPOM) -- surface distance from the crotch at the inner thigh landmark to the bark of th;.: waist at the omph i 1' Ion.

EAR BREADTH (EARBDlTi) -- maximum breadth of the ear perpendicular to its long axis.

EAR LENGTH (ENILGTH) - length of the ear from its highest to lowest points on a line parallel to the long axis of the ear.

EAR LENGTH ABOVE TRAGION (EARLTRAG) -- distance from the tragion landmark to the top of the ear on a line parallel to the long axis of the ear.

EAR PKOT!?USION (EARPROT) -- horkontal distance between the mastoid process and the outside edge of the right ear at its most lateral point.

ELBOW CIRCUMFERENCE (ELBCIRC) -- circumference of the elbow in a plane perpendicular to the long axis uf !he arm at the level of the olecranon center landmark, with the arm straight at the side.

ELBOW REST HEIGHT (ELRHGHT) -- vertical distance between the sitting surface and the olecranon landmark on the bottom of the flexed elbow.

EYE HEIGHT, SIITING (EYEli'TSIT) -- vertical distance between the sitting silrface and the ectocanthus landmark at the outer corner of the eye.

FOOT BREADTH, IiORIZONTAL (FTBRHOR) -- maximum breadth of the standing foot between the first and iitth metatarsophalangeal landmark protrusions.

FOOT LENGTH (FOOTLGTII) -- distance between the tip of the longest tce and the back of the heel of the standing foot.

FOREAKM CIRCUMFERENCE, FLEXED (FCIRCFL) -- niaximum circumference of the forearm just above the elbow crease with the elbow flexed 90 degrees and the fist tightly clenched.

~-OKI' ,A~M-I~OI<I~IIKM I3U EAL>'i'l I (I~OKf~OKIlR) -- maximum horizontal distance bctwecn the lateral right forcarrn and the lateral left forearm.

FOREARM-IfAN[> 1,EKGTI 1 (FOK11D1,G) -- horizonta: distance between the back of the tip of the elbow to the tip of the middle finger.

FUNCTIONAL LEG LENGTli (FNCLEGLG) -- straight-line distance when seatcd with the leg ex!ended bctwecn the footrest surface of the anthropometer and the posterior wrface cf the body.

GLUTEAL FURROW HEIGHT (GLUFURHT) -- vertical distance between the standing surface and the lowest point of the gluteal furrow under the buttocks.

HAND BREADTH (HANDBRTH) -- maximum breadth of the hand between the metacarpal I1 and metacarpal V.

WAND CIRCUMFERENCE (HANDCIRC) -- maximum circumference of the hand at the level of the metacarpal 11 and metacarpal V.

HAND LENGTH (HANDLGTH) -- length of the hand between the stylion landmark on the wrist and the tip of the middle finger.

HEAD BREADTH (HEADBRTH) -- maximum horizontal breadth of the head above the attachment of the ears.

HEAD CIRCUMFEXENCE (HEADCIRC) - maximum circumference of the head above the attachment of the ears and ridges of the eyebrows.

HEAD LENGTH (HEADLGTH) -- maximum iength of the head between the glabella landmark and the opisthocranion.

HEEL ANKLE CIKCUMFERF"" 3 (HLAKCIRC) - circumference of the foot at the ank!e and base of the heel.

HEEL BREADTH (HEYLBRTH) -- maximum horizontal distance between the medial and lateral poi:rts on the inside and outside of the heel.

HIP BREADTH (HIPBRTH) - horizontal distance between the hips at the level of the lateral buttock landmarks.

HIP BREADTH, SITnNG (HIPBRSIT) -- lateral maximum hip or thigh breadth (whichever is broader) of a seated subject.

ILIOCRISTALE HEIGHT (ILCRSIT) -- vertical distance between the standing surface and the iliocristale landmark on thetop of the right side of the pe!vis.

INTERPUPILURY BREADTI f (INPUPBTti) -- horizontal distance between the two pupils.

INERSCYE 1 (IPiSCYEI) -- distance acrnss the back between the top of the right and left axillary fdd posterior landmarks.

INTERSCYE 2 (INSCYEZ) -- distance between the right arid left midscye landmarks on the back.

KNEE CIRCUMFERENCE (KNEECIRC) -- horizontal circumference of the knee at the level of the midpatelfa landmark (standing).

KNEE HEIGHT, MIDPATELLA (KNEEHTMP) -- vertical distance between the standing surface and the center of the knee at the midpatella landmark.

KNEE HEIGHT, SITTING (KiiEEHTSI) -- vertical distance between the bottom of the planted foot and the suprapatellar landmark (located scanding).

LATERAL. FEMORAL EPICONDYLE HEIGHT (LATFEMEP) -- vertical dista~lce between the standink s d a c e and the lateral femoral epicondyle landmark on the outside of the knee.

