Evaluation of the Australian Wage Subsidy Special Youth ...
Evaluation of the Australian Wage Subsidy Special Youth ... Evaluation of the Australian Wage Subsidy Special Youth ...
206 resp 14 (1.66) (1.27) Catholic 0.061 -0.006 (0.37) (-0.03) Presbyterian 0.322 0.321 (1.34) (1.33) Methodist 0.015 0.262 (0.06) (1.01) Other Christian 0.074 0.057 (0.27) (0.21) Other religion 0.176 0.142 (0.80) (0.58) No religion 0.138 0.145 (0.67) (0.74) Observations 1283 1283 Student’s t statistics in brackets; * significant at 5%; ** significant at 1% Weighted with combined weight for attrition, non-response and survey design. NOTE 1: Results Column 1 are sourced from Table 4 and Table 5 pages 18-21 Richardson (1998). Base categories: European ethnic origin, state interviewed in 1984 NSW/ACT, government school, highest qualification in 1984 year 10 at school, longest job by 1984 is 1 year, lived mostly in state capital city until respondent aged 14, English is first language, father clerical worker when respondent aged 14, mother clerical worker when respondent aged 14, religion brought up in is Anglican.
207 6.2 Results of weighting the PSM Weighting the Heckman bivariate probit is a relatively simple task of ensuring the weights are used in estimating the model. In contrast, weighting the PSM requires a little more attention. Frölich et al. (2001) is one of the few applications of PSM found where weights have been applied to reflect the selective nature of the sample surveyed. As the application of weights in PSM is not generally treated in the literature, details of the approach are made clear. 146 6.2.1 Weighting protocol In order to correctly adjust the PSM for the attrition, two steps need to be taken. The weighting protocol steps are shown in Table 6.2. The probit model of SYETP used to estimate the propensity scores, must be weighted with the attrition weight. The common support for the weighted propensity needs to be checked. Once again, the minimum and maximum of the weighted SYETP propensity scores are used to define the common support. Once the matches have been found for these estimated propensity scores, then the final match needs to be re-weighted. The balance of the treatment group in the population is the only consideration. The weights that formerly applied to the comparison cases are then discarded. Instead, the weight that applies to the SYETP case to which they are matched is donated to the comparison case. In this way, the final match is weighted using only the weights of the treated. This is done so that only the characteristics of the treated have influence, as the parameter of interest is the impact of treatment on the treated. One other important step also applies when weighting. In assessing the standardized bias for the performance of the match, weighted mean and variance statistics must be used to construct the standardized bias. Table 6.2 Weighting protocol steps • Weight the propensity probit • Weight the final match, using only the weight of the treated; where comparison case used with replacement, sum the applicable treated case weights. 146 I am grateful to Michael Lechner who in response to an email query made clear the steps in the protocol.
- Page 171 and 172: 155 5.5.2 Univariate examination of
- Page 173 and 174: 157 lower, the job lengths are only
- Page 175 and 176: 159 Work limited by health 1984 0.1
- Page 177 and 178: 161 The characteristics of the SYET
- Page 179 and 180: 163 para-professional Mother not em
- Page 181 and 182: 165 comparison group where the shar
- Page 183 and 184: 167 5.5.4 Attrition: natural attrit
- Page 185 and 186: 169 both sources that impose change
- Page 187 and 188: 171 para-professional Father not em
- Page 189 and 190: 173 work in later sections, this su
- Page 191 and 192: 175 Table 5.6: Effect of selection/
- Page 193 and 194: 177 appropriate to discard these fr
- Page 195 and 196: 179 Australia/Tasmania. Amongst tho
- Page 197 and 198: 181 Table 5.5a Summary statistics b
- Page 199 and 200: 183 5.6.1.2 Effects of the non-resp
- Page 201 and 202: 185 3 years + -0.35 -0.47 -0.34 -0.
- Page 203 and 204: 187 5.7 Multivariate analysis of ef
- Page 205 and 206: 189 proportion of time spent unempl
- Page 207 and 208: 191 post-school qualification, and
- Page 209 and 210: 193 Generally, those variables foun
- Page 211 and 212: 195 longj0 Longest job by 1984 < 1
- Page 213 and 214: 197 adopted in order to maintain co
- Page 215 and 216: 199 6: Study 4 Weighting to counter
- Page 217 and 218: 201 Table 6.1, part A Employment eq
- Page 219 and 220: 203 Methodist 0.133 0.261 (0.77) (1
- Page 221: 205 CEP referrals 1984 0.143* 0.128
- Page 225 and 226: 209 The distribution of the propens
- Page 227 and 228: 211 Table 6.3 Weighted probit used
- Page 229 and 230: 213 (0.76) Tradesperson mtrad 0.20
- Page 231 and 232: 215 Table 6.5 Summary statistics fo
- Page 233 and 234: 217 Table 6.7 Matching results, sin
- Page 235 and 236: 219 6.3 Discussion The comparison o
- Page 237 and 238: 221 the selection into SYETP and th
- Page 239 and 240: 223 Heteroskedasticity is a violati
- Page 241 and 242: 225 Table 7.1, Part A Employment eq
- Page 243 and 244: 227 (1.26) (1.28) (1.16) Mothers oc
- Page 245 and 246: 229 Other Post-School qualification
- Page 247 and 248: 231 7.1.2 Exclusion restriction in
- Page 249 and 250: 233 Finally, the third panel of new
- Page 251 and 252: 235 Table 7.2 summary of changes to
- Page 253 and 254: 237 schooling that was statisticall
- Page 255 and 256: 239 (0.25) (0.21) 3 years + -0.50 -
- Page 257 and 258: 241 Table 7.4 Summary of distributi
- Page 259 and 260: 243 7.2.2 Propensity score matching
- Page 261 and 262: 245 original model, but with the pe
- Page 263 and 264: 247 maintained, then the Heckman bi
- Page 265 and 266: 249 Table 7.6 Weighted Probit used
- Page 267 and 268: 251 Table 7.7 Summary of distributi
- Page 269 and 270: 253 8: Summary and Conclusions The
- Page 271 and 272: 255 over which it ran, the review m
206<br />
resp 14<br />
(1.66) (1.27)<br />
Catholic 0.061 -0.006<br />
(0.37) (-0.03)<br />
Presbyterian 0.322 0.321<br />
(1.34) (1.33)<br />
Methodist 0.015 0.262<br />
(0.06) (1.01)<br />
O<strong>the</strong>r Christian 0.074 0.057<br />
(0.27) (0.21)<br />
O<strong>the</strong>r religion 0.176 0.142<br />
(0.80) (0.58)<br />
No religion 0.138 0.145<br />
(0.67) (0.74)<br />
Observations 1283 1283<br />
Student’s t statistics in brackets; * significant at 5%; ** significant at 1%<br />
Weighted with combined weight for attrition, non-response and survey design.<br />
NOTE 1: Results Column 1 are sourced from Table 4 and Table 5 pages 18-21 Richardson (1998).<br />
Base categories: European ethnic origin, state interviewed in 1984 NSW/ACT, government school, highest<br />
qualification in 1984 year 10 at school, longest job by 1984 is 1 year, lived mostly in state capital city until<br />
respondent aged 14, English is first language, fa<strong>the</strong>r clerical worker when respondent aged 14, mo<strong>the</strong>r<br />
clerical worker when respondent aged 14, religion brought up in is Anglican.