IOE EPrints

Does education have an impact on mothers' educational attitudes and behaviours? [Wider Benefits of Learning Research Report No. 16]

Feinstein, Leon and Sabates, Ricardo (2006) Does education have an impact on mothers' educational attitudes and behaviours? [Wider Benefits of Learning Research Report No. 16]. [Report]

Text (WBL Research Report No. 16)
WBLResRep16.pdf - Published Version

Download (615Kb) | Preview


The different elements of parents' educational attitudes and behaviours have been well documented and are identified as having a significant effect on children's levels of educational achievement. It therefore seems likely that some of the intergenerational effects of education may be transmitted through parents' educational attitudes and behaviours. However, empirical research to date has not been robust enough to test whether there is a causal effect of education on such attitudes and behaviours. This report fills the gap by providing a rigorous estimate of the educational effect. Whilst initial analysis shows an association between the age at which a mother leaves full-time education and her subsequent educational attitudes and behaviours, through using instrumental variable (IV) methodology we find that this link is not the result of causal effects of post-compulsory education but rather it is due to underlying related differences between mothers who stayed on in education and mothers who did not. An interesting implication of these results is the emphasis they place on the positional importance of education – as educational levels for those with lower educational aspirations increase, individuals with positional ambition increase their education further in order to maintain their relative advantage. The results of our research strongly indicate the possibility that the apparent relationships shown by ordinary least squares methodology (OLS) may be spurious as indicators of a causal relationship. This should be taken into account when drawing inferences from OLS studies, particularly those with few control variables. If policy reform causes individuals to self-select within the system, then educational interventions based on observed links between education and outcomes may not generate the expected results.

Item Type: Report
Divisions: IOE Departments > Departments > Centre for Longitudinal Studies
IOE Departments > Departments > Quantitative Social Science
Depositing User: Atira Pure
Date Deposited: 24 Jul 2014 10:04
Last Modified: 29 Jan 2015 10:09
View Item View Item