Publication - Research and analysis

Scottish Study of Early Learning and Childcare - ELC leavers: phase 2 report

Published: 27 Aug 2020

Findings from the second phase of the Scottish Study of Early Learning and Childcare (SSELC), a research project established to evaluate the expansion of early learning and childcare (ELC) in Scotland.

108 page PDF

1.1 MB

108 page PDF

1.1 MB

Contents
Scottish Study of Early Learning and Childcare - ELC leavers: phase 2 report
Appendix D – Regression analysis

108 page PDF

1.1 MB

Appendix D – Regression analysis

Tables D1 and D2 show the results of logistic regression analysis of whether a child has delayed development on at least two domains of the Ages and Stages Questionnaire and of raised / high score on the Strengths and Difficulties Questionnaire total difficulties scale.

Logistic regression analysis is a method of summarising the relationship between a binary 'outcome' variable and one or more 'predictor' variables. It allows us to estimate the odds of a child having a score of '1' on the outcome variable (as opposed to '0') from knowledge of their scores on the predictor variables. In the model shown in Table D1 the score of '1' on the dependent variable refers to exhibiting delayed development on two or more of the ASQ domains, while a '0' refers to exhibiting no delayed development, or delayed development on just one of the domains.

Logistic regression allows us to consider multiple relationships at the same time and to identify those relationships between a predictor variable and the outcome variable which remain statistically significant even when we take into account other predictor variables. For those variables that do remain significant we can say that they show an independent association with the outcome variable while controlling all other factors in the model.

Tables D1 and D2 show how the odds for each category of each predictor variable compare with the odds for the reference category. An odds ratio of greater than 1 indicates that, holding all other factors constant, there is an increased likelihood of a child in that category being in the category '1' for the outcome variable compared with a child in the base category. For example, in Table D1, the odds ratio of 4.09 for the category 'Male' means that boys are more likely than girls (the base category) to exhibit delayed development on two or more of the ASQ domains (and the odds of a boy exhibiting delayed development are 4.09 times those for a girl, holding all other factors constant). Conversely, an odds ratio of below 1 means they have lower odds of exhibiting delayed development than respondents in the reference category.

Because data are taken from a sample, we recognise that the odds ratios are only estimates, so we also include confidence intervals around each estimate. If the survey were to be repeated, we would expect the true value to fall within these odds ratios 95 times out of 100.

Two measures of statistical significance are provided. The first is for the comparison between a particular category and the base category, while the second is for the variable as a whole. Where the independent variable has just two categories, these are the same. A significance level of 0.05 or less indicates that there is less than a 5% chance we would have found these differences between the categories just by chance if in fact no such difference exists, hence we can say that we are 95% sure there is a relationship between the predictor and outcome variables. A level of <0.001 indicates that there is a less than 0.1% chance, so we can say that we are 99.9% sure that the relationship exists. For the purposes of Tables 9 and 10, we described a level of significance of less than 0.01 as "highly significant", of between 0.01 and 0.05 as "moderately significant, and of between 0.05 and 0.10 as "marginally significant".

The Nagelkerke R-square value provided at the bottom of each model is a rough indication of the proportion of variation in the outcome variable explained by the predictor variables in the model. In each of the models this is between 0.2 and 0.26, which is fairly typical for this type of analysis. This means that there is a lot of variation in the data which is not explained by the variables (and nor would we expect it to be).

Both models have been tested for stability through the systematic removal of variables to check for changes in odds ratios and significance of other variables, and checks on the covariation of independent variables, and both were found to be stable. The variable for frequency of sleeping through the night was not included in the models because of its strong correlation with other variables in the model, which would have affected the overall stability.

