Financial capability is captured in the Wealth and Assets Survey against 6 dimensions.
These 6 dimensions of financial capability relate to making ends meet, planning ahead, organised money management, controlled spending, staying informed and choosing products.
Scored on a scale from 0 to 10 individuals performed best at making ends meet (7.0) and least well at planning ahead (2.3).
It was unusual for individuals to score highly, relative to others, on all 6 financial capability dimensions (1%), and more than 1 in 5 (22%) did not score highly on any.
Levels of financial capability were generally higher among older individuals, those living in a couple without children, and those from higher socio-economic groups, although this differed by dimension.
Levels of financial capability were higher for individuals living in households with the highest levels of total wealth and this was especially marked in relation to financial wealth. Individuals living in the 2 lowest wealth quintiles also scored highly at organised money management.
Financial capability describes the skills, knowledge and behaviours individuals need to make informed decisions and take positive action about their finances. The Money Advice Service, which has responsibility for developing a new Financial Capability Strategy for the UK, defines financial capability as the interaction between internal factors such as ability, attitudes and motivations and external factors, such as the accessibility of financial services. The following article uses data from the third wave of the Wealth and Assets Survey (relating to 2010 to 2012) to define and measure financial capability among adults living in Great Britain. It is based on all adults who responded to the survey in person.Back to table of contents
In the third wave of the Wealth and Assets Survey, new questions designed to measure financial capability were included for the first time. These were based on the first national measure of financial capability in the UK, which was undertaken in 2005, for the then Financial Services Authority1. The new questions supplemented questions already asked elsewhere in the survey that directly or indirectly measured aspects of financial capability.
A total of 27 measures were analysed to identify underlying dimensions of financial capability captured by the survey. These comprised individual survey questions and measures derived from one or more survey questions, and included attitudinal questions, self-reported behaviours and objective measures. The measures considered had a strong focus on the behaviours underpinning financial capability, as opposed to knowledge or skills. This reflects the findings from the formative research to the first national measure, which found that focussing on behaviours ensured the relevance of financial capability dimensions to the widest cross-section of people, including by income.
Of these measures, 18 mapped onto 6 emerging financial capability dimensions2. The remaining measures were unrelated to these dimensions and excluded from subsequent analysis. The resulting dimensions were interpreted from 2 previous studies that measured financial capability using data from other surveys3. The 6 dimensions are:
Making ends meet:
A person's ability to live within their means. Someone who is highly capable on this dimension can make their money last until the end of the month, is keeping up with the payments on household bills and any credit commitments and has more money in savings than they owe in consumer credit.
The extent to which someone makes provision for future expenditure from current income. Someone who is highly capable on this dimension saves up for major planned expenditure and makes sure they have money saved for a “rainy day”.
Organised money management:
The extent to which someone knows how much they spend day to day. A highly capable person on this dimension knows, to within 1 or 2 pounds, their bank balance or the cash they have available for everyday spending.
A person's attitudinal preference for longer-term financial security over current spending capacity. A highly capable person on this dimension is someone who prefers to protect a good standard of living for the future rather than prioritise spending for today.
The extent to which someone keeps up to date with changes in the wider economy. Someone who is capable on this dimension keeps a close eye on several sources of economic (like inflation), monetary (such as interest rates), and fiscal (such as taxation) conditions.
The sources of information (if any) someone uses when buying a financial product that most influence their purchase decision. This dimension comprised a single survey question which was only asked of survey respondents who had bought a financial product within the last 2 years. A highly capable person will have shopped around, consulted best buy information or used a comparison website. The strongest survey measures for each dimension are shown in Table 14.
Table 1: Strong indicators of financial capability dimensions
- The loading indicates the strength of correlation between the measure and the component.
- Loadings were not produced for choosing products, as this dimension was based on a single survey question.
- Loadings are rounded to three decimal places.
Some of these indicators relate to single survey questions. Examples include, how well you are keeping up with bills and credit commitments and that you feel you are organised when it comes to managing money. Other indicators comprise of 2 or more closely related survey questions. For example, your time preference for spending is derived from 3 questions about attitude: if you tend to live for today, find it more satisfying to spend money than save for the long term and would rather have a good standard of living today than save for retirement. How accurately you know how much spending money you have derives from 2 related questions: the bank balance on your day-to-day bank or building society account and, if you don’t have an account, how much money you have, excluding any savings.