JATEPAL MALLEOLUS HEIGHT (LATMALHT) -- vertical distance between the standing surface and the lateral ma!leolus on the outside of the ankle.

LOWER TEIGI I CIRCUMFZRENCE (LOT'CIRC) -- horizontal circumference of the thigh at the level of the supra?atellar lardmark.

hlENT3N-SELLION LENGTH (MENSELL) -- distance between the inenton landmark at the bottom of the chin and the sellion landmark at the deepest point of the nasal root depression.

MIDSHOULDER IIEIGHT, SITTING (MSHTSIT) -- vertical distance between the sitting surface and !he midshoulder landmark at the middle of the top of the right shou!der.

NECK-BUSTPOINT/THELlON LENGTH (NKBPLGTH) -- distance between the trapezius landmark at the side of the neck and the bustpoint landmark on women or the nipple o r r n x n .

NECK CIRCUMFERENCE (NECKCIRC) -- circumference of the neck at the infrathyroid landmark (Adam's apple).

NECK ClfiCUltiFERENCE, RISE (NECKCRCB) -- circumference at the base of the neck at the anterior and lateral rieck landmarks.

NECK f IEIGIIT. IA'ITRA!, (NECKliTLT) -- vertical distance between tte standing surface and the trapezius landmark at the side of the ncck.

OVER1 IEAI) FINGERTIP REACI? (OVl iDF.TR1 i ) -- verticd distance between the stantfiny: wr fxe and the tip of thc right middle fingerwhen the arms are cxtcntJcd ovcrt1:aci ;rnd me;rwred again5t rhc wall. The cuhject stands facing a wall- rnounted sc;ilc with t>o~ti arms extcndcd overhead pi~rallcl to c;tch other. The toes are 20 crn from the w;ill and the feet are about 10 cm apart. l'llc palms o f the hantls rest on the scale. A block is p!aced apiinst the tip of the finger tc: establish the measurement. 'I'hc measurement is taken at the maximum p i n t of quiet respiration.

OVERHEAD FINGERTIP REACH, EXTENDED (OVHFRHE) -- vertical distance between the standing surtace and the tip of the right middle finger when the arm is extended overhead as high as possible and measured on a wall scale. The subject stands an his/her toes facing a wall-mounted scale with both arms parallel and extended overhead as far as possible. The toes are 20 cm from the wall and the feet are about 10 cm apart. The palms of the hands rest on the scale. A block is plactd against the tip of the finger to establish !he measurement. The maurement is taken at the maximum point of quiet respiration.

OVERHEAD FINGERTIP REACH, SIlTING (OVHDFRHS) -- vertical distance between the sitting surface and the tip of the right middle f ing~r when the arm is extended overhead and is measured on a wall scale. The subject sits erect on a flat surface 40.8 c n high with the right arm and hand extended vertically overhead as far as possible and the palm of the band facing forward Neither the back nor the arm touches the wall. A block placed at the tip of the middle finger spans the distance between the finger and the wall and establishes the measurement on the wall scale. The meawrement is made at the maximum point of quiet respiration.

POPLITEAL HEIGHT (POPHGHT) -- vertical distance between the foot surfax 2nd the bottom of the thigh just behind the knee. The subject is seated with the thighs parallel and the knees flexed 90 degrees.

RADIALE-STYLIOFi LENGTH (RASIZ) -- distance between the radial2 landmark on the elbow and the stylion landmark on the wrist.

SCYE CIRCUMFERENCE (SCYECJRC) - vertical circumference of the upper arm measured with a tape through the armpit passing over the acromion landmark on the tip of the sko:lldcr.

SCYE DEmf (SCYEDPTH) -- vertical surface distance along the spine between the cervicale landmark on the back of the neck and the scye level at midspine landmark.

SHOULDER CIRCUMFERENCE (SHOUCIRC) -- horizontal circumference of tL;e sh->ulders at the level of the maximum protrusion of the right deltoid muscle.

SHOULDER-ELBOW LENGTH (SHOUELLT) - distance between the acromion landmark on the tip of the shoulder and the olecranon landmark at the hottom of the elimw flexed to 90 degrees.

Si IOULDER lXNGI31 (Sf 1OULG'TI I ) -- surface distance between the trapezius la1 dmark at the base of the neck and the acromion landmark at the iip of the shoulder.

SITTING HEIGHT (SITTIGHT) -- vertical distance between the sitting surface and the top of the head.

