Ms Yeni (+62)-8573-5555-759 Swimwear Muslimah Cheap Malaysia
Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they...
Transcript of Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they...
![Page 1: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/1.jpg)
Using Statistics in Social Research
![Page 2: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/2.jpg)
![Page 3: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/3.jpg)
Scott M. Lynch
Using Statistics in SocialResearch
A Concise Approach
123
![Page 4: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/4.jpg)
Scott M. LynchDepartment of SociologyPrinceton UniversityPrinceton, New Jersey, USA
ISBN 978-1-4614-8572-8 ISBN 978-1-4614-8573-5 (eBook)DOI 10.1007/978-1-4614-8573-5Springer New York Heidelberg Dordrecht London
Library of Congress Control Number: 2013945760
© Springer Science+Business Media New York 2013This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part ofthe material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,broadcasting, reproduction on microfilms or in any other physical way, and transmission or informationstorage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodologynow known or hereafter developed. Exempted from this legal reservation are brief excerpts in connectionwith reviews or scholarly analysis or material supplied specifically for the purpose of being enteredand executed on a computer system, for exclusive use by the purchaser of the work. Duplication ofthis publication or parts thereof is permitted only under the provisions of the Copyright Law of thePublisher’s location, in its current version, and permission for use must always be obtained from Springer.Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violationsare liable to prosecution under the respective Copyright Law.The use of general descriptive names, registered names, trademarks, service marks, etc. in this publicationdoes not imply, even in the absence of a specific statement, that such names are exempt from the relevantprotective laws and regulations and therefore free for general use.While the advice and information in this book are believed to be true and accurate at the date ofpublication, neither the authors nor the editors nor the publisher can accept any legal responsibility forany errors or omissions that may be made. The publisher makes no warranty, express or implied, withrespect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
![Page 5: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/5.jpg)
For Mike and Betsy, my parents
![Page 6: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/6.jpg)
![Page 7: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/7.jpg)
Preface
This book began as a collection of lecture notes and has evolved over the lastseveral years into its current form. My goal in putting the book together was toremedy some problems I perceive with extant introductory statistics texts. To thatend, this book has several features that differentiate it from standard texts. First,typical introductory statistics texts are now commonly over 500 pages long. To beblunt, I don’t believe that students actually read texts when they are so lengthy.So, although a 500C-page book may be chock-full of examples, illustrations, andinteresting side information, such detail is offset by the fact that students will notread it. Here, I have tried to provide limited, but sufficient, examples of concepts andto keep the actual text limited to about 200 pages. In doing so, I have not covered anumber of statistical methods that are not commonly used, like Fisher’s exact test,Kruskal’s Tau, and the Spearman correlation. While these, and similar, methods arecertainly still used, they have limited application in contemporary social research.Thus, they can be investigated in detail by students when needed in the future (e.g.,for a thesis or other research paper) via other sources.
Second, in this book, I integrate statistics into the research process, bothphilosophically and practically. From a philosophical perspective, I discuss thedifficulties with concepts like “proof” and “causality” in a deductive scientific andstatistical framework. From a practical perspective, I discuss how research papersare constructed, including how statistical material should be presented in tables andfigures, and how results and discussion sections should be written.
Third, I emphasize probability theory and its importance to the process ofinference to a greater extent than many books do. Most books discuss probabilitytheory, but I feel its relationship with statistics is underdeveloped. In this book,probability is painstakingly shown to be the basis for statistical reasoning.
Fourth, I have included a number of homework problems at the end of almostevery chapter, along with answers to the first half of them in each chapterin an appendix. There are several differences between the homework problemsand answers I provide and those found in typical texts. First, I do not provideas many exercises as is commonly found in texts. In most chapters, there areapproximately 10 exercises. The exceptions are Chap. 5, with 50 items, and Chap. 6,
vii
![Page 8: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/8.jpg)
viii Preface
with 40 items. My rationale for this is that for developing an understanding ofmost concepts, I do not believe students need to perform dozens of exercises: oncethey have done one or two, especially ones that are word based and involve morethought in implementation, they understand the process. With problems involvingmore complicated (or at least tedious) calculations—like the chi-square test ofindependence, ANOVA, correlation, and simple regression—performing repetitivecalculations may even be detrimental, fostering a “losing the forest for the trees”problem.