Notes for defining financial capability in the Wealth and Assets Survey
Atkinson, A. McKay, S. Kempson, E. and Collard, S (2006) ‘Levels of Financial Capability in the UK: Results of a baseline survey’ Consumer Research 47. London: Financial Services Authority.
The type of analysis used to produce the dimensions was principal components analysis, except for choosing products, which was based on a single survey question. The methodological Appendix has more information.
Ibid. and Hayes, D. Collard, S. and Kempson, E (2015) ‘Understanding the profile of those most at risk of detriment as a result of low financial capability: Analysis of the Money Advice Service Financial Capability Survey’ London: Money Advice Service
A maximum of the 3 strongest indicators is shown for each dimension. All indicators are shown in the accompanying Excel table (235.5 Kb Excel sheet).
An individual’s financial capability for the 6 dimensions is measured on a scale from 0 to 10. Among adults in Great Britain in 2010 to 2012, the average (mean) capability scores ranged from a low of 2.3 for planning ahead to 7.0 for making ends meet. Staying informed also carried a relatively low average score of 3.2, while the remaining dimensions were in the middle of the range.
Table 2 shows the overall mean scores for each dimension, along with the mean score among the quarter of adults with the highest scores, the quarter with the lowest scores and those in between. The mean scores for each quartile vary considerably between dimensions due to the different distributions of scores across the population for each dimension. Very few people scored very poorly on organised money management, for example, while a large proportion of people scored zero on staying informed. This means that each category is high or low relative to others on for this dimension, rather than to any objective, external measure or to scores on other financial capability dimensions.
Table 2: Mean financial capability scores by dimension: Great Britain 2010 to 2012
|Making ends meet||7.0||Low||4.2|
|Organised Money management||6.7||Low||3.9|
|Source: Wealth and Assets Survey – Office for National Statistics|
|1. Means are rounded to one decimal place|
|2. <0.1 indicates a score of greater than zero but less than 0.05|
Download this table.xls
Levels of financial capability by socio-demographic characteristics
Sex and age
Figures 1 and 1a shows how the financial capability score for each of the 6 dimensions varies by sex and age. While the variations between men and women and across the age groups appear small, they are statistically significant in all cases. As such, we find that men score slightly better at making ends meet, planning ahead, and choosing products. They also score more highly than women on the staying informed dimension. Conversely, women score slightly better than men at organised money management and controlled spending.
When other characteristics are held constant (using a technique known as regression analysis)1, the variations by sex remain evident (235.5 Kb Excel sheet). All other things equal, men score relatively well compared with women at making ends meet, staying informed and choosing products, while women score significantly better at organised money management and controlled spending. When other factors are controlled, women also score marginally, but statistically significantly, better than men at planning ahead; this diverges from what we saw above in the simple breakdown. While statistically significant, the effect of sex is generally small. The exception is staying informed, for which, all other things being equal, men score better by approximately three-quarters of a point.
The variation by age is somewhat more striking (Figure 1a). Capability on the making ends meet and controlled spending dimensions increases steadily in the population of Great Britain with increasing age.
For planning ahead, staying informed and choosing products, 16 to 24 year olds score particularly poorly. As such, we observe a fairly marked increase at age 25 to 34. This corresponds with a tendency to move towards independent living and increasing financial independence in young adulthood. This increase is followed by a relatively small improvement with each increasing age group in the planning ahead dimension. Conversely the choosing product dimension shows a steady decline with age while the staying informed dimension has no pattern and remains volatile though the age groups. While lower financial capability in choosing products among older people indicates a greater reliance on single sources of information (possibly due to lower internet access2), financial advisers or even direct marketing, this may in turn reflect greater experience or financial confidence.
Organised money management is the only dimension where capability remains relatively flat across the age groups, although there is modest increase between people aged 16 to 24 and 25 to 34. The comparatively subdued variation observed here in relation to organised money management is repeated for the breakdowns by most other characteristics, as later sections show.
When other characteristics are held constant in regression analysis, the influence of age holds true for most of the dimensions (235.5 Kb Excel sheet). In particular, the steady increase with age in capability on the controlled spending dimension, from low capability among 16 to 24 year olds, remains clear. Meanwhile, the rather lower choosing products score and high levels of capability at making ends meet observed among the 65s and over are clearer still when other factors are controlled. However, there are 2 notable divergences. In relation to making ends meet, scores are slightly (albeit significantly) lower among people aged 25 to 44 than the youngest age group. And planning ahead scores decline among the 45 to 64 year olds. This suggests that factors other than age explain the earlier steady improvement observed in relation to these dimensions. These might include household type and income, which tend to vary with age3.
