Appendix A: Statistical Tests

The tests of significance used in this E.D. TAB are based on Student's t statistics for the comparison of individual estimates and for bivariate relationships and mean comparisons. To test the differences between estimates, unbiased estimates of standard errors were used, derived by jackknife replication methods. To test for a difference between two subgroups in the population proportion having a particular characteristic, say P₁ versus P₂, the test statistic is computed as:

where p_i is the estimated proportion of subgroup i (i = 1, 2) having the particular characteristic and s.e(p_i ) is the standard error of that estimate. Thus, if p₁ is for the 39 percent of male students who had after-school arrangements, with a standard error of 0.9, and p₂ is for the 39 percent of female students who had after-school arrangements, with a standard error of 0.9, then the t-value is equal to 0.00.

For the aforementioned comparison, the decision rule is to reject the null hypothesis (i.e., that there is no difference between the two groups in the population in terms of the proportion having the characteristic) if , where is the value such that the probability that a Student's t random variable with df degrees of freedom exceeds that value is .

Top