Thus, a second difference between my exercises and those often found in texts isthat I have tried to construct exercises that are interesting, that often may be solvedin different ways, that are realistic in the sense that they often require thinking like afull-fledged researcher would, and that often involve information that cannot simplybe dropped into a formula in a rote fashion. For example, in the exercises for one-way ANOVA modeling, I have three exercises without solutions (and three withsolutions). In one, the raw data are provided, and the student must compute sums ofsquares and complete the ANOVA table. This problem tests the student’s ability toperform the basic computations. In another, a partial ANOVA table is provided,and all the student must do is understand the structure of the ANOVA table tocomplete it (i.e., SST D SSB C SSW and df .T / D df .B/ C df .W /) Finally,in another, the student is provided with group-level statistics like the mean andvariance and must recover (two of) the sums of squares by “undoing” the variancecalculations before constructing the ANOVA table. This problem tests the student’sdeeper understanding of the connection between variances and sums of squares.
A third difference between my exercises and those found in other texts is that Iuse the exercises in some cases to introduce additional material not covered directlyin the chapter and as further examples of concepts and methods discussed in thechapter. As an example of the former strategy, the last exercise in Chap. 6 introducesthe basic ideas underlying capture-recapture methodology. The detailed answers toexercises exemplify the latter strategy: the answers are detailed much as a solutionto an example problem would be within a chapter in a typical text. They are placedat the end of the chapter so that they do not disrupt the flow of the text and so thatstudents can work the problems and check their thought and computation processesas they work them.
A final difference between this text and others is that I routinely use simulated orcontrived data, rather than real data, to illustrate concepts. To be sure, I use real datafrom the General Social Survey heavily in both examples and exercises. However,while real data are useful for giving students concrete examples, I think simulateddata are often more useful when trying to illustrate difficult ideas, especially whenshowing how unknowns can be estimated. In real data, the unknowns are unknown!Also, it is generally the case that real data are messy. It is easier, for example, toillustrate the notion of row and column percentages in cross-tabs using roundernumbers than may be found in real data. Perhaps as a better example, it is easierto illustrate the idea of within versus between sums of squares in ANOVA modelingby creating contrived data in which the within sum of squares is 0 and contrasting itwith data in which the between sum of squares is 0.
![Page 9: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/9.jpg)
Preface ix
Layout for a Course
The book is intended as a text for a semester-long course on introductory statisticsfor social science students. Princeton semesters are relatively short (12 weeks),and our courses have one-day-per-week “precepts” in which the large class isbroken into smaller groups (of about 15 students each) that are led by preceptors(graduate teaching assistants or the course instructor). I use these precepts to teachstudents how to use statistical software and to help students work through homeworkproblems. Given the precepts, there are two 1-h lectures per week. After discounting3 days for exams, that leaves 21 lectures. Table 1 shows the organization of thelectures and the corresponding book chapters.
As the table shows, I give four exams in my course. The first covers thebasics of the research process and descriptive statistics. The midterm exam coversprobability theory and one sample inference. The third exam covers various two-sample statistical tests, like the independent samples t test, and the chi-square testof independence. The final exam covers ANOVA, the correlation, and regressionmodeling.
