Last Updated on November 2, 2026 by Karl Thompson
Why do children from poorer families do worse in education than children from richer families? This post looks at one of the best-known attempts to answer that question using longitudinal data: a study for the Joseph Rowntree Foundation, edited by Alissa Goodman and Paul Gregg and published in March 2010 as Poorer children’s educational attainment: how important are attitudes and behaviour?
It’s useful in two ways. For the education topic, it’s evidence on why the class gap opens and widens. For research methods, it’s a clear example of quantitative, longitudinal research, which makes it a good Methods in Context example for AQA Paper 1 (7192/1).
The study and its methods
The research team (Haroon Chowdry, Claire Crawford, Lorraine Dearden, Robert Joyce, Luke Sibieta, Kathy Sylva and Elizabeth Washbrook) analysed four large longitudinal datasets, each following children over time:
- the Millennium Cohort Study, covering children across the UK
- the Avon Longitudinal Study of Parents and Children, covering children in the Avon area
- the Longitudinal Study of Young People in England, covering teenagers
- the children of members of the 1970 British Cohort Study
A longitudinal study follows the same people over a period of time, measuring them repeatedly. That makes it possible to see when gaps between groups open up and what changes alongside them, which a one-off snapshot can’t show. The study compared the poorest fifth of families with better-off families, and used statistical analysis to estimate which factors were linked to the gap at each stage. It’s very much in the quantitative tradition associated with positivism, though the researchers were careful about claiming cause and effect.
The main findings
The gap starts early and widens. Even by the age of three, there was a considerable gap in cognitive test scores between children in the poorest fifth of families and those from better-off backgrounds. The gap widened as children moved through school, especially during the primary years.
The researchers then looked at which attitudes, behaviours and resources were linked to the gap at each stage:
| Stage | Factors linked to the gap between poorer and better-off children |
|---|---|
| Early years (age 3) | The home learning environment, such as how often parents read to the child. But a much larger share of the gap was left unexplained by the early caring environment, or was linked to other family background factors such as the mother’s age and family size. |
| Primary school | Parents’ aspirations for higher education: at age nine, 81% of the richest mothers hoped their child would go to university, compared with 37% of the poorest mothers. How far parents and children believe their own actions shape what happens to them. Children’s behavioural problems. |
| Secondary school | Parents’ and young people’s expectations of higher education; material resources such as private tuition and access to a computer and the internet; young people’s belief in their own ability at school; whether they found school worthwhile; risky behaviours such as truancy and anti-social behaviour; experience of bullying. |
Source: Goodman and Gregg (eds) (2010), as summarised by the Joseph Rowntree Foundation.
One finding worth remembering: the researchers concluded that policies to improve parenting skills and home learning environments cannot, in isolation, close the cognitive gap between rich and poor young children. Early intervention matters, but it isn’t enough on its own.
Feinstein: an earlier longitudinal study
Leon Feinstein (2003) used the 1970 British Cohort Study, testing children’s development at 22 months, 42 months, 5 and 10. He found clear gaps by family background as early as 22 months. More strikingly, children from poorer backgrounds who scored highly as toddlers tended to fall back, while lower-scoring children from better-off families caught up.
HOWEVER, Jerrim and Vignoles (2013) argued that the most dramatic part of this finding may be partly a statistical artefact called regression to the mean: children who score unusually high or low on one test tend to score closer to the average next time, whatever their background. What survives the debate is the broader point, which the 2010 study supports: gaps by family background appear very early and tend to grow. There’s more on Feinstein in my post on material deprivation and education.
What the study tells us about class differences in achievement
The study’s main message is that home background matters a great deal, and that poverty works partly through attitudes, behaviour and the home learning environment. That makes it useful for showing how material and cultural factors connect: lower aspirations and less reading at home are linked to lower income, rather than being free-floating features of working-class culture.
HOWEVER, a cultural deprivation reading of the aspirations finding should be handled carefully. Lower hopes for university among poorer mothers may reflect a realistic view of the obstacles their children face rather than a lack of interest in education. And the study’s focus on home factors says little about what happens inside schools; see in-school factors for that side of the debate.
Evaluating the study as a piece of research
| Strengths | Limitations |
|---|---|
| Large, representative samples. National cohort studies follow thousands of children, so findings can be generalised with more confidence than a small qualitative study. | Correlation, not causation. Statistical analysis shows which factors go together with the gap, not necessarily what causes it. Unmeasured factors may be doing the work. |
| Change over time. Following the same children shows when the gap opens and widens, which cross-sectional data can’t. | Attrition. People drop out of longitudinal studies over the years, and if poorer families drop out more often, the remaining sample becomes less representative. |
| Several datasets. Using four separate studies is a form of triangulation: findings that appear in more than one dataset are more reliable. | Measuring attitudes. Aspirations and beliefs are measured through survey answers, which may not capture what people really think and do. Interpretivists would want qualitative data on what aspirations mean to families. |
| Policy relevance. Quantitative findings about when gaps appear are useful to policymakers deciding where to intervene. | Age of the data and the proxy. The children studied grew up in the 2000s or earlier, and “the poorest fifth” is a measure of income, not of social class as such. |
How this could come up in the exam
This study is a good example to use in a Methods in Context question (Question 05 on Paper 1), where you’re asked to evaluate a method for researching an issue in education. For example, June 2021 (sat November 2021), Question 05: Applying material from Item C and your knowledge of research methods, evaluate the strengths and limitations of using secondary data to investigate the academic progress of pupils in schools. [20 marks]. The 2010 study is itself a secondary analysis of existing datasets, so it shows both the strengths and the limitations. See my answer to that question.
For the education topic, it’s evidence for home-based explanations in essays such as June 2018, Question 04 on social class differences in achievement. See my essay on that question.
Sociology of Education Revision Bundle
If you like this sort of thing, then you might like my sociology of education revision notes bundle, written by Karl Thompson, who taught A-level sociology for eighteen years and spent twelve as an AQA examiner, marking all three papers on the current specification. It contains the following:
- 34 pages of revision notes
- mind maps in pdf and png format – 9 in total, covering various topics within the sociology of education
- short answer exam practice questions and exemplar answers
- how to write sociology essays, including 7 specific templates and model answers on the sociology of education
Signposting and Related Posts
This post is relevant to both the sociology of education and research methods. For more on the method, see longitudinal studies: definition, advantages and disadvantages. For the figures on class and achievement today, see social class and educational achievement statistics.
Please click here to return to the homepage – ReviseSociology.com
Sources
Goodman, A. and Gregg, P. (eds) (2010) Poorer children’s educational attainment: how important are attitudes and behaviour? York: Joseph Rowntree Foundation. JRF summary.
Feinstein, L. (2003) Inequality in the early cognitive development of British children in the 1970 cohort. Economica, 70(277), 73–97.
Jerrim, J. and Vignoles, A. (2013) Social mobility, regression to the mean and the cognitive development of high ability children from disadvantaged homes. Journal of the Royal Statistical Society: Series A.
Pingback: Longitudinal Studies | ReviseSociology