Table D1: Logistic regression model of delayed development on at least two domains of the Ages and Stages Questionnaire
Odds Ratio Confidence interval Sig. (compared with base) Sig. (overall)
Sex of child   <0.001
Male 4.09 (2.64 - 6.34) <0.001
Female <0.001
Long term health condition (reported by either parent or keyworker) 4.81 (3.01 - 7.70) <0.001
Yes
No (+missing)
Parental long-term condition   0.359
Yes 0.76 (0.42 - 1.37) 0.359
No (+ missing)
Single / couple parent household   0.581
Single parent household 1.15 (0.70 - 1.87) 0.581
Couple parent household (+ missing)
Number of siblings < 16 in household   0.039
None / One (+ missing) 0.60 (0.37 - 0.97) 0.039
Two or more
Age of parent / respondent   0.613
20 to 29 1.39 (0.66 - 2.91) 0.385
30 to 34 1.47 (0.77 - 2.80) 0.245
35 to 39 (+ missing) 1.54 (0.80 - 2.96) 0.193
40+
Language spoken at home   0.029
Not English / dual language 2.13 (1.08 - 4.19) 0.029
English (+ missing)
Area deprivation (SIMD) of home address   0.290
40% most deprived 1.28 (0.81 - 2.01) 0.290
Less deprived (+ missing)
Equivalised household incrome   0.770
Lowest 40% 1.08 (0.66 - 1.75) 0.770
Middle / higher incomes (+ missing)
Whether respondent in work / training   0.149
No 1.44 (0.88 - 2.38) 0.149
Yes (+ missing)
Highest qualification of respondent   0.009
None / lower school (Standard Grade, etc.) 2.16 (1.32 - 3.52) 0.002
Upper school / post-school/pre-HE (Highers, HNC, etc.) (+ missing) 1.23 (0.75 - 2.02) 0.415
Degree / HE
Parental mental wellbeing (SWEMWBS)   0.099
Low (1 s.d below the mean) 1.53 (0.92 - 2.55) 0.099
Average / high (+ missing)
Feelings about amount of support from friends / family outside of household   0.303
Don't get enough / don't get any 1.34 (0.77 - 2.34) 0.303
Get enough / don't need (+ missing)
Parental self-efficacy   0.260
Coping well as a parent less often 0.74 (0.43 - 1.25) 0.260
Coping well as a parent most or all of the time (+ missing)
How many hours sleep per 24 hours   0.221
< 10 hours 1.70 (0.92 - 3.15) 0.088
10 hours 1.05 (0.65 - 1.69) 0.859
11+ hours (+ missing)
Ever breastfed   0.193
Yes (+ missing) 0.75 (0.48 - 1.16) 0.193
No
Home learning environment scale   0.934
Lowest quartile (least frequent activities) 0.98 (0.63 - 1.53) 0.934
Higher quartiles (+ missing)
Confusion, hubbub and order scale (CHAOS)   0.219
Highest tertile (most chaotic) 1.64 (0.94 - 2.88) 0.084
Middle tertile (+ missing) 1.38 (0.78 - 2.45) 0.270
Lowest tertile (least chaotic)
Total hours of childcare (formal and informal)   0.340
> 30 0.75 (0.43 - 1.33) 0.326
> 18, up to 30 0.67 (0.38 - 1.18) 0.164
Up to 18
Intercept   <0.001
0.03 (0.01 - 0.08) <0.001
n = 1301
Nagelkerke R square = 0.256  
Table D2: Logistic regression model of raised / high score on the Strengths and Difficulties Questionnaire total difficulties scale
Odds Ratio Confidence interval Sig. (compared with base) Sig. (overall)
Sex of child   <0.001
Male 2.38 (1.47 - 3.85) <0.001
Female <0.001
Long term health condition (reported by either parent or keyworker) 3.58 (2.29 - 5.61) <0.001
Yes
No (+missing)
Parental long-term condition   0.900
Yes 1.03 (0.63 - 1.69) 0.900
No (+ missing)
Single / couple parent household   0.518
Single parent household 1.22 (0.67 - 2.22) 0.518
Couple parent household (+ missing)
Number of siblings < 16 in household   0.431
None / One (+ missing) 1.24 (0.73 - 2.11) 0.431
Two or more
Age of parent / respondent   0.253
20 to 29 0.95 (0.48 - 1.85) 0.871
30 to 34 1.20 (0.74 - 1.94) 0.467
35 to 39 (+ missing) 0.72 (0.39 - 1.33) 0.298
40+
Language spoken at home   0.753
Not English / dual language 0.89 (0.42 - 1.88) 0.753
English (+ missing)
Area deprivation (SIMD) of home address   0.567
40% most deprived 0.87 (0.53 - 1.42) 0.567
Less deprived (+ missing)
Equivalised household incrome   0.124
Lowest 40% 1.52 (0.89 - 2.61) 0.124
Middle / higher incomes (+ missing)
Whether respondent in work / training   0.111
No 1.47 (0.91 - 2.38) 0.111
Yes (+ missing)
Highest qualification of respondent   0.974
None / lower school (Standard Grade, etc.) 1.06 (0.60 - 1.87) 0.854
Upper school / post-school/pre-HE (Highers, HNC, etc.) (+ missing) 1.06 (0.63 - 1.77) 0.829
Degree / HE
Parental mental wellbeing (SWEMWBS)   0.005
Low (1 s.d below the mean) 2.05 (1.25 - 3.36) 0.005
Average / high (+ missing)
Feelings about amount of support from friends / family outside of household   0.470
Don't get enough / don't get any 0.81 (0.46 - 1.43) 0.470
Get enough / don't need (+ missing)
Parental self-efficacy   0.136
Coping well as a parent less often 1.43 (0.89 - 2.28) 0.136
Coping well as a parent most or all of the time (+ missing)
How many hours sleep per 24 hours   0.316
< 10 hours 1.50 (0.84 - 2.69) 0.171
10 hours 0.91 (0.55 - 1.50) 0.695
11+ hours (+ missing)
Ever breastfed   0.266
Yes (+ missing) 0.77 (0.48 - 1.22) 0.266
No
Home learning environment scale   0.106
Lowest quartile (least frequent activities) 1.43 (0.93 - 2.21) 0.106
Higher quartiles (+ missing)
Confusion, hubbub and order scale (CHAOS)   0.295
Highest tertile (most chaotic) 1.52 (0.87 - 2.65) 0.141
Middle tertile (+ missing) 1.50 (0.85 - 2.65) 0.165
Lowest tertile (least chaotic)
Total hours of childcare (formal and informal)   0.245
> 30 1.23 (0.71 - 2.10) 0.459
> 18, up to 30 0.69 (0.40 - 1.21) 0.196
Up to 18
Intercept   <0.001
0.03 (0.01 - 0.07) <0.001
n = 1313
Nagelkerke R square = 0.214  

How to access background or source data

The data collected for this social research publication:

☐ are available in more detail through Scottish Neighbourhood Statistics

☐ are available via an alternative route

☒ may be made available on request, subject to consideration of legal and ethical factors. Please contact socialresearch@gov.scot for further information.

☐ cannot be made available by Scottish Government for further analysis as Scottish Government is not the data controller.


Contact

Email: socialresearch@scotland.gov.scot