Figure 2 shows how levels of financial capability vary by the highest education level someone has achieved. Marked variations are found for 3 dimensions: planning ahead, staying informed and choosing products. For each of these, those without qualifications scored least well, with financial capability increasing steadily for each group with higher qualifications. Those with degrees also score well at making ends meet compared with those without. This may in part reflect the greater earning power of people with degrees and therefore the greater resources they have available4. Financial capability is explored further by economic activity, socio-economic group and household income.
The patterns by education level are confirmed in further analysis that takes other characteristics into account (235.5 Kb Excel sheet). The higher levels of capability among those with some qualification compared with those with none are particularly marked in relation to staying informed and choosing products. In turn, the additional advantage of those with degrees is especially notable in relation to the staying informed dimension; all other things being equal, people with degrees scored significantly higher on this dimension when compared to those with no qualifications and other qualifications.
Figure 3 shows moderate variation in financial capability by the type of household someone lives in. Single and couple households without children tend to be towards the high end of the range on each dimension. The exception is choosing products, for which single-adult households score less well than other groups. Couples with dependent children score particularly well at staying informed and choosing products. At the other end of the range, lone parents with dependent children scored less well on the whole. This is particularly noticeable in relation to making ends meet relative to other households, possibly indicating that these households are likely to have high (unavoidable) demands on their incomes, and this in turn is reflected in a low planning ahead score. The one exception is organised money management, for which lone-parent households are at the higher end of the range. This indicates that they were especially likely to be keeping an eye on the money they had available.
When other characteristics are held constant in regression analysis, these patterns largely hold true (235.5 Kb Excel sheet) . However, there are some small differences. In particular, the variation for organised money management is more marked when other factors are taken into account, with couples with children more clearly scoring poorly on this dimension. Living in a single adult household is associated with higher capability at making ends meet and planning ahead than all other types of living arrangement, independently of other factors. This may reflect, respectively, the fewer demands on disposable income and fewer complications for future finances in single-adult households. Finally, all other things being equal, people living in households containing more than one family stand apart as controlled spenders, their household arrangements perhaps reflecting a need to spend carefully.
Figure 4 shows variations in financial capability according to the location of the main residence of the household in which an individual lives. It shows Scotland, Wales and the 9 English regions (with London shown separately; the figures for the South East exclude London). The variation in capability by location is clearest in relation to the staying informed dimension. For this dimension, people living in the South East of England are at the high end of the range, while those in Scotland are at the low end followed by those in the North East of England. The pattern is not consistent across the dimensions, for example, in relation to making ends meet, people living in Scotland (along with those in the South East and South West of England) score towards the high end of the range, with those in the North West of England and London at the low end. Meanwhile, those in the East Midlands score at the high end of the range in relation to choosing products, a dimension on which people living in Wales score comparatively poorly.
Mean financial capability scores, by region: Great Britain 2010 to 2012
It is possible that the different socio-demographic and economic profiles of the different regions and countries of Great Britain will partly explain these variations. However, after taking a range of other characteristics into account in regression analysis, the general pattern of levels of financial capability holds true (235.5 Kb Excel sheet). For example, people living in Scotland continue to be at the low end of the range in relation to organised money management and in particular staying informed (for which people living in Scotland score about three-quarters of a point less than their counterparts living in the East Midlands). Living in London is independently associated with a low making ends meet score (possibly reflecting the higher cost of living in London5) and living in Wales shows a low choosing products score. Finally, all other things being equal, people living in the West Midlands continue to score significantly better than other parts of the country at organised money management.
However, some differences are apparent, in particular living in the East of England is independently associated with scoring relatively poorly at staying informed; this was less clear before controlling for other factors.
Notes for levels of financial capability in the general population
In addition to the characteristics reported here, the regression analysis (235.5 Kb Excel sheet) included a range of other measures.
Office for National Statistics (2014) Internet Access – Households and Individuals 2014. Statistical Bulletin. Newport: Office for National Statistics.
For example, Table A38, Horsfield, G (2011) Family Spending: A report on the 2010 Living Costs and Food Survey. Newport: ONS
Walker, I., & Zhu, Y. (2013) The impact of university degrees on the lifecycle of earnings: some further analysis. Department for Business Innovation and Skills.
Padley, M. Marshall, L. Hirsch, D. Davis, A. and Valadez, A (2015) A minimum income standard for London. London: Trust for London.
