« Previous | Contents | Next »
Listen
CHAPTER 1 INTRODUCTION
1.1 About the Growing Up in Scotland study
Growing Up in Scotland ( GUS) is a major longitudinal study launched in 2005 with the aim of tracking a group of children and their families from the early years, through childhood and beyond. Funded by the Scottish Government, 1 its main aim is to provide information to support policy-making, but it is also intended to be a broader resource that can be drawn on by academics, voluntary sector organisations and other interested parties.
This report provides information from the second sweep of the survey, conducted between April 2006 and March 2007.
1.2 The GUS cohorts
GUS is based on two cohorts of children, originally sampled from Child Benefit Records, and both recruited at the same time. The first, larger, cohort is based on a sample of children born between June 2004 and May 2005 (referred to hereafter as the birth cohort); the second cohort is based on a sample of children born between June 2002 and March 2003 (referred to as the child cohort). Further information about how the cohorts were sampled and the overall sample design can be found in Appendix A.
1.3 Fieldwork at sweep 1
For the first year of the study, interviewers sought to contact the 'main carer' of the child named in the child benefit records who, in virtually all cases, proved to be the child's mother. These initial interviews were timed to take place when the child was aged either 10 months or 34 months old.
The first round of fieldwork for the study took place between April 2005 and May 2006 and results from this were published early in 2007. Response to the survey was overwhelmingly positive: among those eligible to take part in the first sweep, interviews were achieved with 81% of families in the birth cohort, yielding an achieved sample of 5,217, and with 80% of those in the child cohort, an achieved sample of 2,858.
1.4 Fieldwork at sweep 2
This report presents data from the second sweep of the survey, carried out between April 2006 and May 2007, when the cohort children were aged approximately 22 and 46 months old.
An attempt was made to recontact all those families who had participated at sweep 1 and interviews were successfully completed with 4,512 families from the birth cohort and 2,500 from the child cohort. Although there is inevitably a degree of attrition in the sample, this nevertheless represents a very respectable response rate of 88% for the birth cohort and 89% for the child cohort. As in sweep 1, the data have been weighted to adjust for non-response bias (more information on weighting is detailed in Appendix A).
As well as the main interview, at sweep 2, interviews were also undertaken with the resident partner of the main respondent. Given that in the vast majority of cases the main respondent was the child's natural mother, most of the partner interviews (97%) were conducted with the child's natural father. The partner's interview was shorter than, and used a selection of questions from, the main interview. A total of 2,975 partner's interviews were successfully completed in the birth cohort and 1,541 in the child cohort. These figures represent response rates of 80% and 78% respectively.
1.5 Content of the report
There are three main types of analysis presented in the report:
- Comparisons of the answers given by the main respondent at sweep 1 and sweep 2 (where the same questions were asked at both sweeps). This includes both straightforward comparisons of the proportions giving particular responses at each sweep, and analysis of whether the answers given by individual respondents changed between sweeps or not.
- Comparisons of the responses of main respondents with the answers given by their partners at sweep 2. 2 Again, this includes both straightforward comparisons of the proportions of main respondents and partners giving particular responses, and analysis of whether the answers given by individual couples vary or not.
- Analysis of the answers of main respondents and/or partners given at sweep 2 by factors that might help explain these answers (for example, the age or educational background of the respondent).
Although at this stage the opportunities for longitudinal analysis are still limited, some evidence of change is possible at the level of both the sample as a whole and the individual family. In Chapter 2, for example, we examine changes in family structure and household composition, while Chapter 8 explores use of childcare, including changes over the previous 12 months. But the survey also included a number of new topics and questions and these are also presented. For example, Chapter 4 details information collected on the diets and eating habits of children in the younger cohort, and Chapter 10 examines data collected on the early experiences of pre-school education amongst children in the older cohort. In addition, information is presented in Chapter 7 on the attitudes and use of different parenting techniques including specific methods of behaviour control. Data collected in the partner's interview is included in this report but usually in the context of comparison between partner data and data collected in the main interview.
1.6 Next steps
Fieldwork for the third sweep of the survey was launched in April 2007 and a high proportion of families continue to participate.
Data from each sweep of the survey will be lodged with the Economic and Social Data Service ( ESDS) Data Archive following initial publication of results by the Scottish Government, ScotCen and CRFR. There will be an ongoing programme of dissemination and utilisation associated with the study, details of which are available from the project website at www.growingupinscotland.org.uk.
Further details about the survey methods are included in the technical notes in Appendix A and can be found on the project website. Copies of the full sweep 1 and sweep 2 questionnaires are also available from the site.
1.7 A note on the interpretation and presentation of results
Only statistically significant differences (between sub-groups) are commented on in the text. This is true at the 95% confidence limit. In other words, we can be 95% certain that the difference observed is not due to chance and there is a 95% chance that the true value across all children in the sub-group (as opposed to just those in the sample) falls within this margin. 3
Within the main body of the report, most results have been rounded to whole numbers. In the tables and graphs, results have been rounded to one decimal place.
Further details on the analysis and interpretation of the results can be found in the technical notes in Appendix A.
« Previous | Contents | Next »