1. (a) Current Place Size Tolerance

(b) Age-16 Place Size Tolerance

(c) Age-16 Place Size Tolerance

Current Place Size

2. (a) A correct table (1) must be percentaged in the direction of the independent variable only/other sibling status and (2) must show that onlies performed better than others on a test of academic achievement.

(b) To create a spurious association, an extraneous variable must be a cause of both sibling status and academic achievement; in other words, the variable must be antecedent to the independent and dependent variables. Among the six possible control variables, only the first three—(1) parental educational achievement, (2) rural/urban residence, (3) gender—are clearly antecedent to both sibling status and academic achievement. The other three variables—(4) parental help with homework, (5) parental expectations about the child’s future, and (6) child-care arrangements prior to elementary school (either nursery school or a relative’s care)—are intervening; that is, these variables would occur between the time that a child’s sibling status is determined and the time that his or her academic achievement is measured. All three antecedent variables might create a spurious association; however, parental education and rural/urban residence are most plausible. The following table controls for parental education. If the original association is spurious, then the relationship will be zero or near zero in each of the partial tables.

Parental Educational Achievement

(c) Any of the three intervening variables might interpret the relationship between sibling status and academic achievement. Consider, for example, parental help with homework; parents with only one child would have more time available to help with homework than parents with more than one child; and helping a child with his or her homework should facilitate academic achievement.

3. When the other independent variables are controlled, college first-year G.P.A. is predicted to (a) increase 0.30 if the student is female rather than male, (b) increase 0.50 for every unit increase in high school G.P.A., and (c) decrease 0.01 for every hour of weekly employment.

4. In Figure 16.7, seniority (years in mill) has two indirect paths leading to union sentiment: one via wage level and deference for an indirect effect of (.17) x (.22) x (−.22) = −.008, and the other via wage level, deference, and activism for an indirect effect of (.17) x (.22) x (−.35) x (.44) = −.006. The total effect of seniority on union sentiment is the sum of the direct and indirect effects: (−.220) + (−.008) + (−.006) = −.234.