SLEEVE LENGTl1: SPINE-ELBOW (SLLSPEL) -- horizontal surface distance between the midspine landmark and the olecranoqcenter landmark at the tip of the raised elbow. The measurement is made while the arms are held in a horizontal position, parallei to the standing surface, and joined by bringing the fists together.

ctmce SLEEVE LENGTH: SPINE-SCYE (SLLSPSC) -- horizontal surface di.. between the midspine landmark and the posterior-diagonal-scye lrindmark at the back of the raised right arm. The measurement is made whi!e the arms are held in a horizontal position, parallel to the standing surface, and joined by bringing the fists together.

SLEEVE LENGTH: SPINE-WRIST (SLLSPWR) -- horizontal surface distance from the midspine landmark, acrcss the olecranoqcenter landmark at the tip of the elbow to the dorsal wrist landmark. The measurement is made while the arms are held in a horizontal position, parallel to the standing surface, and joined by bringing the Ests together.

SLEEVE OUTSEAT4 (SLOUTSM) -- straight-line distance between the acromion landmark on the tip L; !he shoulder and the stylion landmark on the wrist, measured with the arm is straight a! the side and the palm facing foruard.

SPAN (SPAN) -- distance between the tips of the third fingers when the arms are stretched out horizontally.

STATURE (STATURE) -- vertical distance between the standing surface and the top of the herrd.

STRAP LENGTH (STRLGTH) -- distance from the right hustpoint for women or nipple for men over the back of the neck to the left bustpoint or nipple. The tape passes over the leh and right lateral neck landmarks.

SUPKASTERNALE HEIGHT (SUPSTRHT) -- vertical distance between the standing surface and the suprasternale landmark at the lowest point of the notch at the top of the breastbone.

E;. (TlI RIB IiEICIiT (TENRIBIIT) -- vertical distance betvieen the standing surface and the tenth rib landmark at the bottom of the ribcage.

'rlIIG11 (7lRClJMf;T~Ki~NC'E (T11Gi1CIKC) -- circumference o f the thigh at it3

juncture with the buttock.

'1'1 11(;11 C2.13KANCI3 ('1'1 I(; I K L K) -- vertical distancc hetwcen the s i t t iq surface and the higheht point on the top of the thigh.

TI1UMB BREADTli ( T l i U M ~ I ~ K ) -- muimum breadth of the thumb perpendicular to its long axis.

107 THUMBTIP REACH (THbiBTPR) - horizontal distance between a wall against which the posterior trunk is in contact and the tip of the thumb when the arm is extended anteriorly.

. : TROCHANTEXION* HEIGHT (TROCHHT) -- vertical distance between the ~tanding surface and the trochanterion landmark on the hip.

In earlier publications also referred to as trochanteric height.

109 VERTICAL TRlJNK Cl RCUMFERENCE (Assc)* * (VTCASCC) -- vertical circumference of the trunk on a line passing between the buttocks and through the ~ y t c h , and over :he bustpoint/thelion landmark and midshoulder landmark.

Aircrew Standardization Committee

VERTICAL TRUNK ClRCUMFERENCE (USA)**' (VTCUSA) -- vertical circumference of the trunk on a line passing over the maximum protrusion of the L,r t o c ~ s and through !he crotch, and over the bustpint/thelion landmark and gj$shoulde~ landmark.

U. S. Army.

WAIST BACK LENGTH, NATLXAL INDENTATION (WSTBLNI) -- vertical surface distance txtween the cervicale landmark on the hack of the neck and the posterior waist (natural indentation) landmark.

WAIST BACK L.ENGTH, OXlPllALION (WSTBLOM) -- vcrtical surface distance between the cervicale landmark on the back of the neck and the posterior waist landmark at the level of the navel (omphalionh

WAIST l3READT)I (WSTBRTIi) -- horizontal breadth of the waist at the level of the center of the navel (ornphalion).

WAIST CIRCUMFERENCE, NATURAL INDENTATIOS (WSCIRCNI) -- horizonta! circumference at the level of the natural indentation.

WAIST CIKClJhlFI3KENCE. OXiPtiALION (WSCIRCOiM) -- horizontal distance ;irou;~d thc torw ut ~ h c level of the center of the nave! (ornphalion).

WAIST' I>lilyI'I{ (WSTDIJPTII) -- horizontal distancc txtween the front and back of the waist ;it thc level of the ccntcr of the n;ivcl (omphalion).

WAIST FKO!JT I.[:N(;TlI, NATURAL INDENTATION (WSTFKLNI) -- vertical surf;lce di\t;~ncc hctwctn thc nntcrior ncck landmark at the front of the ncck and the anterior wakt (n;ituraI indcnt.ltion).

WAIST i.XOluT I i:N<;Tii, OMPtIAC.ION (WSTFR1,OM) -- vertical surface dktancc hctwecn the anterior neck landmark and center of the navel (omphalion).

120 WAIST HEIGHT, OMPHALION (WSTHOM) - verrial distance between the standing surface ;nd the center of the navel (omphalitln).

