For courses in Statistical Methods for the Social Sciences. Statistical methods applied to social sciences, made accessible to all through an emphasis on concepts Statistical Methods for the Social Sciences introduces
statistical methods to students majoring in social science disciplines.
With an emphasis on concepts and applications, this book assumes no
previous knowledge of statistics and only a minimal mathematical
background. It contains sufficient material for a twosemester course.
The 5th Edition uses examples and exercises with a variety of
¡°real data.¡± It includes more illustrations of statistical software for
computations and takes advantage of the outstanding applets to explain
key concepts, such as sampling distributions and conducting basic data
analyses. It continues to downplay mathematics—often a stumbling block
for students—while avoiding reliance on an overly simplistic
recipebased approach to statistics. Features
About the Book  Introduce social science students to statistical methods
o Strong
focus on real examples help students learn the fundamental concepts of
sampling distributions, confidence intervals, and significance tests. o Low technical level in first nine chapters makes the text accessible to undergraduate students. o Integration of descriptive and inferential statistics features from an early point in the text. o Strong emphasis on regression topics outlines
special cases of a generalized linear model, supported by a wide
variety of regression models (such as linear regression, ANOVA,
logistic, and regression). o Emphasis on concepts on advanced topics
(such as regression and ANOVA) underline the importance of interpreting
output from computer packages rather than complex computing formulas. o NEW! Greater integration of statistical software. Software
output shown now uses R and Stata instead of only SAS and SPSS,
although much output has a generic appearance. The text appendix
provides instructions about basic use of these software packages. o NEW! Companion website (found at www.pearsonglobaleditions.com/Agresti)
now features the data sets analyzed in the text in generic form to copy
for input into statistical software. Special directories there also
have data files in Stata format and in SPSS format so they are ready for
immediate use with those packages. Answers to Select OddNumbered
Exercises are available at the companion website. o NEW! New examples and exercises
ask students to use applets to help learn the fundamental concepts of
sampling distributions, confidence intervals, and significance tests.
The text also now relies more on applets for finding tail probabilities
from distributions such as the normal, t, and chisquared. The excellent
applets cited, can be found at www.pearsonglobaleditions.com/Agresti.  Emphasis on concepts and applications
 Descriptive statistic chapter
gives students early exposure to contingency tables, regression,
concepts of association, and response and explanatory variables.
 Relative frequency concept is introduced in chapter four and briefly summarizes three basic probability rules occasionally applied in the text.
 Full focus on t distribution
makes results consistent with software output and emphasizes that the
normality assumption for the t distribution is mainly needed for small
samples with onesided inference.
 Confidence intervals
present methods for proportion before the mean. This allows students to
learn the basic concept of a confidence interval without being
confronted with too many topics all at once.
 Comparing Two Groups chapter introduces
ideas of bivariate analysis, highlights the distinction between
response and explanatory variables, defines independent and dependent
samples, discusses how to compare two groups with a difference or a
ratio of two parameters, and shows the general formula for finding a
standard error of a difference between two independent estimates.
 ANOVA focus explains the ideas behind the F test and gives examples before presenting the sums of square formulas.
 UPDATED! Regression modeling chapter
now has a new section using case studies to illustrate how research
studies commonly use regression with both types of explanatory
variables. The chapter also has a new section introducing linear mixed
models.
 Logistic regression is explained in a less technical way so it¡¯s widely understood by all students.
New to this Edition
About the Book ¡¤ Greater integration of statistical software.
Software output shown now uses R and Stata instead of only SAS and
SPSS, although much output has a generic appearance. The text appendix
provides instructions about basic use of these software packages. ¡¤ New examples and exercises
ask students to use applets to help learn the fundamental concepts of
sampling distributions, confidence intervals, and significance tests.
The text also now relies more on applets for finding tail probabilities
from distributions such as the normal, t, and chisquared. The excellent
applets cited, can be found at www.pearsonglobaleditions.com/Agresti. ¡¤ ANOVA coverage
has been reorganized to put more emphasis on using regression models
with dummy variables to handle categorical explanatory variables. ¡¤ Companion website (found at www.pearsonglobaleditions.com/Agresti)
now features the data sets analyzed in the text in generic form to copy
for input into statistical software. Special directories there also
have data files in Stata format and in SPSS format so they are ready for
immediate use with those packages. Answers to Select OddNumbered
Exercises are available at the companion website. Content Updates ¡¤ Chapter 5 has a new section that introduces maximum likelihood estimation and the bootstrap method. ¡¤ Chapter 13
on regression modeling now has a new section using case studies to
illustrate how research studies commonly use regression with both types
of explanatory variables. The chapter also has a new section introducing
linear mixed models. ¡¤ Chapter 14 contains a new section on robust regression covering standard errors and nonparametric regression.
