Socioeconomic status

Attempts to identify the social class of adolescents based on parental occupation and education has in the past been hampered by large amounts of missing and incomplete data (Currie et al 1997). In the absence o f well established and reliable measures for adolescents, several different approaches were made to the measurement of socio-economic background:

Chapter 4. Baseline methodology and descriptive statistics

4.2.2.1 Parental Occupation

Participants were asked to describe the occupation of the parents that they lived with in order to allow coding using the Registrar General's social classification (Office of Population Censuses and Surveys, 1991) scheme using the questions:

'Does your father (or a stepfather) live with you most o f the time?' If Yes: 'Does your father (or stepfather i f you live with him) have a paid jo b ? ' If Yes 'What is

his job?'. Questions were repeated for mother's job.

4.2.2.2 Parental Education

This was assessed with a single question asking 'Did your parents go to college or university after leaving school?' Response options were: Yes both o f them, Yes, one o f them. No and Don't know.

4.2.2.3 Neighbourhood tvpe

Postcodes were collected in order to make use o f a small area measure of deprivation. Information from the 1991 census allows classification of neighbourhoods at the enumeration district level (approximately 20 households) in terms of levels of unemployment, overcrowding, lack o f access to a car and rented accommodation (Townsend Index: Townsend et al 1988).

4.2.2.4 Home affluence Scale (H A S P

A home affluence scale was developed as an alternative approach to the measurement of social status. Participants were asked about family ownership of one or more vehicles {'Do you have a car or van at home? Yes more than one car or van/Yes one car or van/No we do not own a car or van), a computer {(Do you have a home computer at home? (Do not include playstations or other computers that can only be used fo r games). Yes/No) home ownership {'Thinking about the house you live in at the moment, do your parents own it or rent it? ( i f they have a mortgage, tick they own it). They own it/They rent it/Don't know')

and eligibility to receive free school meals ('Do you yourself have the option o f free lunchtime meals at school? Yes/No). The four material indicators were then

Chapter 4. Baseline methodology and descriptive statistics

each of the following: the family owning a car, owning 2 cars, the family owning their home, the family owning a computer, and the pupil not having the option of free school meals. This generated a scale with possible scores ranging from 0-5. The affluence scale had the advantage of being a household-level measure which asked questions to which adolescents were likely to know the answers, and completion rates were expected to be higher than those for education and occupational data.

Validation o f the HASC The HASC was validated against the other measures of socio-economic status. Details of the validation o f this scale can also be seen in Wardle, Robb and Johnson (in press). Analyses were carried out to examine the completion rates, levels of socioeconomic bias in completion, internal reliability and external validity of the HASC. Table 4.4 shows the completion rates for each of the questions in the HASC, parental occupation and education and postcode. The lowest completion rates were for the questions on parental employment. 22% o f students did not live with a father or stepfather, so paternal information was not completed. 6% did not live with a mother or stepmother. An additional 10% of fathers and 25% of mothers were not in paid employment at the time. A further 12% o f students gave answers to the paternal occupation question that provided insufficient information to code paternal social class (e.g. “He works at Cadbury’s”) or failed to give a description. Similarly 13% of information on mothers’ occupation was uncodeable or missing. The education questions were better completed, but still, 24% o f students responded that they didn’t know whether their parents had a college education. Postcodes, used to derive area-level data, were reported by almost 90% of students. The material indicators were all completed by more than 90% o f the students, with car and computer ownership having completion rates above 99%.

In order to examine the question of differential completion in relation to SES, completion rates were examined at the school level to see whether they were higher in schools with a more affluent student population. Less affluent schools have significantly poorer completion rates for many o f the items (See table 4.4).

Chapter 4. Baseline methodology and descriptive statistics

There was a gradient for parental education, paternal and maternal occupation and postcode. Housing tenure completion was also linked with school type, and free school meals showed an inverse gradient, such that girls from the more affluent schools were less likely to know whether they were eligible or not. The other material measures, car ownership, and computers, showed no significant gradient across the three school types.

Table 4.4 Item completion rates by school type

SES level o f school

High Middle Low Total

Linear trend for school type

Mother's occupation codeable 66.0 5&8 48.3 57.7 =24.9 p=0.00 Father's occupation codeable 69.8 55.8 48.1 57.9 _{x '[i =37.8 p=0.00} Parents’ college attendance 85^ 74.5 67.2 75.8 =36.9 p=0.00

Postcode 99.0 92.2 83.7 91.6 =59.0 p=0.00

Housing tenure 94.6 93.7 8&0 92.1 =11.5p=0.00

Car ownership 99.5 99.2 99.5 99.4 _{x '[i =0.00 p=0.99}

Free School meals 97.2 98.7 98.9 98.3 =3.8 p=0.05

Computer ownership 99.5 99.5 99.5 99.5 =0.00 p=0.99

Inter-correlations between the items were all modest but statistically significant (see table 4.5). The cronbach’s alpha of the final scale was moderate (a = .53). The three groups with the lowest affluence scores were combined to produce a more even distribution with roughly equal numbers o f participants in each group.

Table 4.5. Correlations between individual material indicators and with the total score (all significant at p<0.01)

Tenure Computer FSM Total HASC

Car (0,1,2) 0.26 0.32 0.27 0.75

Tenure (0,1) - 0.15 0.27 0.56

Computer (0,1) - 0.16 0.56

Chapter 4. Baseline methodology and descriptive statistics

In order to assess the external validity o f the scale, we examined associations with occupational social class and parental education where these were available. HASC scores were modestly correlated with paternal social class (rs=.34, p<.001), maternal social class (rs=.30, p<.001), and parental education (rs=.27, p<.001). External validity was also evaluated by testing whether children whose postcodes indicated that they came from an area o f greater socio­ economic deprivation reported lower levels o f affluence. Townsend Index scores for the area of residence show a strong association with affluence scores, with a correlation of r=-.42, p < .01 between the HASC score and the Townsend Index.