121 WAIST HEIGHT. S I I T N G , NATLRAL INDENTATION (WSIITSTNI) -- vertical distance between [he sitt~ng surface and the right natural indentation of the waist.

122 WAIST tfElGli'T, S I T I S C , O?.lPtiAL-ION (U'SliTSTO\I) -- bertical distance between the sitting surface and the center of the navel (omphalion).

WAIST-HIP LENGTH (WSHIPLTH) -- vertical distance between the right waist landmark (ornphalion) and the rigtit lateral buttock landmark.

WAIST, NATURAL INDENTATION -- WAIST, OMPHALION LENGTH (WSNIWSOM) -- s ~ r f a c e distance between the right waist (natural indentation) landmark and the rignt waist (omphalion) landmark.

WEIGHT (WEIGHT) -- in centigrams (. 1 kilograms).

WRIST- CEYTEd OF GRIP L E S G l l I (WRCTRCRL) -- horizontal distance between the stylion landmark on the wrist and the hole in the center of the gripped dowel

WRIST CIRCLr!dFERESCE (WRISCIRC) -- circumference of the wrist perpendicular to the long axis c,f the farearm at the level of thc stylion landmark.

'.'i RIST HEIGI i T (N HIStICf fT) -- vertical distance hctwcea the standing surface and the stylion 1:lndmark on the wrist when the arm is held straight down.

WRIST tiEiGIl'T, S I l 7 1 S G (WRISi1TST) -- vertical distance between the floor and !he stylion landmark on the wri\t of a ceated suhject uhen the arm is held straight down. Sote that the scat height u\ed in measuring was 45.5 cm.

WRIST-1;C'nS;Y FINGER I.I:NG?Ii (WH19i~?1GI.) -- di\tancc hetween the stylion o n the wrnt and the tlp of the Index fincer.

W I ~ I S ~ I ' ~ W A I . l . I I . S ( ; ' l I i (WKW:\I I 1 N) -- horimnt;~l tli\r;~rlcc hctuc.cn a wall .rp,linv wh~ch the po\lcrlor trunk is In ccmt;ict and thc \tylron I;tnclriiark on tllc writt when thc :.rnt i \ cxtcnded anteriorly .ind the hutttuks and shoulders are ;~gain\t the WII.

1 ne rollowng measurements are iuusrrareu In v w a r rnuex - neau mcasurcmrnl~, pp. 50-51.

BIGONIAL BREADTH HEADBOARD (BIGBRH) - straight-he distance between the right and left gonion landn-ia;b at the corners of the jaw.

BIINFRAORBITAL BREADTI3 HEADBOARD (BIINORBH) -- straight-line distance between the right and left infraorbitale landmarks at the bottom edge of the bony eye socket.

BIOCtJIAR BREADTH MAXI MUM HEADBOARD (BIBCBRMH) -- straight- line distance between the right and left ectoorbitale landmarks behind each boay eye socket at the level of the outer corners of the eyes.

BITRAGION BREADTH HEADBOARD (BTRBDTHH) - straight-line distance between the right and left tragion landmarks on the cartilaginous flaps in front of each earhole.

BIZYGQMAIIC BREADTH HEADBOARD (BIZYBRH) - straight-line distance between the right and left zygion landmarks at the most lateral point of the zygomatic arch.

LIP LENGTH HEADSOARD (LIPLGTHH) - straight-line distance between the right and left cheilion landmarks at the corners of the mouth.

MAXIMUM FRONTAL BREADTH HEADBOARD (hftLYFROZJH) -- straight- line distance between the right and left zygofrontale landmarks at the upper margins of each bony eye socket.

MENTON-CKINIOPJ L,ENGTI1 i1EADBOARD (MENCRINii) -- straight-line distsnce between the menton landmark at the twttom of the chin and the crinion landmark at the lowest point of the hairline on the forehead.

MENTON-SELLION LENGTH 11EADBOARD (MENSELLH) - straight-line distance between the mcnton landmark and the sellion landmark at the deepest point of the nasal root depression.

MENTON-St!WJASAI.E LXNGTIt IIEADBOAKD (MENSIJBNIi) -- straight- line distance between the menton landmark on the twttom of the chir, and the suhnasale landmark under the now

MINlMtJM FHONTAI, iIHtiA1Y111 1tt<ADi30AKD (MINFRONII) -- straight- line distance hcfwecn the rrght and left fronto~ct~i~ntt;rle lantlmarks on thc tcmp~ral crest.. on each tide of the forehead.

NOSE BREADTI1 IIEADROARD (NOSEBRTli) -- straight-line distance between right ;ind left ahre 1andm;lrks on the ~icles of the nosttiis.