Lecture Topic Chapter
1 Overview of research 1,22 Data acquisition 33 continued4 Summarizing data 45 continued* First exam none6 Probability theory 57 continued8 Central Limit Theorem 69 One sample tests10 Confidence intervals11 Review none* Mid-term exam12 Two-sample tests/intervals13 continued14 Chi-square 715 ANOVA 816 Correlation & simple regression 917 continued* Third exam none18 Multiple regression 1019 continued20 continued21 Tables and figures 11
Table 1. Typical course structure based on book
![Page 10: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/10.jpg)
x Preface
Acknowledgements
I have a number of people to thank for helping with the creation and revision ofthis book over the last 5 or 6 years. Most importantly, I thank my wife, Barbara,for listening to me talk about this project, and my undergraduate course moregenerally, over the last decade, for offering ideas for the book, and for beinggenerally supportive. I thank my long-time colleague and friend Scott Brown foralso providing numerous ideas for the book. I thank my colleagues, Mitch Duneier,for his help in revising the discussion of qualitative data and methods in Chap. 3, andMatt Salganik for providing general advice throughout the book. I thank my editor,Marc Strauss, and his assistant, Hannah Bracken, for their support in the publicationprocess. Finally, I would like to thank the preceptors for my course over the lastseveral years (since 2008), including Edward Berchick, Mariana Campos Horta,Phillip Connor, Elaine Enriquez, Lauren Gaydosh, Kerstin Gentsch, JoAnne Golan,Aaron Gottlieb, Angelina Grigoryeva, Patrick Ishizuka, Adam Moore, ManishNag, Rourke O’Brien, David Pedulla, and Erik Vickstrom. Aside from the usualresponsibilities of lecturing, grading, and meeting with students, these preceptorsalso helped find errors in the book and make suggestions for its improvement. Anyerrors that remain, however, are mine.
![Page 11: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/11.jpg)
List of Symbols
! Logical if/then: A ! B means “If A is true, then B istrue”
) Logical/mathematical “therefore”: Negation (logical not) symbol. Ex: p.A/Cp.:A/ D 1
(the sum of the probabilities of event A and not eventA is 1). Ex: :.:A/ D A
� Equivalence: If A � B , then B is just another way ofexpressing A
Š Factorial. kŠ D k.k�1/.k�2/ : : : .k�.k�2//.1/.0Š/,with 0Š D 1 by definition
C.n; x/ � �nx
�Combinatorial expression for selecting x items out ofn items without replacement and in which order doesnot matter. Calculation: nŠ
xŠ.n�x/Š
n Commonly used to represent a sample size. N is oftenused to represent a population size
xi Generic representation of the i th value of the variablex. Example: If the variable x is body weight, x3 wouldbe the body weight of the third person in the sampleunder considerationPn
iD1 Summation of items labeled i D 1 : : : n. i is the indexof the summation loop; n is the total number of itemsbeing summed
Nx Sample meanMD MedianMO ModeIQR Interquartile ranges2 Sample variance; s is the standard deviation
xi
![Page 12: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/12.jpg)
xii List of Symbols
f .x/ May be generic for a function of x, like a probabilitydensity function, or may represent the frequency of x
� Population mean�2 Population variance; � is the population standard de-
viations Sample standard deviationEi Generic event i that may occur in a trialp.Ei / Probability of event i occurring in a trialS Sample space, sometimes2 Is an element of. Ex. S1 2 S2 means that sample space
S1 is a subset of sample space S2
p.A; B/ � p.A \ B/ Joint probability of events A and B
p.A [ B/ Probability of event A or B , where “or” is inclusive(and/or)
p.AjB/ Conditional probability of event A given B is trueP.n; x/ Permutation of x items selected without replacement
from n items in which order of selection/arrangementmatters. Computed as: nŠ
.n�x/Š
z Usually a standard normal variablep Often the probability parameter in a binomial distri-
bution. Can also be a “p” value, the probability ofobserving the sample data under the null hypothesis
ex The exponential function� “Is distributed as”� Nx The mean of the sampling distribution of means (i.e.,
the mean of the distribution of sample means underrepeated sampling from the population)
� Nx The standard deviation of the sampling distribution ofmeans. Under the Central Limit Theorem, � Nx D �=
pn
s:e: Standard error. Usually an estimate of � NxH0 Usually the null hypothesis˛ The value at which one is willing to reject the null
hypothesis. If p < ˛, we reject the nullt A distribution similar to the normal distribution but
with fatter tails and a degree of freedom parameterthat, for our purposes, captures the extent to which s
is a good estimate of �
![Page 13: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/13.jpg)
List of Symbols xiii
t˛=2 The value of t for which one half of the mass ˛ fallsbeyond in the t distribution
1 � ˛ The level of confidence in a confidence intervalmin.a; b/ Minimum function. Ex. min.1; 2/ D 1
�2 Chi-square. �2 is a probability distribution and is usedin the chi-square test of independence and the lack-of-fit test in this book
NNx The overall sample mean. Usually called the grandmean in ANOVA
R2 R-squared. Proportion of variance explained in anANOVA or regression model by grouping/predictorvariables
MSE Mean squared error from ANOVA and regression mod-eling
F F is a probability distribution used in ANOVA andregression modeling
cov.x; y/ The covariance of x and y
