Addison-Wesley / Prentice Hall

Statistics

Browse available resources for Statistics:



Statistical Methods for the Social Sciences, 4/E
Alan Agresti, University of Florida
Barbara Finlay

ISBN-10: 0130272957
ISBN-13: 9780130272959

Publisher: Prentice Hall
Copyright: 2009
Format: Cloth; 624 pp
Published: 12/28/2007

Suggested retail price: $128.00
Buy from myPearsonStore

The book presents an introduction to statistical methods for students majoring in social science disciplines.  No previous knowledge of statistics is assumed, and mathematical background is assumed to be minimal (lowest-level high-school algebra).

 

The book contains sufficient material for a two-semester sequence of courses.  Such sequences are commonly required of social science graduate students in sociology, political science, and psychology. Students in geography, anthropology, journalism, and speech also are sometimes required to take at least one statistics course.

The author is successful in his goal of introducing statistical methods in a style that emphasized their concepts and their application to the social sciences rather than the mathematics and computational details behind them.

 

 

1.  Strong emphasis on regression topics.  Moreover, a wide variety of regression models (such as linear regression, ANOVA, logistic

regression) are taught in the same format, essentially as special cases of a generalized linear model.

 

2.  Emphasis on concepts, rather than computing formulas.  Advanced topics such as regression and ANOVA emphasize interpreting output from computer packages rather than complex computing formulas.

 

3.  Integration of descriptive and inferential statistics from an early point in the text.

 

4.  A technically correct presentation.

The fourth edition has an even stronger emphasis on concepts and applications, with greater attention to "real data" both in the examples and exercises. The mathematics is still downplayed, in particular probability, which is all too often a stumbling block for students.  On the other hand, the text is not a cookbook. Reliance on an overly simplistic recipe-based approach to statistics is not the route to good statistical practice.

 

Changes in the Fourth Edition:

 

Since the first edition, the increase in computer power coupled with the continued improvement and accessibility of statistical software has had a major impact on the way social scientists analyze data. Because of this, this book does not cover the traditional shortcut hand-computational formulas and approximations.  The presentation of computationally complex methods, such as regression, emphasizes interpretation of software output rather than the formulas for performing the analysis. Teh text contains numerous sample printouts, mainly in the style of SPSS and occasionaly SAS, both in chapter text and homework problems. This edition also has an appendix explaining how to apply SPSS and SAS to conduct the methods of each chapter and a website giving links to information about other software.

 

http://www.stat.ufl.edu/~aa/social/data.html

 

 

This edition contains several changes and additions in content, directed toward a more modern approach.  The main changes are as follows:

 

  • There is a stronger focus on real examples and on the integration of statisical software. This includes some new exercises that ask students to use applets (located at http://www.prenhall.com/statistics , search for Agresti/ Finlay under "Statistical Methods for the Social Sciences" ) to help learn the fundamental concepts of sampling distributions, confidence intervals, and significance tests.
  • This edition has a somewhat lower technical level in the first nine chapters, to make the book more easily accessible to undergraduate students.  To help with this, some notation has been simplified or eliminated.

The author, in this new edition, uses the symbol se for estimated standard errors, rather than the notation of sigma-hat with subscript having the estimator symbol. Although not quite as informative, this will again make results consistent with software output, and help students connect the idea of the se for the various inferential methods they see.

 

The author uses capital Y only as notation for a variable and lower-case for observed values and sample statistics; thus, y-bar, rather than Y-bar, which is consistent with the lower-case used throughout for the standard deviation and other statistics.

  • Chapter 3 on descriptive statistics has a separate section for measures of positions, such as percentiles and related topics such as the box plot and outliers. It also has a short section on bivariate descriptive methods. This gives students an early exposure to contingency tables and to regression and the point that, in practice, there's almost always more than one variable of interest. This section also introduces them to the concepts of association, and response and explanatory variables.
  • Chapter 4 has a new section that introduces the relative frequency concept and briefly summarizes three basic probability rules that are occasionally applied in the text.
  • The inference material on means in chapters 5-7 has been changed to rely completely on the t distribution, rather than using the z  for large samples and t  for small samples. Thsi makes results consistent with software output. The author continues to emphasize that the normality assumption for the t distribution is mainly needed for small samples with one-sided inference.
  • partly because of the change in using the t distribution always for inference about the mean, chapter 5 on confidence intervals now presents methods for the proportion (which relies only on the normal distribution) before the mean. This way, the students can learn the basic concept of a confidence interval using the information they've just learned at the end of chapter 4 about the normal distribution as a sampling distribution (i.e. for a proportion, the margin of error multiplies the standard error by a z-score rather than a t-score). This delays introduction of the t distribution by a section, so students are not confronted with too many new topics all at once.
  • Chapter 7 on comparing two groups has a new section introducing ideas of bivariate analysis, reminding students of the distinction between response and explanatory variables, defining independent and dependent samples, discussing how to compare two groups with a difference or a ratio of two parameters, and showing the general formula for finding a standard error of a difference between two independent estimates. Section 7.3 introduces the concept of a model.
  • Chapter 12 on ANOVA explains the ideas behind the F test and gives an example before presenting the sums of squares formulas.
  • Chapter 15 provides a less technical explanation of logistic regression and introduces its extensions for  nominal and ordinal response variables.
  • Chapter 16 includes new sections on longitudinal data analysis and multilevel (hierarchical) models.