Figure 5 shows how individuals’ financial capability varies depending on the level of their household’s income. For this analysis, income was divided into quintiles: 5 equal groups describing the lowest 20% of individuals by household income up to the highest 20%. There is a clear relationship between capability and income on 4 of the 6 financial capability dimensions: making ends meet, planning ahead, staying informed and choosing products. For these dimensions, higher incomes are associated with higher capability scores. There is relatively little variation by income in relation to organised money management and controlled spending.
This general pattern of variation remains evident after other characteristics are held constant using regression analysis (235.5 Kb Excel sheet). If anything, the higher scores at making ends meet, planning ahead and choosing products for people living in higher income households are evidenced more clearly. The variation in capability at organised money management and controlled spending is also clearer, with lower incomes associated with higher levels of capability on these dimensions, most likely driven by the need to watch their money more closely than others. People living in the lowest income households perform slightly, if significantly, less well at staying informed, all other things being equal, with the differences between the remaining groups on this dimension being marginal.
Figure 6 shows a very mixed pattern of financial capability based on a person’s economic activity status. People who have retired from paid work stand out as particularly financially capable, relative to others, on 2 dimensions: making ends meet (for which, at 8.0, they score a point higher than the average for the population as a whole); and controlled spending.
Retirees also perform comparatively well at planning ahead. Their capability on these dimensions may reflect a particular need for retirees to manage on fixed incomes, combined with greater experience of financial management (given that retirees will tend to be older). Notably, the unemployed score comparatively poorly at making ends meet and planning ahead, most likely reflecting the low incomes of the unemployed relative to others.
People in paid work score highly relative to other groups on the choosing products dimension, although the differences are not great. Notably, the self-employed are the most capable of all the economic activity status groups on the staying informed dimension which may reflect a greater direct influence of these factors and the fewer protections from them and, therefore, a need for heightened awareness.
When exploring the independent influence of economic activity on levels of financial capability, using regression analysis, it was necessary to combine the self-employed with employees, and retirees with those who were inactive for some other reason. In this analysis (235.5 Kb Excel sheet) the unemployed continue to score less well than other groups at making ends meet and controlled spending, in addition to scoring poorly at organised money management (although the differences between the groups are not large). All other things equal, people defined as unemployed perform very well at staying informed, scoring some 0.8 points higher than those who were not working for any other reason. This may reflect a greater relevance of wider economic factors for people who are actively looking to enter or re-enter the jobs market.
When examining financial capability by the socio-economic group someone belongs to, we find that higher socio-economic groups are generally associated with higher capability, with lower socio-economic groups tending to score less well (Figure 7). The variation is most marked in the 2 lowest scoring dimensions, planning ahead and staying informed, where people in higher managerial, administrative and professional occupations scored more than twice as highly as those who have never worked or are long-term unemployed.
The variation is again muted for organised money management and controlled spending, with those who have never worked or are long-term unemployed at the low end of the range.
This pattern holds true in regression analysis (235.5 Kb Excel sheet). Notably, working in semi-routine or routine occupations, or being classed as long-term unemployed or having never worked, is associated with 0.6 fewer points at staying informed compared with people from higher managerial, administrative and professional occupations, even after taking other factors into account. Having never worked or being long-term unemployed is associated with half a point lower score on the organised money management dimension than the 2 highest socio-economic groups.
Where the pattern does diverge from the one described above, it relates to small employers and own account workers. All other things equal, this group scores relatively poorly at making ends meet and relatively well at choosing products, for which they score better than people from higher managerial, administrative and professional occupations. These findings may reflect, respectively, the less regular income often associated with self-employment and greater importance of making the right financial product choices within this context.Back to table of contents
Figure 8 shows how an individual’s financial capability scores vary by the amount of wealth their household holds. Total wealth and its 4 components are considered: net financial wealth (any financial assets, such as savings, less any financial liabilities); total private pension wealth (comprising occupational and personal pension wealth); wealth from physical assets (such as household contents, vehicles and other valuables); and net property wealth (property, such as housing assets, less any property liabilities).
Financial capability tends to be higher where household wealth is higher, especially in relation to making ends meet, planning ahead, staying informed and choosing products. This variation is particularly marked by net financial wealth in relation to making ends meet and planning ahead. Additionally, there is notable variation in controlled spending capability for financial wealth and total wealth, and to a lesser extent pension and property wealth. Here again, higher wealth tends to be associated with higher capability.