r Pearson’s correlation coefficientzf Fisher’s z transformed correlation coefficient� Approximately equal toOa An estimate of a. Read: “a-hat”˛ Intercept term for the simple regression model. The
estimate for it is Oˇ Slope term for the simple regression model. The sti-
mate for it is Oei Error term for the i th individual in the regression
modelbk Alternate representation for regression coefficients. b0
is the intercept, b1 is the slope in simple regression,and subsequent subscripts are for additional variablesin multiple regression models
�2e The MSE of the regression model. Usually has a
“hat” above it, because it is estimated and not directlyobserved
![Page 14: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/14.jpg)
![Page 15: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/15.jpg)
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Goals of This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2 Overview of the Research Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.1 What Is NOT Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.2 Replication as Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.3 Stages of Research and Scientific Paper Structure . . . . . . . . . . . . . . . . . 7
2.3.1 What Is a Perspective? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82.3.2 What Is a Theory? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.3.3 What Is a Proposition? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.3.4 What Is a Research Question? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.3.5 What Is a Hypothesis? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Some Summary and Clarification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.5 What Research Cannot Do: Proof . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.7 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.8 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3 Data and Its Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173.1 Qualitative Data Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.1 The “Unit of Analysis” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.2 Quantitative Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2.1 Sampling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.2.2 Response Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263.2.3 Instrument and Item Construction . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.4 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343.5 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4 Summarizing Data with Descriptive Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . 374.1 Summarizing Nominal Level Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
xv
![Page 16: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/16.jpg)
xvi Contents
4.2 Summarizing Interval and Ratio Level Measures . . . . . . . . . . . . . . . . . . 394.2.1 Measures of Central Tendency . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394.2.2 Measures of Dispersion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
4.3 Summarizing Bivariate Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484.4 What About Ordinal Data? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514.5 The Abuse and Misuse of Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554.7 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554.8 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5 Probability Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 575.1 Probability Rules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.1.1 Total Probability and Bayes’ Theorem . . . . . . . . . . . . . . . . . . . . 615.2 How to Count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 635.3 Probability Density/Mass Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
5.3.1 Binomial Mass Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685.3.2 Normal Density Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715.3.3 Normal Approximation to the Binomial . . . . . . . . . . . . . . . . . . 75
5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775.5 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 785.6 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
6 Statistical Inference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 836.1 The Central Limit Theorem and Inferential Statistics. . . . . . . . . . . . . . 836.2 Hypothesis Testing Using the z Distribution . . . . . . . . . . . . . . . . . . . . . . . 876.3 Hypothesis Testing When � Is Unknown: The t Distribution . . . . . 906.4 Confidence Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 906.5 Additional Hypothesis Testing Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 956.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 986.7 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 986.8 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
7 Statistical Approaches for Nominal Data: Chi-Square Tests . . . . . . . . . . 1077.1 The Chi-Square (�2) Test of Independence. . . . . . . . . . . . . . . . . . . . . . . . . 110
7.1.1 The Test Applied to the Region and MaritalStatus Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
7.2 The Lack-of-Fit �2 Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1137.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1147.4 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1157.5 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
