Sociology 521: Social Statistics and Quantitative Methods I Spring 2012 Wed. 2 5, Kap 305 Computer Lab Instructor: Tim Biblarz Office hours (Kap 352): W, 5 6pm, F, 10 11, and by appointment (213) 740 3547; biblarz@usc.edu Teaching Assistant: Michela Musto Office hours (Kap 305 or 355): M, 1 2, W, 12 1, and by appointment (213) 740 4257; mmusto@usc.edu Course Website Course information will be posted to our course website at http://blackboard.usc.edu. Introduction Sociology 521 is a course about how sociologists apply statistics to different kinds of data to answer questions about ourselves and our social world. Statistics are mathematical tools that allow us to find regularities, substantive meaning, and sociological stories in data that on first glance may look as though they have neither rhyme nor reason. Quantitative research methodology typically involves a set of steps: 1) developing a conceptual model of the questions that we are interested in; 2) developing an instrument, like a questionnaire, to measure the dimensions described in our conceptual model; 3) selecting a study sample; 4) administering the survey; and then 5) inputting the results in coded form to create a dataset, or data matrix. We then statistically analyze the dataset to find answers to our original questions. The types of statistics covered in Soc. 521 will allow us to describe the distribution of variables; estimate general characteristics of populations based on the characteristics of samples; and test hypotheses about the nature of relationships among two or more variables in the population, based on the relationships among variables observed in samples. These topics comprise the areas of univariate and bivariate (plus limited multivariate) descriptive statistics, and inferential statistics. For those interested, Soc. 521 is followed next semester by Soc. 621, where we will focus on multivariate models that allow us to mathematically express and statistically explore the interrelationships among a large number of variables. Throughout this semester, we will make distinctions between: 1) statistics for the analysis of quantitative variables; 2) statistics for the analysis of categorical, or qualitative variables; and 3)
statistics for the analysis of quantitative and qualitative variables in combination. Soc. 521 gives a bit more time to statistics for the analysis of categorical variables, with a special section on loglinear models for contingency tables. In this class, each of you will have access to both small and large scale survey data sets. Homework assignments and exams will require, among other things, that you apply the statistics presented in class to these survey data using SPSS statistical software. Recommended Course Materials 1) Agresti, Alan and Barbara Finlay. 2009. Statistical Methods for the Social Sciences, Fourth Edition. Upper Saddle River, NJ: Prentice Hall, Inc. This introductory statistics text is written in a clear, no nonsense style by a statistician (Agresti) and a sociologist (Finlay) who specialize in the presentation of statistical ideas to nonstatisticians. The book s main emphasis is the application of statistics to the social sciences. 2) Fowler, Floyd J. 2008. Survey Research Methods, Fourth Edition. Newbury Park, CA: Sage Publications. This is a brief summary of current standards and procedures for conducting surveys sampling, question design, and interviewing. It pays particular attention to how various design decisions affect the quality of survey data. 3) Green, Samuel. B and Neil J. Salkind. 2007. Using SPSS for Windows and Macintosh: Analyzing and Understanding Data, Fifth Edition. Upper Saddle River, NJ: Prentice Hall. Although we will cover many of the ins and outs of SPSS statistical software, this book may be of use for assignments that require the application of SPSS to analyze survey data on computers. Requirements 1) Weekly Homework Assignments There will be 7 homework assignments over the course of the semester. The homeworks include exercises that I have constructed, asking you to apply statistics to concrete research problems using survey data sets. Homeworks will help you keep on top of the materials, actively engage you in the research process, facilitate your ability to calculate and interpret statistics, and provide feedback to the instructor. 2) Midterm and Final Examinations There will be two in class exams. Their purpose is to develop skill at applying statistics to data to answer substantively important questions, and to assess your understanding of course materials and ability to interpret statistical information. No make up or late exams will be accepted unless you have made arrangements with the instructor prior to the examination date. Exceptions will
only be made if you provide clear evidence that circumstances beyond your control prevented your taking the exam. 3) Student Presentation Projects (more about these early in the semester) Grading Assignment Percentage of Total Grade Homework Assignments 45% Midterm Exam 20% Final Exam 20% Presentation project 15% Total 100% Schedule WEEK 1 I. What are statistics, and what are they doing in sociology? II. Introduction to surveys: Conceptualization and measurement III. Statistics for describing the distribution of categorical variables, grouping, graphing **Agresti and Finlay Chs. 1, 2** **Agresti and Finlay Ch. 3, pp. 31 38** WEEKS 2 3 I. Statistics for describing relationships between categorical variables, and measuring association in twoway contingency tables (crosstabulations) II. Statistical inference: Significance tests and confidence intervals for a percentage/proportion, and for difference between two percentages/proportions; chi square test of independence in contingency tables **Agresti and Finlay Ch. 7, pp. 183 191, 201 220** **Agresti and Finlay Ch. 8** WEEK 4
Controlling for variables through multi way contingency tables: spurious relationships, conditional relationships (also called moderating relationships or moderators, and statistical interactions), and Simpson s Paradox WEEKS 5 7 Loglinear Models for contingency tables (week 7 workshop on presentation projects) **Agresti and Finlay Ch. 8** **selected articles posted on website** WEEK 8 MIDTERM EXAM WEEK 9 I. Statistics for describing central tendency and dispersion of quantitative variables II. Statistical inference: Confidence intervals for estimating quantitative characteristics of populations based on samples III. Descriptive and inferential statistics for comparing means across groups IV. Hypothesis testing **Agresti and Finlay Ch. 3, Pp. 38 59** **Agresti and Finlay Chs. 4 5** **Agresti and Finlay Ch. 7, pp. 191 201** WEEK 10 SPRING RECESS WEEKS 11 12 ANOVA, Two Way ANOVA **Agresti and Finlay Ch. 12** WEEKS 13 14 I. Statistics for describing relationships among quantitative variables: Correlation and linear regression II. Inferential statistics for correlation and regression **Agresti and Finlay Ch. 9** WEEK 15
Reading tables from journal articles WEEK 16 Student Presentations Assignment Schedule Homework 1 (univariate stats, graphing, CI s) posted Jan. 11, due Jan. 18 in class Homework 2 (2 way contingency tables, t tests, CI s, χ 2 ) posted Jan. 18, due Jan. 27 in Michela s box Homework 3 (multi way contingency tables) posted Feb. 1, due Feb. 8 in class Homework 4 (loglinear models) posted Feb. 8, due Feb. 22 in class Student Presentation Project posted Feb. 22 Midterm Exam in class Feb. 29 Homework 5 (quant variables, test for differences between groups) posted March 7, due March 21 Homework 6 (ANOVA) posted March 21, due April 4 Homework 7 (regression) posted April 11, due April 18 Student Presentation Projects due April 25 Final Exam Monday May 7, 2 4pm