There are exceptions, however. People living in the least wealthy households by financial wealth score comparatively well compared to those in wealthier households on the choosing products dimension. People living in the second-least wealthy households by property wealth also score particularly well on this dimension. This is likely to reflect the limited population on which this dimension is based, such that only the most financially capable of people living in less wealthy households are likely to have actively reviewed and changed their financial products in the previous 2 years. People who live in the second-least wealthy households by financial wealth score poorly, however, at staying informed.
Regression analyses (235.5 Kb Excel sheet) show a particularly marked increase in financial capability on 3 dimensions – planning ahead, staying informed and choosing products – between people living in households with the 2 lowest pension wealth quintiles and those with higher pension wealth. In relation to controlled spending, the picture by pension and physical wealth is reversed, such that with increasing wealth capability tends to decline, perhaps because the need to control spending is diminished in these instances. People living in households with the second highest wealth of these types score particularly well at choosing products, all other things equal.
Mean financial capability scores, by quintile of household wealth: Great Britain 2010 to 2012
Otherwise, the patterns described above broadly hold true when other factors are controlled. Of particular note is the strength of the independent influence of a household’s financial wealth on someone ‘s financial capability, especially in relation to making ends meet and planning ahead. All other things equal, living in a household in the highest financial wealth quintile is associated with 2 extra points at making ends meet and 1.5 extra points at planning ahead, compared with those living in the lowest-wealth households.
When other characteristics are controlled, the variation on the organised money management dimension by household pension and property wealth is also more clearly defined, although the absolute impact on financial capability scores remains relatively small (less than 0.2 points difference across the range).Back to table of contents
The process undertaken to define the financial capability dimensions captured in the survey is intended to identify more or less discrete dimensions. Nonetheless, these dimensions are likely to co-vary (or ‘correlate’) with each other to a greater or lesser extent. For example, someone with high levels of capability on one dimension may be somewhat more likely to have high levels of financial capability on another. Analysis confirms that the correlation between each pair of dimensions is positive, such that as financial capability scores on one dimension increase they also increase on another. Table 3 shows the size of the correlations, where 0 would indicate that the 2 dimensions are not at all correlated and 1 indicates a perfect correlation.
As we should expect, the correlations between dimensions are generally weak. Only one pair of dimensions has a correlation which exceeds 0.5: making ends meet and planning ahead. This suggests that these 2 dimensions are influenced by an underlying factor. Given the composition of these dimensions, this might be interpreted as individuals’ financial liquidity or, perhaps, their wealth status more generally. In other words, we need to take into consideration that financial capability on the first 2 dimensions may be limited or facilitated by the underlying financial resources individuals have available. This is supported by the findings of the regression analyses, in which capability on these 2 dimensions varied strongly with household income, and especially financial wealth.
Table 3: Pairwise correlations between financial capability dimensions
|Pearson’s correlation coefficient ( r )||Making ends meet||Planning ahead||Organised money management||Controlled spending||Staying informed||Choosing products|
|Making ends meet||1||0.63||0.03||0.36||0.19||0.03|
|Organised money management||0.03||0.11||1||0.41||0.19||0.06|
|Source: Wealth and Assets Survey – Office for National Statistics|
|1. Correlations are rounded to two decimal places|
|1. All correlations are statistically significant (p<.05)|
Download this table.xls
Capability across multiple dimensions
Looking across all of the dimensions simultaneously, it was very unusual for someone to score highly relative to others on all dimensions. Figure 9 shows that, for the quarter of adults with the highest scores (defined in Table 2), excluding choosing products, only 1% score highly on the remaining 5 dimensions. It also shows that 2 in 5 (42%) do not score highly on any and a further 28% only score highly on 1 dimension.
When including choosing products, and limiting our analysis to those who had bought a product in the last 2 years, less than 1% score highly on all dimensions, although only 22% don’t score highly on any. This suggests that people who had bought a financial product in the last 2 years are more financially capable on the whole than people who had not. A further 30% are highly capable on 1 dimension and 1 in 10 are highly capable, compared with others, on 4 or more dimensions.
Figure 9 also shows the number of dimensions on which individuals have low financial capability scores relative to other people. A significant minority of people (3 in 10) don’t have a low score on any of the dimensions. However, nearly 1 in 10 people have low financial capability on 4 or more dimensions, regardless of whether choosing products were included or excluded. These individuals are likely to be at most risk of detriment as a result of low financial capability.