8 Comparing Means Across Multiple Groups: Analysisof Variance (ANOVA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1178.1 The Logic of ANOVA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1178.2 Some Basic ANOVA Computations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1198.3 A Real ANOVA Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1228.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
![Page 17: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/17.jpg)
Contents xvii
8.5 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1248.6 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
9 Correlation and Simple Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1279.1 Measuring Linear Association . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
9.1.1 The Covariance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1279.1.2 The Pearson Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1299.1.3 Confidence Intervals for the Correlation . . . . . . . . . . . . . . . . . . 1319.1.4 Hypothesis Testing on the Correlation . . . . . . . . . . . . . . . . . . . . 1339.1.5 Limitations of the Correlation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
9.2 Simple Linear Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1349.2.1 Logic of Simple Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1349.2.2 Estimation of the Model Parameters . . . . . . . . . . . . . . . . . . . . . . 1369.2.3 Model Evaluation and Hypothesis Tests . . . . . . . . . . . . . . . . . . 137
9.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1399.4 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1409.5 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
10 Introduction to Multiple Regression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14310.1 Understanding Causality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
10.1.1 The Counterfactual Model and ExperimentalMethods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
10.1.2 Statistical Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15110.2 Multiple Regression Model, Estimation,
and Hypothesis Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15310.2.1 Total, Direct, and Indirect Association . . . . . . . . . . . . . . . . . . . . 15610.2.2 Suppressor Relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
10.3 Expanding the Model’s Capabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16210.3.1 Including Non-continuous Variables . . . . . . . . . . . . . . . . . . . . . . 16210.3.2 Statistical Interactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16410.3.3 Modeling Nonlinear Relationships . . . . . . . . . . . . . . . . . . . . . . . . 166
10.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16910.5 Items for Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16910.6 Homework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
11 Presenting Results of Statistical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17511.1 Making Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17511.2 Making Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17811.3 Writing the Results Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18111.4 Writing the Discussion Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18311.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
12 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
A Statistical Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189A.1 The Z (Standard Normal) Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189A.2 The t Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
![Page 18: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/18.jpg)
xviii Contents
A.3 The Chi-Square (�2) Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193A.4 The F Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194A.5 Using the Tables and Interpolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
B Answers to Selected Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199B.1 Chapter 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199B.2 Chapter 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199B.3 Chapter 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200B.4 Chapter 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201B.5 Chapter 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203B.6 Chapter 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207B.7 Chapter 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213B.8 Chapter 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215B.9 Chapter 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216B.10 Chapter 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220B.11 Chapter 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
![Page 19: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/19.jpg)
List of Figures
Fig. 3.1 Illustration of concepts of reliability and validity of measurement 27
Fig. 4.1 Example of a barchart for the race data shown in Table 4.1 . . . . . . . 38Fig. 4.2 Stem-and-Leaf plot of education data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Fig. 4.3 Horizontal version of stem-and-leaf plot for education . . . . . . . . . . . . 42Fig. 4.4 Example of a histogram for the education data. . . . . . . . . . . . . . . . . . . . . 43Fig. 4.5 Another approach to plotting a histogram (mean is
solid line; median is dotted line; mode is dashed line) . . . . . . . . . . . . . 43Fig. 4.6 Example of a boxplot for the education data (Q1, Q2,
and Q3 are highlighted with dashed lines). . . . . . . . . . . . . . . . . . . . . . . . . . 46
