51

# never married parents is the greatest with mean scores for children from intact families

13.47 points higher than the mean for students from divorced parents.

# The data in Table 1 indicate that students from intact families have higher unadjusted

mean scores than students from all other family structure types for all academic

achievement measures. It is worth mentioning that the lowest mean scores for all

achievement variables are associated with the never married category. Also noteworthy is

the fact that unadjusted mean scores for students whose parents are divorced are slightly

lower than the scores for children from intact families for three of the four outcome

measures — reading, math, and combined math/reading variables. The largest difference

in unadjusted means across family structure types is indicated by the 1st wave follow-up

math variable in which the mean for students from intact families is 13.5 points higher

than the mean for students whose parents never married. As the name indicates, each

unadjusted mean score in Table 1 is not adjusted to take into account the effect any other

demographic characteristics. In evaluating the regression coefficients in Table 2, it is

important to remember that the regression equation produces coefficients after controlling

for all other effects in the equation. For example, the regression equation renders

coefficients for race after adjusting for any differences in parental involvement, SES,

gender etc.

In Table 2 below, family structure is shown to have a small but statistically

significant effect on academic achievement. Standardized family structure coefficients

range from a high of B =.18 (p < .001) for the 1^{st }wave follow-up math score to a low of