9. Table 8 26 shows an SPSS printout for some analyses for a data set taken from the 1991 General Social Survey, with variables race and party identification.

a) Form the conditonal distribution on party identification, for each race. Interpret

b) Report the expected fiequency for the first cell, and show how SPSS obtained it

c) Test the hypothesis of independence between party identification and race. Report the test statistic. P-value, and interpret.

d) Use the adjusted residuals to describe the evidence about association

e) Summarize the association between race and whether one is a Democrat or a Republi- can. by computing the odds ratio or by companng percentages. Interpret. TABLE 8.26 Count Exp Val PARTY Adj Res Row democr indep repub Total RACE black 103 15 11 129 58.4 15.8 54.8 8.5 -.2 -8.4 white 341 105 405 851 385.6 104.2 361.2 -8.5 .2 B.4 Column Total 444 120 416 980 Chi-Square Value DF Significance ---- Pearson 79.431 2 Likelihood Ratio 90.331 2 .00000 .00000 TABLE 8.27 Teenage Birth Control Strongly Strongly Premantal Sex Disagree Disagree Agree Agree Always wrong 81 (7.6) 68 (3.1) 60 (-4.1) 38 (-48) Almost always wrong 24 (23) 26 (18) 29 (-0.8) 14 (-2.8) Wrong only sometimes 18 (-27) 41 (10) 74 (2.2) 42 (-1.0) Not wrong at all 36 (-61) 57 (-4.6) 161 (24) 157 (6.8) Source 1991 General Social Survey.