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

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

241<br />

Table 7.4 Summary <strong>of</strong> distribution <strong>of</strong> propensity, exclude CEP referrals<br />

Distribution <strong>of</strong> estimated propensity for SYETP treatment group<br />

Percentiles Smallest<br />

1% .0096799 .0080714<br />

5% .0209195 .0096799<br />

10% .0391096 .011608 Obs 104<br />

25% .0818465 .0178796 Sum <strong>of</strong> Wgt. 104<br />

50% .1352024 Mean .1653081<br />

Largest Std. Dev. .1161794<br />

75% .2308325 .4278012<br />

90% .3124363 .456956 Variance .0134977<br />

95% .4151957 .4614879 Skewness 1.059129<br />

99% .4614879 .6016818 Kurtosis 4.17753<br />

Distribution <strong>of</strong> estimated propensity for comparison group<br />

Percentiles Smallest<br />

1% .0003765 .0000136<br />

5% .0025803 .0000267<br />

10% .005414 .0000539 Obs 1179<br />

25% .0176855 .0000611 Sum <strong>of</strong> Wgt. 1179<br />

50% .0492093 Mean .0743632<br />

Largest Std. Dev. .0793953<br />

75% .1044401 .4422829<br />

90% .1755594 .4788667 Variance .0063036<br />

95% .2357188 .4821222 Skewness 2.031504<br />

99% .3915129 .5711446 Kurtosis 8.35458<br />

The new matching results are shown in column 2 <strong>of</strong> Table 7.5. To better facilitate<br />

comparison, <strong>the</strong> former results with CEP referrals included are shown in <strong>the</strong> first column.<br />

The matching results are now compared, to show <strong>the</strong> effect <strong>of</strong> removing CEP referrals<br />

from <strong>the</strong> specification. The employment effect falls slightly in size, from eight percentage<br />

points to six, and statistical significance drops very low. The number <strong>of</strong> common support<br />

cases discarded from <strong>the</strong> SYETP treatment group does not change. However <strong>the</strong> number<br />

<strong>of</strong> SYETP matched falls slightly, and <strong>the</strong> number <strong>of</strong> comparison cases used to match to<br />

<strong>the</strong>m also falls, with more comparison cases used with replacement than before. The<br />

mean difference in <strong>the</strong> propensity scores after matching is <strong>of</strong> <strong>the</strong> same magnitude as<br />

formerly, however <strong>the</strong> standard deviation is now slightly larger. The mean standardized<br />

bias has risen from 11.04 to 16.15. Overall, <strong>the</strong> efficiency <strong>of</strong> <strong>the</strong> match is judged to be<br />

poorer than before.

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

Saved successfully!

Ooh no, something went wrong!