 

 

1.Introduction

    1.1 Introduction to statistical methodology

    1.2 Descriptive statistics and inferential statistics

    1.3 The role of computers in statistics

    1.4 Chapter summary

2. Sampling and Measurement

    2.1 Variables and their measurement

    2.2 Randomization

    2.3 Sampling variability and potential bias

    2.4 other probability sampling methods *

    2.4 Chapter summary

3. Descriptive statistics

    3.1 Describing data with tables and graphs

    3.2 Describing the center of the data

    3.3 Describing variability of the data

    3.4 Measure of position

    3.5 Bivariate descriptive statistics

    3.6 Sample statistics and population parameters

    3.7 Chapter summary

4. Probability Distributions

    4.1 Introduction to probability

    4.2 Probablitity distributions for discrete and  continuous variables

    4.3 The normal probability distribution

    4.4 Sampling distributions describe how statistics vary

    4.5 Sampling distributions of sample means

    4.6 Review: Probability, sample data, and sampling distributions

    4.7 Chapter summary

5. Statistical inference: estimation

    5.1 Point and interval estimation

    5.2 Confidence interval for a proportion

    5.3 Confidence interval for a mean

    5.4 Choice of sample size

    5.5 Confidence intervals for median and other parameters*

    5.6 Chapter summary

 6. Statistical Inference: Significance Tests

    6.1 Steps of a significance test

    6.2 Significance test for a eman

    6.3 Significance test for a proportion

    6.4 Decisions and types of errors in tests

    6.5 Limitations of significance tests

    6.6 Calculating P (Type II error)*

    6.7 Small-sample test for a proportion: the binomial distribution*

    6.8 Chapter summary

7. Comparison of Two Groups

    7.1 Preliminaries for comparing groups

    7.2 Categorical data: comparing two proportions

    7.3 Quantitative data: comparing two means

    7.4 Comparing means with dependent samples

    7.5 Other methods for comparing means*

    7.6 Other methods for comparing proportions*

    7.7 Nonparametric statistics for comparing groups

    7.8 Chapter summary

8. Analyzing Association between Categorical Variables

    8.1 Contingency Tables

    8.2 Chi-squared test of independence

    8.3 Residuals: Detecting the pattern of association

    8.4 Measuring association in contingency tables

    8.5 Association between ordinal variables*

    8.6 Inference for ordinal associations*

    8.7 Chapter summary

9. Linear Regression and Correlation

    9.1 Linear relationships

    9.2 Least squares prediction equation

    9.3 The linear regression model

    9.4 Measuring linear association - the correlation

    9.5 Inference for the slope and correlation

    9.6 Model assumptions and violations

    9.7 Chapter summary

10. Introduction to multivariate Relationships

    10.1 Association and causality

    10.2 Controlling for other variables

    10.3 Types of multivariate relationships

    10.4 Inferenential issus in statistical control

    10.5 Chapter summary

11. Multiple Regression and Correlation

    11.1 Multiple regression model

    11.2 Example with multiple regression computer output

    11.3 Multiple correlation and R-squared

    11.4 Inference for multiple regression and coefficients

    11.5 Interaction between predictors in their effects

    11.6 Comparing regression models

    11.7 Partial correlation*

    11.8 Standardized regression coefficients*

    11.9 Chapter summary

12. Comparing groups: Analysis of Variance (ANOVA) methods

    12.1 Comparing several means: One way analysis of variance

    12.2 Multiple comparisons of means

    12.3 Performing ANOVA by regression modeling

    12.4 Two-way analysis of variance

    12.5 Two way ANOVA and regression

    12.6 Repeated measures analysis of variance*

    12.7 Two-way ANOVA with repeated measures on one factor*

    12.8 Effects of violations of ANOVA assumptions

    12.9 Chapter summary

13. Combining regression and ANOVA: Quantitative and Categorical Predictors

    13.1 Comparing means and comparing regression lines

    13.2 Regression with quantitative and categorical predictors

    13.3 Permitting interaction between quantitative and categorical predictors

    13.4 Inference for regression with quantitative and categorical predictors

    13.5 Adjusted means*

    13.6 Chapter summary

14. Model Building with Multiple Regression

    14.1 Model selection procedures

    14.2 Regression diagnostics

    14.3 Effects of multicollinearity

    14.4 Generalized linear models

    14.5 Nonlinearity: polynomial regression

    14.6 Exponential regression and log transforms*

    14.7 Chapter summary

15. Logistic Regression: Modeling Categorical Responses

    15.1 Logistic regression

    15.2 Multiple logistic regression

    15.3 Inference for logistic regression models

    15.4 Logistic regression models for ordinal variables*

    15.5 Logistic models for nominal responses*

    15.6 Loglinear models for categorical variables*

    15.7 Model goodness of fit tests for contingency tables*

    15.9 Chapter summary

16. Introduction to Advanced Topics

    16.1 Longitudinal data analysis*

    16.2 Multilevel (hierarchical) models*

    16.3 Event history analysis*

    16.4 Path analysis*

    16.5 Factor analysis*

    16.6 Structural equation models*

    16.7 Markov chains*

Appendix: SAS and SPSS for Statistical Analyses

Tables

Answers to selected odd-numbered problems

Index

 

 

    For an introductory course in statistics for undergraduate or graduate students in departments of sociology, political science, and psychology, geography, anthropology, journalism, and speech.

