This study of sibling resemblance in educational attainment seeks to determine whether the effect of family background on educational attainment differs according to family structure, birth order, and sex.
In order to address these research problems, I analyze sibling data from both waves of the National Survey of Families and Households and use a form of the multiple indicator, multiple cause (MIMIC) model proposed by Hauser and Goldberger (1971). The MIMIC model includes a latent common family background variable which takes into account family influences common to pairs of siblings but not captured by the set of measured family background variables.
In the first part of the analysis, I consider the educational attainment and social background of sibling pairs coming from two-parent dual-earner families, two-parent single-earner families, and single-parent single-earner families. I did find the effects of family background on educational attainment to differ among these three family groups. However, I also found family background effects to be more similar among the two-parent families than among the single-parent families. Single-parenthood seems to alter the effects that family background has on children's educational attainment much more than does mother's employment.
In the second part of the analysis, I attempt to determine whether family background affects educational attainment differently according to the birth order and sex of children. I limit my analysis to two-parent families and consider the educational attainment and social background of four types of sibling pairs (older-brother younger-brother, older-brother younger-sister, older-sister younger-sister, and older sister-younger brother). For the most part, I found family background effects to be similar among all children, regardless of birth order and sex. However, I did find sibling pairs with an older brother to exhibit a greater degree of resemblance in educational attainment; a resemblance not explained by the measured background factors. This finding lends support to Benin and Johnson's (1984) facilitation hypothesis and leads me to conclude that family background does affect children's educational attainment differently according to birth order and sex.