224 NOSE PROTRUSION HEADBOARD (NOSEPRH) - straight-line distance between the pronasale landmark on the tip of the nose and the subnasale landmark under the nose.

225 SUBNASALE-SLLLION HEADBOARD (SBNSSELH) -- straight-line distance between the s u b n ~ u l e landrnark under the nose and the sellion landmark at the deepest point of the n ~ s i l root dttprrs\ion.

226 A M E TO BACK o r i;LXI) ( A ~ 9 H k d j -- iiorizontal disiaix: Z-?Pw@n the alare landmark on the side of ?he ncstrils and the vertical plane tangent to the back of the head.

227 ALARE TO TOP OF HEAD ( U R E T ) -- vertical distance between the alare landmark at the side of the nostril and the horizontal plane tangent to the top of the head.

228 CHEILION TO BACK OF HEAD (CIiEILU) -- horizontal distance between the cheilion landmark at tke corner of the mouth and the vertical pime tangent to the back of the h e d .

229 CliEILION T O TOP OF 1iEAD (('1114LT) --vertical dihtance between the cheilion landmark at the corner of the mouth and the horizontal plane tangent to the top of the head.

2.30 CRINION TO R A f K OF 11EAD (CIIINIONX) -- horizontal distance bc,tween the crinion landmark at the lowe5t point c:i the hairline on the forehead and the vertical plane tangent to thc back of'the head.

23 1 CRINION TO TOP OF l1EAD (CKINIONZ) -- vertical distance between the crinion landmark at the lowcst point of the hairline on the forchead and the horizontal planc tancent to the top of the head.

2 lJ(n'~>OKI3ITA1.I1 ' 1 -0 I3AClr; O F t 4 I i : l I ) (1X'TORIlIl) -- horizontal distance between the ccttw)rhitale landrnark Iwhind the bony eve socket at the Izvcl of the outcr corncr of thc clic and thc vertical pl;rnc tangent to the hi1c.k of thc hcad.

GLABELLA TO BACK OF HEAD (GLABX) -- horizontal distance between the glabella landmark on the forehead between the eyebrows and the vertical plane tangent to the back of the head.

GLABELLA TO TOP OF HEAD (GLABZ) - vertical distace between the g!abeila landmark on the forehead between the eyebrows and the horizontal plane tangent to the top of the head.

GONION TO BACK OF HEAD (GONIONB) -- horizontal distance between the gonion landmark at the comer of the jaw and the vertical plane tangent to the back of the head.

GONION TO TOP OF HEAD (GONIONT) - vertical distance between the gonion landmark at the corner of the jaw and the horizontal plane tangent to thc top of the head.

INFRAORBITALE TO BACK OF HEAD (INFORBE) -- horizontal distance between the infraorb;tale landmark on the bony eye socket under !he eye and the vertical plane tangent to the back of the head.

INFRAORBITALE TO TOP OF HEAD (INFORBT) -- vertical distance betweerl the infraorbitaie landmark on the bony eye socket under the eye and the horizontal plane tangent to the top of the head.

MENTON TO BACK OF HEAD (MENTONX) -- horizontal distance between the menton landmark at the bottom of the chin and the vertical planc tangent lo the back of the head.

MENTON TO TOP OF HEAD (MENTONZ) -- vertical distance betwecn the menton landmark at the bottom of the chin and the horizontal plane tangent to the top of the head.

PROMENTON TO BACK OF HEAD (P,MENTONX) -- horizontal distance between the promcnton landmark at the chin and the vertical plane tangent to the back of the head.

PROMENTON TO TOP OF ,';AD (PMENTONZ) - vertical distance between the promenton landmark at the chin and the horizontal plane tangent to the top of !he head.

PRONASA1.I; TO BACK OF i 1 M D (PRONASX) -- horizontal distance between the pronasale landmark at the tip of the nose and the vertical plane tangent to the back of [he head.

PRONASALE TO TOP OF i1EAD (PHONASZ) -- vertical distance hetween the pr.nnawle landmark at the tip of the nose 3 r d the horizontal plane tangent to the top of the head.

SELLION TO BACK OF HEAD (SELLIONX) - horizontal distance between the sellion landmark at the deepest p i n t of the nasal root depression and the verticd plane tangent to the back of the head.

SELL;: I N TO TOP OF HEAD (SELI-10x2) -- vertical distance between the sellion landmark at the deepest point of the nasal root depression and the horizontal plane [angent t i ) rhe top of the head.

STOMION TO BACK OF HEAD (STOMIONX) -- horizontal uistance between the stornion landmark at the center oi the mouth in the midsagittal plane and the vertical plane tangent to the back of ihe head.

