Table 9.12, from the 1991 General Social Survey, reports subjects' religious affiliation in 1991 and at age 16, for categories (1) Protestant, (2) Catholic, (3) Jewish, (4) None or other.

a. Test whether the proportions classified as Catholic differed in 1991 and at age 16. Construct a 90% confidence interval for the change in the proportion classified Catholic. Interpret.

b. Show that the symmetry model has G = 32.2 with df = 6. Use residuals to analyze transition patterns between pairs of religions.

c. Show that the quasi-symmetry model has G = 2.0 with df = 3. Interpret.

d. Test marginal homogeneity. Show that the small P-value mainly reflects the large sample size and is due to a small decrease in the proportion classified Catholic and an increase in the proportion classified None or Other, with no evidence of change for other categories.

e. Fit the quasi-independence model, and interpret.

Transcribed Image Text:

Table 9.12 Affiliation at Age 16 1 2 34 3. 4 1 Religious Affiliation Now 2 3 4 863 30 50 320 1 1 27 8 23-3 1880 52 28 1 0 33 Source: 1991 General Social Survey.