Midterm 2 Review Sheet Directions: Page 5 provides data from a sample of GSS 2002 respondents. A codebook is also provided, which corresponds to the data. Use the information contained in the data and the codebook to answer exam questions 1 - 6. Page 5 also contains some useful formulas which will help you perform calculations, as well as some descriptive statistics on the variables in the data set. Page 6 contains a bivariate table of highest degree earned by social class. 1. Using the grid below, create a bivariate table showing the relationship between the nominal variables in the data set. Decide which of the variables could logically be the independent variable and which could be the dependent variable. Make the independent variable the column variable. In your table, include both column percentages and the column frequencies as well as all appropriate marginals. Be sure that your table includes all of the essential elements of a table. (7 points) 2. Using the table you constructed for question 1, calculate the appropriate measure of association for these variables. Interpret the meaning of the measure of association. What does it say about the strength and direction of the association between the two variables? How can this association be described using plain English (or everyday non-technical language)? (6 points)Soci52 Midterm 2 PID#: ____________________ 1 3. Suppose you were interested in testing a theory about education and television viewing. Imagine that you thought that an individual’s hours of daily television viewing depended on the respondent’s educational attainment. Use data from the GSS sample to estimate a simple linear regression model testing this idea. For this question only calculate the slope and the intercept of the linear regression model. (8 points)Soci52 Midterm 2 PID#: ____________________ 2 4. Interpret the meaning of the slope and the intercept of the linear regression model that you estimated for question 3. What does it say about the strength and direction of the association between the two variables? How can this association be described using plain English (or everyday non-technical language)? (4 points) 5. Calculate two more statistics: the coefficient of determination (r2) and the coefficient of correlation (r) for two variables that were included in the linear regression model that you estimated in question 3. (4 points)Soci52 Midterm 2 PID#: ____________________ 3 6. Interpret both the coefficient of determination (r2) and the coefficient of correlation (r). What do these statistics say about the strength and direction of the association between the two variables? (4 points) 7. Use the bivariate table of degree by class (found on page 7) to calculate the proper measure of association between these two variables. (4 points)Soci52 Midterm 2 PID#: ____________________ 4 8. Interpret the meaning of the measure of association. What does it say about the strength and direction of the association between the two variables? How can this association be described using plain English (or everyday non-technical language)? (3 points)Soci52 Midterm 2 PID#: ____________________ 5 GSS Data ID educ postlife sex tvhours 1 14 2 1 1 2 19 2 1 1 3 15 2 2 12 4 12 1 1 0 5 16 1 2 1 6 15 2 1 4 7 14 1 1 2 8 12 1 2 6 9 12 2 1 4 10 16 1 1 2 Useful Statistics: Variable Mean Variance educ 14.50 4.944 postlife 1.50 0.278 sex 1.30 0.233 tvhours 3.30 12.678 Useful Formulas: 121EEE−=λ NdNsNdNs+−=γ 2XYXSSb = ; )(XbYa −= 1)(22−−=∑NXXSX 1))((−−−=∑NYYXXSYX 222YXYXSSSr = YXYXSSSr = Codebook ID: RESPONDENT’S IDENTIFICATION NUMBER EDUC: HIGHEST YEAR OF SCHOOL COMPLETED What is your highest year of school completed? Value Labels: CODE ACTUAL VALUES POSTLIFE: BELIEF IN LIFE AFTER DEATH Do you believe there is life after death? Value Labels: 1 YES 2 NO SEX: RESPONDENT’S SEX The sex of the respondent. Value Labels: 1 MALE 2 FEMALE TVHOURS: HOURS PER DAY WATCHING TV On an average day, about how many hours do you personally watch television? Value Labels: CODE ACTUAL VALUESoci52 Midterm 2 PID#: ____________________ 6 The following bivariate table shows the relationship between a respondent’s highest degree earned (recoded into three categories) and a respondent’s subjective social class (recoded into three categories): RECODE OF DEGREE * RECODE OF CLASS CrosstabulationCount36 27 1 641 6 1 84 13 2 192 8 1 1143 54 5 102HIGH SCHOOLJUNIOR COLLEGEBACHELORGRADUATERECODE OFDEGREETotalLOWERCLASSMIDDLECLASSUPPERCLASSRECODE OF