Fig. 5.1 Sample Venn diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60Fig. 5.2 Heuristic for the law of total probability. . . . . . . . . . . . . . . . . . . . . . . . . . . . 61Fig. 5.3 Tree diagram showing outcomes of three coin flips . . . . . . . . . . . . . . . . 69Fig. 5.4 Some binomial distributions (n D 10) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70Fig. 5.5 Pascal’s triangle. Each row shows the number of ways
to obtain x successes out of n trials for different valuesof x, where x increases from left to right from 0 to n . . . . . . . . . . . . . . 70
Fig. 5.6 Some normal distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73Fig. 5.7 Some common areas under the normal distribution at
given numbers of standard deviations from the mean . . . . . . . . . . . . . . 74Fig. 5.8 Normal distribution area shown in z table vs. the area
of interest in IQ example. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75Fig. 5.9 Illustration of continuity correction factor in binomial
distribution with n D 10 and p D :5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Fig. 6.1 Depiction of repeated sampling from the populationand the concept of a sampling distribution. . . . . . . . . . . . . . . . . . . . . . . . . . 84
Fig. 6.2 Histograms of sampling distributions of the mean forvarious sample sizes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Fig. 6.3 Some t distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
xix
![Page 20: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/20.jpg)
xx List of Figures
Fig. 6.4 Illustration of concept of “confidence.” Plot shows that95 % of the sample means that can be drawn from theirsampling distribution will contain � within an intervalranging 2 standard errors around Nx. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Fig. 6.5 Some confidence intervals from samples of differentsizes (based on the z and t distributions) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
Fig. 9.1 Scatterplot of hypothetical variables x and y. . . . . . . . . . . . . . . . . . . . . . . 128Fig. 9.2 Scatterplot of hypothetical variables x and y with
deviations from means superimposed for a single observation. . . . . 129Fig. 9.3 Scatterplot of education and income (2004 GSS Data). . . . . . . . . . . . . 130Fig. 9.4 Scatterplot of standardized education and income with
reference lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Fig. 10.1 Path model for the relationship between parents’education, respondent’s education, and respondent’s income. . . . . . 152
Fig. 10.2 Correlations between respondent’s education andincome for the sample (solid line), by level of parent’seducation (dots). Dashed line is the average ofcorrelations at different levels of parent’s education(levels of schooling discussed in text). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
Fig. 10.3 Path model for the relationship between parents’education, respondent’s education, and respondent’sincome with standardized coefficient estimates (GSSdata from regression example). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
Fig. 10.4 Path model for the relationship between age,education, and income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Fig. 10.5 Expected income by years of schooling and race (GSSdata 1972–2010, ages 30–64) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
Fig. 10.6 Mean income by years of schooling with valuesweighted by sample size (GSS data 1972–2010, ages 30–64) . . . . . 167
Fig. 10.7 Mean income by education with linear, quadratic, andcubic fits superimposed (GSS data 1972–2010) . . . . . . . . . . . . . . . . . . . . 168
Fig. 11.1 Income and schooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179Fig. 11.2 Observed and predicted mean income by years of
education (n = 12,924, 1972–2006 GSS data forpersons ages 30–54; income in 2006 dollars). . . . . . . . . . . . . . . . . . . . . . . 180
Fig. 11.3 Observed and predicted mean income by years ofeducation (n = 12,924, 1972–2006 GSS data forpersons ages 30–54; income in 2006 dollars). . . . . . . . . . . . . . . . . . . . . . . 180
Fig. A.1 The standard normal distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190Fig. A.2 The t distribution with 1 degree of freedom . . . . . . . . . . . . . . . . . . . . . . . . 192Fig. A.3 The �2 distribution with 5 degrees of freedom . . . . . . . . . . . . . . . . . . . . . 193Fig. A.4 The F distribution with 3 and 10 degrees of freedom . . . . . . . . . . . . . . 195
![Page 21: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/21.jpg)
List of Tables
Table 1 Typical course structure based on book . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Table 2.1 Correspondence of the parts of the research processwith the parts of a scientific paper. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Table 2.2 An example of the fallacy of affirming the consequent. . . . . . . . . . . 13Table 2.3 An example of the logically valid argument structure
of modus tollens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Table 3.1 Cross-tabulation of hypothetical political party votesby race. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Table 3.2 Alternative cross-tabulation of hypothetical politicalparty votes by race. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Table 3.3 Rankings of different modes of data collection alongseveral dimensions (1 D best; 3 D worst). . . . . . . . . . . . . . . . . . . . . . . . . . 21
