02.06.2014 Views

Evaluation of the Australian Wage Subsidy Special Youth ...

Evaluation of the Australian Wage Subsidy Special Youth ...

Evaluation of the Australian Wage Subsidy Special Youth ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

164<br />

5.5.3.2 The effect <strong>of</strong> sample reduction on <strong>the</strong> difference between SYETP and comparison<br />

groups<br />

Overall, in <strong>the</strong> previous section it can be seen that <strong>the</strong> effect <strong>of</strong> <strong>the</strong> sample reduction is<br />

not equal on <strong>the</strong> treatment and <strong>the</strong> comparison groups’ pr<strong>of</strong>ile for <strong>the</strong>se characteristics<br />

examined. This is crucial as <strong>the</strong> aim <strong>of</strong> later modelling <strong>of</strong> <strong>the</strong> treatment effect is <strong>the</strong><br />

comparison <strong>of</strong> <strong>the</strong> treatment and comparison groups. The effect <strong>of</strong> sample reduction on<br />

<strong>the</strong> difference in pr<strong>of</strong>iles <strong>of</strong> <strong>the</strong> treatment and comparison groups is now explored. This<br />

provides a different perspective on how <strong>the</strong> data reduction impacts on <strong>the</strong> analysis.<br />

The contrast between <strong>the</strong> treatment and comparison group is summarised in <strong>the</strong> absolute<br />

mean difference that is shown in Table 5.4. The first column <strong>of</strong> Table 5.4 gives <strong>the</strong> presample-<br />

reduction differential while <strong>the</strong> second column is <strong>the</strong> post-sample-reduction<br />

statistic. Statistically significant differences between <strong>the</strong>se means are shown with a star<br />

indicating a one percent level <strong>of</strong> significance. Almost all <strong>the</strong> differences are statistically<br />

significant. However <strong>the</strong> differences between <strong>the</strong> treatment and comparison prior to<br />

sample reduction are not <strong>the</strong> same as those present after sample reduction.<br />

The lack <strong>of</strong> correspondence between <strong>the</strong> treatment and comparison group before sample<br />

reduction is quite great. Table 5.4 shows <strong>the</strong> pr<strong>of</strong>ile prior to sample reduction to be<br />

markedly different for <strong>the</strong> treatment and comparison groups for certain characteristics,<br />

and <strong>the</strong> scale <strong>of</strong> contrast can be illustrated by considering <strong>the</strong> figures underlying column 1<br />

<strong>of</strong> Table 5.4. Far fewer <strong>of</strong> those who enter <strong>the</strong> SYETP were married in 1984 (3.9 per cent)<br />

than amongst <strong>the</strong> comparison group (11.7 per cent). The educational pr<strong>of</strong>ile differs for<br />

those on SYETP compared to <strong>the</strong> comparison group. Fewer on SYETP have post-school<br />

qualifications (15.5 per cent) than in <strong>the</strong> comparison group (26.7 per cent), while more <strong>of</strong><br />

those on SYETP have schooling <strong>of</strong> year 10 (36.1 per cent), year 11 (16.1 per cent) and 12<br />

(19.4 per cent) than do those in <strong>the</strong> comparison group (at 29.9 per cent, 12.1 per cent and<br />

15.1 per cent respectively). The work history pr<strong>of</strong>ile <strong>of</strong> those who entered SYETP is also<br />

quite different to that <strong>of</strong> <strong>the</strong> comparison group. The share <strong>of</strong> those on SYETP with work<br />

experience <strong>of</strong> less than a year is 58.1 per cent, which contrasts strongly with <strong>the</strong>

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!