Читать книгу Elementary Regression Modeling - Roger A. Wojtkiewicz - Страница 33

1 Appendix A describes how to access the High School Longitudinal Study data that are used to create the examples in this book. Access the data at the National Center for Educational Statistics website and download the variables as specified. The data are also available on the SAGE website for this book.

2 Create the working dataset, as specified in Appendix A, by selecting only those cases where W2STUDENT>0 and X2CONTROL>-6. The resulting dataset should include 20,133 cases.

3 Re-create the table of differences where the independent variable is family structure (FAMSTRUCT) and the dependent variable is math score (X2TXMTSCOR).

4 Re-create the table of differences where the independent variable is family structure and the dependent variable is the log-odds of top 25% in math (HIGHMATH).

5 Create a new table of differences where the independent variable is family structure and the dependent variable is socioeconomic status (X2SES). How do students vary in SES by family structure category?

6 Create a new table of differences where the independent variable is family structure and the dependent variable is private high school (PRIVATE). How do students vary in the log-odds of attending private high school by family structure category?

1. The brief discussion of basic statistics in this book is presented to provide a base for the more complex material that follows. For more thorough discussions of basic statistics, see Frankfort-Nachmias and Leon-Guerrero (2014) and Linneman (2014) at the undergraduate level and Gordon (2012) and Agresti and Finlay (2009) for noncalculus-based discussions at the graduate level.

2. See Fox (2015), Chapter 14, and Kutner, Nachtsheim, Neter, and Li (2005), Chapter 14, for fuller explanations about the necessity for using logistic regression when the dependent variable is dichotomous.