Table 3.4 Color composition of samples of four marbles takenfrom a population of 10 marbles, half black and half white. . . . . . 23
Table 3.5 Types of sample frames and some problems with them. . . . . . . . . . 25
Table 4.1 Values of race for a (contrived) sample of n D 100
persons (W D white; B D black; O D other) . . . . . . . . . . . . . . . . . . . . . . . 38Table 4.2 Sample of n D 100 values of years of schooling for
persons ages 80C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Table 4.3 Sample of n D 100 values of years of schooling for
persons ages 80+ sorted to facilitate finding median. . . . . . . . . . . . . . 40Table 4.4 Variance calculations for education data. Dividing
the sum at the bottom right by 99 (n � 1) yields 13.06. . . . . . . . . . . 47Table 4.5 Cross-tabulation of sex and race for the 2010 GSS. . . . . . . . . . . . . . . 49
Table 5.1 Elements of Bayes’ Theorem in disease example. . . . . . . . . . . . . . . . . 63Table 5.2 Some helpful shortcuts for calculating combinations. . . . . . . . . . . . . 71
Table 6.1 Results of simulation examining distributionsof sample means . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
xxi
![Page 22: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/22.jpg)
xxii List of Tables
Table 6.2 Proportion of “95 %” confidence interval estimatescapturing the true mean by type of interval (z/t ) andsample size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Table 6.3 Various hypothesis testing and confidence intervalformulas following from the CLT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Table 7.1 Crosstabulation of hypothetical religious affiliationand educational attainment data (all religions notshown have been excluded). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Table 7.2 Crosstabulation of marital status and region of thecountry (2006 GSS data). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Table 7.3 Crosstabulation of marital status and region of thecountry (2006 GSS data) with row percentages included. . . . . . . . . 109
Table 7.4 Generic cross-tab of variables X and Y . . . . . . . . . . . . . . . . . . . . . . . . . . . 111Table 7.5 Crosstabulation of marital status and region of the
country (2006 GSS data) with chi square test ofindependence calculations (�2 D 14:65, df D 9; p > :10) . . . . . 113
Table 7.6 Observed and expected counts of racial groupmembers (2000 GSS data) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Table 8.1 Hypothetical distribution of study hours by year in college . . . . . . 118Table 8.2 Another hypothetical distribution of study hours by
year in college . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118Table 8.3 Sums of squares for the first hypothetical set of data.
Note: NNx D 6; SST D PGD3gD1
Png
iD1.xig � NNx/2 D 48 . . . . . . . . . . . . . . 120Table 8.4 Sums of squares for the second hypothetical set of
data. Note: NNx D 6; SST D PGD3gD1
Png
iD1.xig � NNx/2 D 48 . . . . . . . . 121Table 8.5 Format for a generic one-way ANOVA table. . . . . . . . . . . . . . . . . . . . . . 122Table 8.6 ANOVA example: Race differences in educational
attainment (2006 GSS data). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Table 9.1 Generic ANOVA table for a simple regression model. . . . . . . . . . . . 137Table 9.2 ANOVA table for model of income regressed on
education (data from continued example) . . . . . . . . . . . . . . . . . . . . . . . . . 138
Table 10.1 Illustration of a true experimental design applied totwo individuals: i and j . R.T / and R.C / representrandom assignment to treatment and control groups,respectively. � represents change or betweenindividual differences. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Table 10.2 Regression of income on respondent and parentaleducation (GSS data 1972–2010, ages 30–64,n D 20; 409; coefficients and (s.e.) presented) . . . . . . . . . . . . . . . . . . . . 155
Table 10.3 Regression of income on respondent age andeducation (GSS data 1972–2010, ages 30–64,n D 20; 409; coefficients and (s.e.) presented) . . . . . . . . . . . . . . . . . . . . 160
![Page 23: Using Statistics in Social Research978-1-4614-8573... · 2017-08-29 · certainly still used, they have limited application in contemporary social research. Thus, they can be investigated](https://reader033.fdocuments.net/reader033/viewer/2022050513/5f9d6a45ea3d460d9e39caa6/html5/thumbnails/23.jpg)
List of Tables xxiii
Table 10.4 Regression of income on race and education(GSS data 1972–2010, ages 30–64, n D 20;409;coefficients and (s.e.) presented) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Table 10.5 Regression of income on race and education(GSS data 1972–2010, ages 30–64, n D 19;440;coefficients and (s.e.) presented) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Table 11.1 Descriptive statistics for variables used in analysesby sex (2010 GSS). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
Table 11.2 Results of multiple regression analyses of income onrace and education by sex (2010 GSS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Table A.1 Table shows LEFT TAIL probabilities—area from�1 up to Z . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
Table A.2 Table shows ONE TAIL t statistics (t for right tail ˛) . . . . . . . . . . . . 192Table A.3 Table shows RIGHT TAIL �2 statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . 194Table A.4 Table shows critical values of F for one-tail probabilities . . . . . . . 196