Appendix B - Analyses of family salary data, pre-school duration, quality and
impact on child attainment and progress at entry to primary school17
Prepared for HM Treasury November 2003
Introduction
This paper provides an analysis of the impact of family income on young children’s cognitive
attainment. It also makes comparisons with the impact of quality and duration of pre-school.
Using average cost data for different types of pre-school provider, and effect sizes for different
types of provision. The paper also presents some evidence concerning cost benefit for pre-
school provision. The analyses focus in the main on two child outcomes collected at entry to
primary school (age rising five years). Pre-reading and language attainment were chosen
because they show stronger relationships with child and family background in earlier multilevel
analyses (EPPE Technical Papers 8a & 8b). Evidence of the positive impact of pre-school on
Early Number concepts and social behaviour, especially Peer sociability has been provided in
the main EPPE pre-school results.
Sources of data
The EPPE parental questionnaire survey provides indicators of parental salary in the form of
average yearly salary (before tax). This information was collected in the form of seven income
bands (see Tables B1 and B2 below). This limitation means that it is not possible to explore the
relationships between child attainment and more finely differentiated salary levels. In addition, a
total parent salary measure was created, by combining the mother’s and father’s salary bands.
To do this the mid point of each band was used except for the top band (£65000 plus) where the
conservative estimate of £66000 was chosen. The results are shown in Table B3. The use of
bands means that the salary data provide an approximate indication of relative family salary
levels, but cannot be treated as providing accurate information about actual gross salary. The
total salary measure approximates to a continuous variable. To allow further exploration another
parent salary measure was created, which divided the continuous measure into seven broad joint
income categories (see Table B4). This creation of categories allows the inclusion of an
‘unknown’ category for children for whom no parent salary data were collected and increases
sample size in selected analyses. The use of parent salary categories also allows direct
comparisons with the influence of different amounts of pre-school in multilevel analyses.
In addition to parent salary a measure was created concerning parent work status. This covers
no parent working, mother only, father only, and both parents working see Table B5.
Table B1 Mother’s average yearly salary before tax
|
Frequency |
Percent |
Valid Percent |
Cumulative ______Percent_____ | |
|
Valid 1 £0-4999 |
301 |
9.5 |
13.2 |
13.2 |
|
2 £5000-9999 |
362 |
11.4 |
15.9 |
29.0 |
|
3 £10000-19999 |
381 |
12.0 |
16.7 |
45.7 |
|
4 £20000-29999 |
194 |
6.1 |
8.5 |
54.2 |
|
5 30000-39999 |
74 |
2.3 |
3.2 |
57.5 |
|
6 £40000-64999 |
43 |
1.4 |
1.9 |
59.4 |
|
7 £65000+ |
22 |
.7 |
1.0 |
60.3 |
|
9 not applicable / or absent | ||||
|
mother/voluntary |
906 |
28.6 |
39.7 |
100.0 |
|
Total |
2283 |
72.0 |
100.0 | |
|
Missing -100 missing parent | ||||
|
questionnaire |
719 |
22.7 | ||
|
99 missing answer |
169 |
5.3 | ||
|
Total |
888 |
28.0 | ||
|
Total |
____________3171 |
100.0 |
17 Derived from EPPE Parent Questionnaire conducted autumn 2001/spring 2002. The response
rate to the parent survey was over 77%.
72