Suchergebnisse

PAPER REVIEWS A NUMBER OF RECENT STUDIES OF THE INCOMESCHOOLING-ABILITY NEXUS USING SIBLING DATA AND DISCUSSES THE PROBLEM OF IDENTIFICATION IN SUCH STUDIES. SPECIAL EMPHASIS IS PLACED ON THE ROLE OF ERRORS IN VARIABLES, CONCLUDING THAT MODEST ERROR LEVELS CAN ACCOUNT FOR MUCH OF OBSERVED DIFFERENCE BETWEEN TOTAL AND WITHIN-FAMILY ESTIMATES OF RETURNS TO SCHOOLING.

AbstractIn sibling models with categorical outcomes the question arises of how best to calculate the intraclass correlation, ICC. We show that, for this purpose, the random effects linear probability model is preferable to a random effects non-linear probability model, such as a logit or probit. This is because, for a binary outcome, the ICC derived from a random effects linear probability model is a non-parametric estimate of the ICC, equivalent to a statistic called Cohen's κ. Furthermore, because κ can be calculated when the outcome has more than two categories, we can use the random effects linear probability model to compute a single ICC in cases with more than two outcome categories. Lastly, ICCs are often compared between groups to show the degree to which sibling differences vary between groups: we show that when the outcome is categorical these comparisons are invalid. We suggest alternative measures for this purpose.

What struck me most forcefully in Townsend's critique (this issue) was a pattern he observed: A large preponderance of the changes he documented in Sulloway's numeric or graphic descriptions of data resulted in improvements. The later data point was almost always a better fit to Sulloway's predictions than the earlier one. Townsend suggested that further analysis would be appropriate, which made me wonder why he stopped where he did, without doing any statistical tests on his findings. A possible next step would be to calculate the likelihood that the observed proportion of changes in the confirming direction occurred by chance alone.

SummaryRelationships were examined between sibling variables, perceptions of learning environments and aspirations for Australian adolescents from family contexts defined conjointly by human and social capital. Generally, the findings using a regression approach indicated that sibling variables continued to have significant and often complex associations with adolescents' aspirations after taking into account relationships between perceptions of learning environments and aspirations. Also, relationships between sibling variables, perceptions of learning environments and aspirations, varied for adolescents from the different family contexts.