|
1-2 years_________________________________________ |
0.322*_______________ |
0.518*__________ |
|
2-3 years_______________________________________ |
0.381*_______________ |
0.527*__________ |
|
3 years plus______________________________________ |
0.385*_______________ |
0.659*__________ |
*p<0.05, #nearly statistically significant
The results show that increased family income has only a modest impact on pre-reading results
for low attaining children (effect size 0.298, verging on statistical significance) for the highest
income band (£37500 plus) compared with the low attaining no salary group.
The importance of GCA is illustrated clearly, with much higher effect sizes for all average ability
children across all income bands, although the highest income average attaining band once
again show the most positive effects in both Language and pre-reading.
These additional results support the earlier conclusion that the net effects for income are
identified for Pre-reading rather than language and are mainly evident at the highest income
levels.
Duration of pre-school still shows significant and moderate effect sizes net of other influences
especially for language attainment.
Quality and duration
Additional analyses were conducted to explore the influence of quality and duration of
attendance effects on child attainment at the start of primary school. Pre school centres were
divided into three groups low (bottom 20%), average (middle 60%) and high (top 20%) based on
total ECERS-E rating, an observational quality measure. These bands were chosen after looking
at the distribution of scores to clearly distinguish the low and high quality centres. Because the
'cut offs' were based on centre scores (considered the most valid approach) the numbers of
children in the bands varied (around 16% of the sample were in centres with the lowest quality
scores but around 24% in centres with the highest scores). Within each quality band children
were further divided on the basis of duration of attendance. It should be noted that due to the
relatively smaller numbers in the low quality band, the sub divisions by duration are broader.
Therefore direct comparisons for the low quality low duration are not possible.
Table B10 shows the net effects for each of the sub-groups.
Table B10 Comparison of Effect sizes for quality and duration
|
Pre-school group (Compared with no duration, no quality i.e. the ‘home’ |
Pre-reading |
Language |
|
Low quality low duration (< 24 months)_______________ |
0.254 |
0.602* |
|
Low quality average duration (24-36 months)_________ |
0.293* |
0.540* |
|
Low quality high duration (36 months plus)____________ |
0.368* |
0.529* |
|
Average quality very low duration (< 12 months)______ |
0.153 |
0.459* |
|
Average quality low duration (12-24 months)__________ |
0.331* |
0.459* |
|
Average quality average duration (24-36 months) |
0.479* |
0.528* |
|
Average quality high duration (36 months plus)________ |
0.545* |
0.672* |
|
High quality very low duration (<12 months)___________ |
0.256*______________ |
0.338*________ |
|
High quality low duration (12-24 months)______________ |
0.381* |
0.526* |
|
High quality average duration (24-36 months)_________ |
0.346* |
0.535* |
|
High quality high duration (36 months plus)____________ |
0.622*______________ |
1.010*________ |
*p<0.05
It can be seen that in comparison with the ‘home’ group all levels of quality and duration show a
significant positive effect compared with none. Overall, longer duration shows a greater benefit
than low duration, irrespective of quality. However, the combination of high quality and high
duration shows a particularly strong effect size (1.01) for language, and fairly large effect for pre-
reading (0.622). To try to distinguish the separate quality effect we can calculate the net
78