Characteristics of people with low capability on multiple dimensions
Regression analysis (235.5 Kb Excel sheet) explores the characteristics that independently relate to being at most risk of detriment as a result of low financial capability (defined as having low financial capability on 4 or more dimensions). All other things being equal, men are slightly more likely to score poorly on multiple dimensions than women, while young adults are substantially more likely than older age groups; it is unclear whether the latter is an ageing or a generation effect.
People living in couples with children (dependent or non-dependent) and lone parents with dependent children are towards the high end of the range. Where someone lives in Great Britain also makes a difference independently of other factors, with people in the East of England, Scotland, Yorkshire and Humberside and Wales at the high end of the range, the latter partly reflecting particularly low average capability at choosing products among people living in Wales. This is particularly interesting as these differences are not explained by the socio-demographic and economic profiles of the population of these regions and countries, but some other factors that are otherwise not captured in the analysis.
Being without educational qualifications, and to a lesser extent having other qualifications compared with a degree-level qualification or above, also significantly increases the likelihood that someone scores in the lowest quartile for 4 or more dimensions, independently of other factors. While someone’s economic activity status is not significant in this analysis, certain socio-economic groups, having semi-routine and routine occupations, tend to be at the high end of the range, followed by those who had never worked or were long-term unemployed and small employers and account workers.
Household income itself is not a significant factor in this analysis. Scoring poorly, relative to others, on multiple dimensions appears to relate more to an individual’s household wealth. This is especially true in relation to their financial wealth (for which there is a steady decrease in the odds of scoring poorly with increasing wealth) and, to a lesser extent, to their pension wealth.
Overall, someone’s age and the financial wealth of their household appear to be the strongest determinants of capability across the dimensions. Independently of other factors, younger adults and people with access to the least financial wealth in their household are likely to be at greatest risk of detriment as a result of low financial capability across the dimensions. It is not clear, however, to what extent wealth influences financial capability or that, conversely, financial capability influences wealth.Back to table of contents
The starting point for this analysis was to identify all survey questions with a potential link to financial capability. This was informed by the 2 previous studies of financial capability (FSA, 2006; MAS, 2015) and through a comprehensive review of the wave 3 questionnaire. This produced a long list of approximately 70 potential survey questions and derived measures to use in the analysis. The vast majority were drawn from the person file, with only a few using household-level measures.
The long list was subsequently reviewed in greater detail for suitability for use in principal components analysis. This took into account their potential relevance to financial capability, potential overlap with other similar variables and their routing and sample coverage within the questionnaire. In particular, only those measures that were asked of all adults, or which could be rebased to adequately represent all adults, could be put forward for the analysis and only those with some bearing on previously identified dimensions were retained. At this stage 43 measures were excluded. Due to the reliance on a number of attitudinal questions it was necessary to limit the subsequent analysis to all adults interviewed in person at wave 3.
The remaining 27 measures were entered into an initial principal components analysis. This was deliberately inclusive of all measures (subject to the limitations described above) in order to test our assumptions about how certain measures were likely to behave. This initial analysis confirmed the need to exclude further variables that did not load onto meaningful factors and these were removed sequentially from subsequent refinements of the initial model. One such variable included the single measure (based on question Sour2p in the questionnaire), which has become the choosing products dimension; others include measures (such as ORisks) for which it is unclear what they intend to measure or which are better treated as covariates (such as measures based on OTrust and IncDrop).
This initial principal components analysis also identified particularly high correlations between 3 financial acuity attitudinal measures (SpendM, GStdNw and LvTDay), which were combined into a single measure (representing ‘time preference’) using a separate principal components analysis. The option of using 1 of these three attitudinal variables to represent all 3 was explored. However, considerable detail and sensitivity was lost in doing so, and the composite measure produced more meaningful dimensions in later refinements.
A final all-measure principal components analysis was subsequently run to verify the remaining dimensions. This analysis produced very high-value diagnostic statistics (KMO = .87; Bartlett's Test p<.001) indicating a highly robust solution and indicated 5 significant components. A further component, which was not significant, appeared to capture financial difficulty (led by the total amount an individual’s household owed in bill arrears). This component was excluded on the basis that it was not significant, resulting in the loss of only 1 further measure. The 5 remaining components were refined in single-component solutions to produce the final solutions reported here. Together with the choosing products measure they make up the 6 dimensions and are based on 18 measures (including composite measures) drawn from 21 survey questions or ONS-derived variables.
Principal components analysis produces scores with a mean value of 0. Scores on all dimensions were re-scaled to take a value ranging from 0 to 10.Back to table of contents
Contact details for this Article
Telephone: +44 (0)1633 456295