Marriage Market , Assortative Mating

1. Combined the results of six papers, what the agreement and disagreement about the trend of educational homogamy in U.S among the different authors (SU&L, Raymo& Xie, and Schwatz)?

2. Is there any difference among the trends of educational homogamy in prevailing marriage and newlyweds?

3. How can we understand the role of population changes in explaining the changes in educational assortative mating, such as the changes in distribution of educational attainment across gender and race, "disturbance" of cohabitation, or imbalanced sex-ratio in some Asian countries?

4. Kalmijn (1991) suggests that "marriage selection in the U.S. is more strongly oriented toward education than toward social origins." In education literature, some authors find that educational attainment is strongly correlated with the family background. So, could it possible that social origins influence the educational attainment to some degree and therefore the significance of social origins is reinforced in assortative mating?

5. What are the implications of increasing/decreasing educational homogamy for social openness?

6. Large body of literature suggests that ethnic intermarriage is one of important indicators of social distance or social openness. If so, what is the trend of ethnic intermarriage in the U.S.? Is there any correlation between educational attainment and ethnic intermarriage? Are impacts of education and race on assortative mating same?

7. Are there any differences of assortative marriage pattern among different birth cohorts? Do cohabitors have same pattern of assortative mating as the married?

8. How about assortative mating pattern by occupational status (i.e. managerial jobs vs. service jobs) in the same level of education (high school grads, college graduate, etc)?

Data and Methods

1. Which kind of data is more appropriate to analyze the trend and causes of assortative mating? Cross-sectional vs. Longitudinal data?

2. How can the census be used to analyze the assortative mating?

3. What is the strength of period measurement and cohort measurement in measuring the assortative mating? For example, prevailing marriage vs. newlywed.

4. What is the log-linear model, parsimonious log-linear model? How these models apply in the dataset?

5. What is the logic of constructing model? Is stepwise the best way to find the best model?

6. What are potential problems of cross-national comparisons?

Additional questions

1) The language used in several of these articles tends to imply that securing a mate with a higher level of education than one's own is a desirable thing. For instance, Schwartz & Mare and Qian both discuss crossing educational "barriers" in marriage and cohabiting. The reasoning probably arises out of the earnings potential associated with education levels. Are there other reasons that one may or may not want to find a more educated partner?

2) Most analyses of assortative mating use log-linear modeling techniques. Why? What are some other ways one might meaningfully model spouse pairing?

3) Demographically speaking, what is a major limitation in the way that patterns of assortative mating are typically analyzed?

4) In Kalmijn's paper, what are the relative roles of structural changes vs. changes in individual preferences on the trends that he describes?

5) How do Schwartz and Mare describe the relationship between age at marriage, timing of marriage relative to schooling, and educational homogamy? Why do they hypothesize weakened educational homogamy with increased ages at marriage?

6) What is a "crossing model"? What is the model able (unable) to control for? When is it appropriate (inappropriate) to apply such a model to study assortative mating?

7) Would the presence of interracial marriage affect Raley's conclusions? Does her conclusion shed some light on what might happen if there were more interracial marriage

8) Raymo and Xie critique the paper by Smits and colleagues as an example of "reading history sideways." What is meant by this expression and why is this a particularly problematic way to evaluate patters of social/demographic change? (note that the paper by Arland Thornton cited in the first week's readings is very helpful - Thornton, Arland. 2001. "The Developmental Paradigm, Reading History Sideways, and Family Change." Demography 38:449-465.)