    Assuming no prior knowledge of statistics, these authors combine their varied backgrounds—one a statistician, the other a social scientist—to introduce statistical methods with a high degree of statistical accuracy and a wealth of examples that are interesting and relevant to social scientists. The Third Edition has been both updated and improved to integrate real-world data into examples and exercises and make coverage more accessible throughout the text.

  • 0135265266Statistical Methods for the Social Sciences, 3/E
    Agresti & Finlay
    © 1997 | Prentice Hall | Cloth; 643 pages | Instock
    ISBN-10: 0135265266 | ISBN-13: 9780135265260
    Brief Description

"This text is readable, understandable, and well-organized. It provides good examples with SPSS output." (Robert Wilson, University of Delaware).

 

"Overall, [Agresti/ Finlay] is a good book for introductory statistics that targets general audiences...it covers most topics you want to cover and allows the instructor to choose which topics to include." (Youqin Huang, State University of New York, Albany)

 

"I originally started using the Agresti/ Finlay book based on its reputation as "the class of the market", in terms of being unfailingly statistically correct and having a "modern" perspective. By "modern", I mean that it is model rather than test oriented, that it gives heavy emphasis to confidence intervals and p-values rather than using arbitrary levels of significance, and that it eschews computational formulae. It has met those expectations..." (Michael Lacey, Colorado State University)

 

"..the book has been a good and helpful resource for me in preparing the class notes and assigning homework qustions. The main concepts to be understood by students are sampling distribution, confidence interval, p-value, linear regression. The book helps in this..." (Arne Bathke, University of Kentucky)

Alan Agresti is Distinguished Professor in the Department of Statistics at the University of Florida. He has been teaching statistics there for 30 years, including the development of three courses in statistical methods for social science students and three courses in categorical data analysis. He is author of over 100 refereed article and four texts including "Statistics: The Art and Science of Learning From Data" (with Christine Franklin, Prentice Hall, 2nd edition 2009) and "Categorical Data Analysis" (Wiley, 2nd edition 2002). He is a Fellow of the American Statistical Association and recipient of an Honorary Doctor of Science from De Montfort University in the UK. In 2003 he was named "Statistician of the Year" by the Chicago chapter of the American Statistical Association and in 2004 he was the first honoree of the Herman Callaert Leadership Award in Biostatistical Education and Dissemination awarded by the University of Limburgs, Belgium. He has held visiting positions at Harvard University, Boston University, London School of Economics, and Imperial College and has taught courses or short courses for universities and companies in about 20 countries worldwide. He has also received teaching awards from UF and an excellence in writing award from John Wiley & Sons.

View a Sample Chapter PDF:

  • Interwrite Personal Response System
    EduCue, Addison-Wesley & Benjamin Cummings
    © 2004 | Benjamin Cummings | Electronic Supplement | Instock
    ISBN-10: 0321267354 | ISBN-13: 9780321267351


  • iClicker Classroom Response System
    iClicker, Addison-Wesley & Benjamin Cummings
    © 2008 | Benjamin Cummings | Electronic Supplement | Instock
    ISBN-10: 032153705X | ISBN-13: 9780321537058


Give your students a choice! PearsonChoices products are designed to give your students more value and flexibility by letting them choose from a variety of text and media formats to best match their learning style and their budget.

Pearson Higher Education offers special pricing when you choose to package your text with other student resources. If you're interested in creating a cost-saving package for your students, see the Packages tab.

    CourseSmart Textbooks Online is an exciting new choice for students looking to save money. As an alternative to purchasing the print textbook, students can subscribe to the same content online and save up to 50% off the suggested list price of the print text. With a CourseSmart etextbook, students can search the text, make notes online, print out reading assignments that incorporate lecture notes, and bookmark important passages for later review. For more information, or to subscribe to the CourseSmart eTextbook, visit www.coursesmart.com.
  • 0135024153Statistical Methods for the Social Sciences, CourseSmart eTextbook, 4/E
    Agresti
    © 2009 | Prentice Hall | Electronic Book; 624 pages | Instock
    ISBN-10: 0135024153 | ISBN-13: 9780135024157
    URL: http://www.coursesmart.com
    Brief Description | Buy from myPearsonStore

Pearson Higher Education offers special pricing when you choose to package your text with other student resources. If you're interested in creating a cost-saving package for your students contact your Pearson Higher Education representative.


Copyright ©2008 Pearson Education. All rights reserved. Legal Notice | Privacy Policy | Permissions