Evaluation of Maintenance Allowance (EMAs): Evaluation of the East Ayrshire Pilot
Appendix 2 Statistical Models
Table A.1: Logit estimation of the likelihood of continuing studying (at school and college) after completing S4: coefficients (and standard errors) (Table 5.14 in the main report)
| Model 1 | Model 2 | Model 3 |
Intercept | 1.46*** (0.12) | 1.98*** (0.22) | 1.64*** (0.33) |
Cohort (ref 98-99) | | | |
99-00 | 0.43* (0.18) | 0.41* (0.18) | 0.58** (0.21) |
00-01 | 0.49** (0.18) | 0.51** (0.18) | 0.65** (0.20) |
Female | | 0.40** (0.15) | ns |
Low-income families | | -0.99*** (0.19) | ns |
Attainment at S4 (scale from 0 to 33) | | | 0.20*** (0.01) |
Missing information on attainment in S4 | | | -0.95* (0.43) |
Contrasts with the control area | | | |
Lanarkshire | 0.49* (0.21) | ns | ns |
Lanarkshire*99-00 | ns | ns | -0.74* (0.33) |
Lanarkshire*00-01 | -0.56* (0.27) | -0.55* (0.28) | -0.63* (0.32) |
Total number of cases (unweighted): 1459 in East Ayrshire and 1256 in Lanarkshire
Note: Only significant coefficients are reported. No significant interaction effects have been found between Lanarkshire and the other independent variables in the model.
Note2: The introduction of previous educational attainment cancel out the effect of gender and family of origin. This is due to the high correlation existing between these variables. Women and pupils from more advantaged social background tend to have higher educational achievement. Thus, after controlling for the effect of educational achievement, the effects of gender and family of origin are no more significant.
Note 3: Confidence levels: *** denotes the effect is significant at the 0.001 level, ** at the 0.01 level and * at the 0.05 level.
Table A.2 Logit estimation of the likelihood of completing S5: coefficients (and standard errors) (Table 5.17 in main report)
| Model 1 | Model 2 | Model 3 | Model 4 |
Intercept | 0.70*** (0.10) | 1.29*** (0.20) | -3.76*** (0.21) | -3.97*** (0.47) |
Cohort (ref 98-99) | | | | |
99-00 | 0.34* (0.14) | 0.31* (0.15) | 0.49* (0.19) | 0.48* (0.20) |
Female | | 0.54* (0.15) | 0.43* (0.19) | 0.39* (0.20) |
Low-income families | | -1.20*** (0.19) | -0.47* (0.24) | -0.78** (0.25) |
Attainment at S4 | | | 0.27*** (0.02) | 0.27*** (0.02) |
Missing information on attainment in S4 | | | -0.78 (0.43) | -0.78 (0.43) |
Grants at school or college | | | | |
EMA | | | | 1.35*** (0.25) |
EMA (missing) | | | | ns |
Bursary | | | | 0.80*** (0.25) |
Bursary 2 | | | | ns |
| | | | |
Contrasts with the control area | | | | |
Lanarkshire | 0.49* (0.16) | ns | -0.08 (0.63) | 0.30 (0.68) |
Lanarkshire*99-00 | -0.59 (0.22) | -0.57* (0.23) | -1.11*** (0.30) | -1.10*** (0.30) |
Lanarkshire*Attainment at S4 | | | 0.06* (0.03) | 0.06* (0.03) |
Lanarkshire *Missing information on attainment in S4 | | | -2.27* (1.81) | -2.34* (1.22) |
Total number of cases (unweighted): 890 in East Ayrshire and 742 in Lanarkshire
Note: Only significant coefficients are reported.
Table A.3: Logit estimation of the likelihood of entering S6: coefficients (and standard errors) (Table 5.18 in main report)
| Model 1 | Model 2 | Model 3 | Model 4 |
Intercept | -0.02 (0.09) | -0.37* (0.16) | -3.87*** (0.37) | -3.84*** (0.41) |
Cohort (ref 98-99) | | | | |
99-00 | ns | ns | ns | ns |
Female | | 0.45** (0.14) | ns | ns |
Low-income families | | -0.94*** (0.15) | -0.23 (0.18) | -0.42* (0.19) |
Attainment at S4 | | | 0.21*** (0.01) | 0.20*** (0.01) |
Missing information on attainment in S4 | | | -0.98* (0.51) | ns |
Grants at school or college | | | | |
EMA | | | | 0.94*** (0.19) |
EMA (missing) | | | | ns |
Bursary | | | | -0.13 (0.21) |
Bursary 2 | | | | ns |
| | | | |
Contrasts with the control area | | | | |
Lanarkshire | ns | ns | ns | 0.30 (0.68) |
Lanarkshire*99-00 | ns | ns | ns | ns |
Lanarkshire *Bursary | | | | -0.91** (0.33) |
Total number of cases (unweighted): 890 in East Ayrshire and 742 in Lanarkshire
Note: Only significant coefficients are reported.
Table A.4: OLS regression of the likelihood of achieving higher educational attainment at school: coefficients (and standard errors) (Table 5.21 in main report)
| Model 1 | Model 2 | Model 3 | Model 4 |
Intercept | 9.99***(0.39) | 10.59***(0.63) | -5.98***(1.03) | -7.15*** (1.17) |
Female | | 1.59**(0.55) | n.s. | n.s. |
Low-income families | | -2.38***(0.56) | n.s. | n.s. |
Attainment at S4 | | | 0.82*** (0.04) | 0.85*** (0.04) |
Missing information on attainment in S4 | | | n.s. | n.s. |
Grants at school | | | | |
EMA | | | | n.s. |
Bursary | | | | 1.25*(0.52) |
Bursary 2 (travel or clothing grant) | | | | n.s. |
| | | | |
Contrasts with the control area | | | | |
Lanarkshire | 1.29*(0.54) | 1.23* (0.53) | 1.27** (0.44) | 1.59** (0.53) |
Total number of cases (unweighted): 716 (375 in East Ayrshire; 341 in Lanarkshire).
Note: Only significant coefficients are reported.