STOMION TO TOP OF lIEAD (STOhIIONZ) --vertical distance. beween the stornion landmark at the center of the mouth in the midsagittal plane and the horizontal plane tangent to the top oi the head.

SUBNASALE TO BACK OF tIEAD (SUBNASX) -- horizontal distance between the subnasalz landmark under the x s e and the vertical plane tangent to the back of the head.

SUBNASALE TO T(>P OF 11E.4D (SUBNASZ) -- vertical distance between the subnasale landmark under the nose and the horizontal plane tangent to :he top of the hcad.

TRAGION TO BACK OF I1EAD (TRAGD) -- horizontal distance between the tragion landmark on the cart~laijnous flap in front of the earhole and the vertical plane tangent to the hack of the head.

TRAGION 'TO TOf' O F t II3AD (TKAGT) -- vertical distance between the trayon landmark on the cart~l;tg~r:ou\ flap in front oi the earhole and the horizontd plane tangent to the :op ot the head.

Z Y G I O N TO 'TOP 0 1 : 1 l L l I ) (ZYG.1') - - bcrt~cal di:;ance hctwcen the q)lt;ion landmark on ~ h c ngomatlc arch and the tior~mntal plane tan,wnt to the top of the ! x . l ( l .

VISUAL INDEX - THE STANDARD MExSVREMENTS (Continued)

(7) m u IEEIorC

( 2 3 ) DU,WDflTff ELIC(J-U;>~I;~~~~IELIU.I

( 0 4 ) OVEnIFAD FDK;L'1L?P RLj',ffI

( 8 5 ) OVERIm FIZcEm"?P w EXx'mEEt'D

( 8 6 ) mT;EXrTP KIII\CH s m J C (102) S E W S E R U K E I C ; t F

(119) WAIST EIGfI', K X L + C , m a l

(120) man', CHI1V\LIOE1

(123) WRI= IEEIQff STT,nrJ

VISUAL INDEX - THE STANDARD MEASt'REME3TS (Continued)

VISUAL LNDEX - THE STASDXRD MEASUREMENTS (Ccntinoed)

VISUAL N E X - THE STANDARD MEASUREMENTS (Cdnrinucd)

VISUAL INDEX - THE STANDARC MEASUREMENTS (Ccntinued)

VISUAL XNDEX - THE STANDARD MEASUREMENTS (Continued)

VISUAL INDEX - THE STANDARD MEASUREMENTS (Continued)

(45) EiCR mm (46) EAR UZXITH ABCrJE TRACIW

(47) EAR FlxmuEIW

(35) s m m: SPINE-Elx%J (96) SLECVE UXXW: SPINE-SCYE

(37) s m XBGTH: SPLNE-WRIST

VISUAL INDEX - THE STAh'DARD .MEASUREMENTS (Continued)

V ! S U AL hi EX - THE

VISUAL INDEX - HEAD MEASUREMENTS

INDEX

Entries in a ital letters are anthropometric measurements and those in P mixed capitsl and ower w e letters represent landmarks on the body. All measurements given in the ~ndcx have pages identified in a articular order. The first page number is for the measurement def i t ion foilowel by the Visual Index, the male bivariate regression table, the female bivariate regresson table, the male stepulse regression table, and the female stepwise regression table. Any additional e~t r ies are for the standard muitiple regression tables. Only dependent variab!es are indexed for the bivariate and ste wise regression tables while 50th dependent and independent variables are identi I! led in the standard multiple re ressiou tables. Entries are not provided for :he correlation tables since each van'a le appears on every third page ill each of the 10 tables.

i

Part 1 of this sen- 11 pp 1 - 51, Part 2 ir pp. 53 - 312; Part 3 is pp. $13 - 572; F u t 4 I. pp. 57s - 853; Part 5 is pp. 854 - 984.

ABDOMINAL EXIESSION DEPIH, SRTlXG (ABEXDPST, 2)

Abdominal point, anterior

ACROMIAL HEIGVT (ACRHGHT, 3)

ALAWK TO TOP OF t i f AD (AIARET, ZT)

ANKLE CIRCUMFERENCE (AYKLClRC, 6)

Anterior superior &ac spine

rVULLA HEIGHT ( M G m , 7)

AXIUARY ARM CIRCUMFERENCE (AXARCIRC 8)

kril)ary ld4 poat erior

BALL OF FOOT CIRCUMFERENCE (BLFTCIRC, 9)

BALL OF FOOT LENGTH (BLFIISTH, 10)

BLAcROMIAL BREADTH (BCRMBDTH, 11)

BICEPS CIRCZ1'MFERENCE, FtEXED (BICIRCR, 12)

B i u p p>int

RIDELTOID RRFADTIi (BIDIBDTH, 13)

