Annex 1: Methodology
This annex provides a detailed account of the methodology applied in this report. It starts with an overview of the Wealth and Assets Survey ( WAS) dataset used for the analysis and reasons for selecting this dataset as the most appropriate for analysis of how wealth is distributed across the population. A detailed account of the survey design and data collection stages is then given. It is followed by information on the sample of Scottish households. This section concludes with consideration of limitations with regard to WAS and the extracted sample of Scottish households.
1.1 Wealth and Assets Survey
The Wealth and Assets Survey ( WAS) is one of the main sources of information on how households in Great Britain (England, Wales and Scotland) are managing economically. The survey has a large sample size and almost complete coverage of Great Britain. The sample of private households in Scotland however excludes households north of the Caledonian Canal, and the Scottish Islands.
The survey focusses broadly on household assets and debts, borrowing and saving, and plans for retirement. Therefore, it allows for a detailed analysis of how wealth is distributed across the population.
The WAS is also a longitudinal survey, which means respondents are interviewed more than once, with about a two year interval in this case. The longitudinal element allows analysis of the levels of change in wealth across all waves from lower levels of wealth and indebtedness (or not). At the time of writing, four waves of the survey have been completed and are analysed in this report. The fifth wave is in progress. Respondents of wave one were invited to take part in a wave two follow up interview two years later. Respondents to wave two were then invited to take part in wave three and so on. In addition to these follow up interviews, a new random sample of addresses was also added to wave three.
Interviews in all waves were conducted using Computer Assisted Personal Interviewing ( CAPI). The data collected are split into two categories: household level and person level. The interviews in the main stage were conducted at the homes of respondents, while the follow up interviews took place over the phone. At the household level the interview was conducted with the household reference person ( HRP), often the head of the household or their partner; at the person level all individuals over the age of 16 and not in full-time education were interviewed separately. In each wave, each household is given a weight in order for the sample to be representative of the whole population. These weights are also assigned to all members of the same household. Estimates represented in this report were computed using these weights.
Wealth is highly skewed. Therefore median values are generally used to report central tendency.
Wealth is measured at household level. This is based on the assumption that wealth, and the economic benefit of wealth, is shared equally across all members of the household, and they benefit equally from household wealth. Analysis of characteristics in different wealth bands however is based on the household reference person.
1.2 The sample of Scottish households
The WAS provides information about the economic well-being of households living in Great Britain including England, Wales and Scotland. This report focuses on Scottish households. These were extracted from the total sample and were used for further analysis. In some cases they have been compared with the total sample of British households. The following table provides details about the time frame of each wave along with a summary of the number of households and individuals interviewed in each wave of the survey in Scotland.
|Survey Period||Number of individuals||Number of households||Number of individuals interviewed at previous wave||Number of households interviewed at previous wave|
|July 2006 - June 2008||6,199 (inc. 1,102 children)||2,833||/||/|
|July 2008 - June 2010||4,391 (inc. 760 children)||1,995||4,169||1,875|
|July 2010 - June 2012||4,540 (inc 733 children)||2,069||3,087||1,417|
|July 2012 to June 2014||4,057 (inc. 617 children)||1,890||3,040||1,420|
Methodological considerations and limitations
There are a number of methodological issues that need to be borne in mind when considering this analysis.
i. The measure of net wealth in the analysis is based on the personal, private wealth of households. It does not include business assets owned by household members, nor does it include rights to state pensions, which people accrue during their working lives and draw on in retirement.
ii. In most cases, the analysis is conducted at the household level. This means that all assets or wealth stock held by individuals living within households have been added together to produce household totals.
iii. Only those designated as 'private households' were sampled for the survey. This means that people in residential institutions, such as retirement homes, nursing homes, prisons, barracks or university halls of residence, and also homeless people, are excluded from the scope of the analysis presented in this chapter.
iv. The Scottish sample is not fully representative of the geography of Scotland, as WAS does not sample households north of the Caledonian Canal or on the Scottish islands.
v. Estimates within the analysis have not been equivalised and therefore are not intended to account for differences in household size or composition.
vi. All estimates within this report are presented as current values at the time of interview, and have not been adjusted for inflation.
vii. In Wave 1, data on physical wealth was only collected for approximately half of the sample. In order to split the population into deciles, only those that have values for all wealth types are used. Analysis based on deciles may therefore not match the total values for the population but are intended instead to represent the distribution of wealth within the population.
viii. Figures and percentages presented in this report have been rounded, and so discrepancies may occur between sums of component items and the totals. This will also occur where variables allow more than one response to be reported. Published percentages were calculated prior to rounding and therefore discrepancies may occur between reported percentages and those that could be calculated from the rounded figures.
It is also important to note that methodological changes have been introduced with the publication of the wave 4 data. Changes to the valuation of pension wealth have been introduced, with the methodology being applied retrospectively. Additionally there have been changes in the financial parameters used to value defined benefit pensions in wave 4 only (see pension Wealth section for further details). For this reason, the analysis in this report differs from that published in previous reports. Any comparison with previously published data is not appropriate. This report uses the latest data for all waves of the survey to present consistent time series.
Email: Andrew White