Education, Employment and Earnings of Secondary School-Leavers ...
Education, Employment and Earnings of Secondary School-Leavers ...
Education, Employment and Earnings of Secondary School-Leavers ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
5.4 Estimated Private Rates <strong>of</strong> Return to <strong>Education</strong>al QualificationsWe now turn to an examination <strong>of</strong> the estimated private rates <strong>of</strong> return for the twoeducational qualifications used here – senior secondary <strong>and</strong> university. These arecomputed using the differences in the estimated coefficients between adjacentqualifications divided by the difference in years. If we define the interval regressionearnings model parameter for a university qualification as γ U <strong>and</strong> for senior secondaryas γ S , the rate <strong>of</strong> return (RoR) to a university qualification is computed as:RoR – University Qualification =UγUYears- γ-SSYears[9]where U Years is total years in education to acquire a university qualification <strong>and</strong> S Yearsis the corresponding number <strong>of</strong> years taken to acquire a senior secondaryqualification. The rate <strong>of</strong> return to senior secondary schooling can then be computedas:RoR – Senior <strong>Secondary</strong> Qualification =SYearsγS- JYears[10]where J Years is total years in education to acquire a junior certification qualification.The sampling variances for the estimated rates <strong>of</strong> return are easily computed using theconventional formula.The estimated rates return reported in table 7 are the exponentiated versions <strong>of</strong> [9] <strong>and</strong>[10] expressed in percentages <strong>and</strong> the corresponding sampling variances areconstructed using the delta method. In respect <strong>of</strong> employees the point estimate for thesenior secondary qualification is just under nine per cent but is situated within a largeconfidence interval <strong>and</strong> is only significant at the 10 per cent level. The comparablepoint estimate for the self-employed is close to five per cent but is even moreimprecisely estimated.At a conventional level <strong>of</strong> statistical significance, theproposition that the sample data were generated from an underlying population with azero return for the senior secondary qualification cannot be rejected for the selfemployed.26