BIOCl.lLAR BREADTH MAXIMUM H W B O A R D (BIOCBRMH, 214) 3650, m, 833, 889,935,955, 964

BlSPlNOUS BREADTH (BISBDTH, 15)

BITKAGION BREADTH HEADBOARD (BTRBDTHH, 215)

BrraAGION CHIN ARC (BITCHARC, 16)

BITRAGION CORONAL ARC (BITCOARC 17)

BITR4GlON CRlNlON ARC (BITCRARC, 18)

BI'TMGION FRONTAL ARC (BITFRQRC, 19)

BITRAGION SUBMAi4QIBUW ARC (BI'EiHAWC, r))

rut pna 14

n, u. sw, 739, MI, 907. Pro, 950954,95cl-'W,3

BL?TOQI- WPLITUL LENGTH (B-TH, 2)

CALF CIRCUMFERENCE (CALFCnRC 3)

CHEST BREADTH (CHSTBDTH, 33)

CHEST CIRCUMFFRENCE (CHSTCIRC, 34)

CHEST CIRCUMFERENCE AT SCYE (CHSTCISC, 35)

CHEST CiRCUMFERENCE BEtOW BREAST' (CHSTCE, 36)

CHEST DEPIW (CHSTDPTH, 37)

CHEST HEIGHT (CHSTHGHT, 3)

Chin

Clavicle

Crinion

CRINlON TO BACK OF HEAD (CRINIOWX, 232)

CRiPilON TO TOP OF HEAD (CRINIOHZ, 231)

CROTCH LENCi'M, NATURAL IHDfNTATlON ( C R C H M , 40)

CROTCH LENGTH, O M P W O N (CRHLOM, 41)

CROTCH LENGTH, POSTERIOR NATURAL 1SD (CRLPM, 42)

CROTCH LENGTH, POSTERIOR OMPPAION (CRLPOM, 43)

Dactylion I1

Dactylioa 111

Delroid point

Dorsal juncture d al l and th@

Dorul jundwc of fmt and Icg

Ear bO(tm

WR 3R EADTIi (EARBDIH, C4)

EAR LENGTH (EARLC,T)i, 45)

E A R LENGTH ABOVE TRAGiON (FARLTRAG. 46)

FRONTOTEMPORALE TO 'IY)P OF HEAD (FRTEhC1.235)

FUNCJ7ONA.L LEG LENGTH (FNCLEGU;, 36)

GbfJh

G L A B U TO BACK OF HEAD ( G U B X , 226)

G l A E U U TO TOP OF HEAD (GLABZ 237)

G L V r U L FURROW HEIGHT (GLUNRH1; TT)

Gluteal humw point

Go&

GOMON TO BACK OF H W (GONIOFCB, 238)

GOMON TO TOP OF HEAD (GONfOKT, 234)

HAND BREADTH (IUNDBRM, 9)

HAND CIRCUMFERENCE (HAMXnRC, 59)

HAND LENGTH (tUNDLGTH, 60)

HEAD BREADTH (H W B R T H , 61)

ECrOORBiTALE TO BACK OF HEAD (ECTORBE, 232)

EC1'00RBITAl.E TO TOP OF HEAD (ECTORBT, 233)

ELBOW CIRCUMFERENCE (ELBCIRC, 4)

Elbow creasc

ELBGW R E S HEIGHT (ELRHGHT, 49)

EYE HEIGHT, SiTTlNG (EYEHTSIT, 50)

Fd~h metatarsophalangeal pro(-usioa

Fu31 met atarsopbalangeal protrusion

FOOT BREADTH, HORIZONTAL (FTBRHOR, 51)

FOOT LENGTH (FOOTU;TH, 52)

).'ORW\RM CIHCLPIFERENCE, FLEXED (FCIFCR, 53)

FOREARM-FOREARM BRCADTII (FORFORBR, 51)

FORFARM }(AND LEN(iTt4 (FT)RtlDL(;, 55)

37, M, M I , MI, 894, W, 954 %3

KNEE E I G H T , SITTING (KNEEKISI, 74)

Knee point

Lateral femoral cplcrpadylr:

LATERAL FEMOaAL EPICOMDYE HEIGHT (UTFEMEP, 75)

LA': ERAL MALLEOLUS HEIGHT ( U M , 76)

LIP LFiN<iTH H ~ W B O A R D (UPISTl iH, 211)

LOWER THIGH CIRCUMFERENCE (UTHClRC 7?)

