1 - Scottish Government



Growing Up in Scotland: Father-child relationships and child socio-emotional wellbeing

Technical Annex

Alison Parkes, Julie Riddell, Daniel Wight and Katie Buston

Contents

Technical Annex 1

1 MULTIVARIABLE ANALYSIS 3

1.1 Description of analysis undertaken 3

1.2 Analytic strategy 3

1.3 Interpreting the regression results 3

1.4 Detailed results of multivariable modelling 4

1.4.1 Multivariable model of current predictors of poor father-child relationship 4

1.4.2 Multivariable model of early childhood factors associated with father-child relationship quality 6

1.4.3 Multivariable model of early and later influences on father-child relationship 8

1.4.4 Multivariable models of associations between father-child relationship and other aspects of child socio-emotional wellbeing 10

MULTIVARIABLE ANALYSIS

1 Description of analysis undertaken

Many of the factors we are interested in are related to each other as well as being related to the outcome variables of interest. An example of this from the current study is that families with an alternative father figure are more likely to contain younger mothers, parents with fewer educational qualifications and households on lower incomes. Simple analysis may identify an association between having an alternative father figure and a less supportive father-child relationship. However, this association may be occuring because of an underlying association between the presence of an alternative father figure and one or more of the socio-demographic factors, which are all indicative of family disadvantage. Thus, it may be actually be greater family disadvantage, rather than the presence of an alternative father figure, that best predicts less supportive father-child relationships.

To take these possible confounds into account, multivariable regression analysis was used. This analysis allows the examination of the relationships between an outcome variable and multiple explanatory variables whilst controlling for the inter-relationships between each of the explanatory variables. This means it is possible to identify an independent relationship between any single explanatory variable and the outcome variable; to show, for example, that there is a relationship between the presence of an alternative father figure and less supportive father-child relationships that does not simply occur because both are more commonly found among disadvantaged families.

2 Analytic strategy

Predictors of poor father-child relationships were first explored using bivariate associations (i.e. simple associations between pairs of measures). Then, in order to reduce the risk of bias and increase statistical power (compared to a complete case analysis), multiple imputation was used to impute missing information, creating 20 imputed data sets. Multivariable logistic regression models used these imputed data sets to predict the likelihood of a poor father-child relationship for any one risk factor, after controlling for other factors in the model. Models were constructed using all predictor measures that had statistically significant bivariate associations with poor father-child relationship at the p ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download