MAXIMUM FRONTAL BREAMH HEADBOARD (MAXFROMi, 218) %,SO, '101,835, m, 456,955, %4

HEAD CIRCUMFEREMCE (HEADCIRC 62)

H W D LENGTH (HLUlLGTH, 63)

HEEL ANKLE ClRCUMERENCE IHLAKUIRC, S j )

HEEL BREADTH (HEELBRTH, 65)

Hccl poin~

HIP BRWDTH (HIPBRTH, 66)

HIP BREADTH, SITTING (HIPBRSTT, 67)

;Sf RAORBITALE T O TOP O F HFAD (INFORBT, 241)

Idrathyoid

111ncr I high point

ImRPIJPIL.IARY BQFADni (lNPlJPU13i. M )

Mcnton

MENTON TO BACK OF HL4D (MEhTONX, 242)

MESTON-SELLION LENGTH HEADBOARD (MENSEU.H. 220)

ME?FTON-SEUION LENGTH (MENSELL, 7Hj

MEhTON-SUBS4SAE LENGTH HEADBOARD (MESSUBS'H, 221)

NOSE PROTRIJ'SION HEADBOARD (W\FPRW. 2 4 1

PRONUALE TO BACK OF HEAD ( P R O M U6)

PRONASALE TO TOP OF H W ( P R 9 N G Z 247)

Ptermm

R&

RADIALE-STYLION LENGTH (RAST1,88)

\ELUOH TO TOP OF HEAD (SEUJONZ 3 ' 3 )

' 4 tOI1U>t 'R C'TR(I,'Mn.RtNCE (\tiOt.:<'IRC', 31)

38, so, 7l3.847, 897,943.9.56, 955

SITl-ING HEIGHT (Sl7THGHT. 94)

SLEE\*E LESGTH SP!%E-ELBOW' IS!..ISYEL. 95)

SUBNASALE TO TOP OF HEAD (SLJNASZ 253)

SupTapale~

Suprasterr&

5LPFUsrZRNALE HEIGHT (SLPSTRHT, l E )

Tcntb Rib

T E . W RIB HEIGHT ('TENRIBHT. 103)

Tr+

lT-AC>iON TO BACK OF HEAD m G B , 25-8)

Trapezius Point

TrtKhnttr

T R O C ~ ~ R l O V HEIGHT (TXOCl.'CT, I@)

VERTICAL TRUNK CIRCZ'MFERENCE (ASSC) (WTASC'C, 1W)

VERTICAL TRUNK (3RCI;UFERESCE (USA) (\7CCSA, 110)

WAIST HEIGHT, NATURAL XNDLM'ATIOH ~ ~ S I H N 1 , 1 1 Y )

W A I S HEIGHT, OMF RAUO;"' (NSTHOM 120)

WRIST HEIGHT, SITnNG (U'RISKIST, m,

w.rsr-INDEX FINGER ESGTH ( W R I Y ~ G L UO)

WRIST-niUMsTlP LENGTH (UUTHKJTH, 131)

UU IST.WALL LENGTH (Vr'R WALLLN, !32)

%Y(rOf R 0 M A I . E T O 7: ,P O F IiI'AI) (2.YT RT. 2'9)

T> . 1 1 q ,!I,( t ~ r a ~ t r t r r p s ) r t r p r . c r r , !I ~ ~ r i < i ~ r t a k r , n r t t h r

:"; A r m y Hat I I k R e r v a r t h , I ) o v e l ~ t j ~ r n r n t anti I . r i < , ~ ~ ~ r v r i t r ~

I c . n t r r r and i 1 1 1 ) h r r n a n r 1 g n c . ~ 1 YI,. H A T I ( K / I H - ' /, i t 1 1 he n a r ~ e a a ~ f r r p g l r t r a p p r + ~ v r d t t ~ r pt lh l L C a t I o n .

TITLE:

ERRATA

1988 Anthropometric Survey of U.S. Army Personnel: Corre lat ron Coef f r c i e n t s and Regress; on Equations

P a r t s ! through 5.

James Cheverud, C l a i r e C. Gordon, 2obert A. Wal l er , Cashel 1 Jacqui sh, Luc i Kohn, A l l en ttoore, and Nvutr Yamashita

~~3!ica::on Year: May 1990

F s r the asove-rererenced T?chnica: Repsrts, note should be made t h a t P e d d and +ace dtrnensions measured w i t h t5e autcmated headboard davlce &ere veccrCo3 t o t h e nearest 0.: mrl:imeter, not t o t he nearest n l l l i m e t e r ds i nd i ca ted 0 r i p : n r l l y On page 25 of Par t I. Canversion s r x e d u r e s are out: lned on the f c l lcwinq page.

khen ca: iu! at:ng reqress ions using headbcard dlmenslons, conversions from O . ! mm t o 1 rnm a r e necessary:

Headboard dimension = dependent v a r i aS!e Stancard dimensicn independent v a r ~ a b l e

IKote: Wheq Mead3oard dimension i s t h e d-pencent var i n b l e , the Standara Es t imate of Er ror 1 % i n 0.1 mm. 1