Primary and Secondary Effects in Class Differentials in Educational Attainment: The Transition to A-Level Courses in England and Wales

(2007) - Michelle Jackson, Robert Erikson, John H. Goldthorpe, Meir Yaish

Journal: Acta Sociologica
Link:: http://journals.sagepub.com/doi/10.1177/0001699307080926
DOI:: 10.1177/0001699307080926
Links::
Tags:: #paper #NCDS #Attainment #Transition #school-to-work
Cite Key:: [@Jackson2007]

Abstract

In this article we start from Boudon’s important, but still surprisingly neglected, distinction between ‘primary’ and ‘secondary’ effects in the creation of class differentials in educational attainment. Primary effects are all those, whether of a genetic or socio-cultural kind, that are expressed via the association between children’s class backgrounds and their actual levels of academic performance. Secondary effects are those that are expressed via the educational choices that children from differing class backgrounds make within the range of choice that their previous performance allows them. We apply a method introduced by Erikson and Jonsson to represent the relationship between primary and secondary effects in analysing class differentials in one crucial transition within the English and Welsh educational system: that which children make at around age 16 and which determines whether or not they will pursue the higher-level academic qualifications –A-levels – that are usually required for university entry. We then use a development of this method that we have earlier proposed in order to produce quantitative estimates of the relative importance of primary and secondary effects as they operate within this transition. We show that secondary effects reinforce primary effects to a substantial extent, accounting for at least one quarter, and possibly up to one-half, of class differentials as measured by odds ratios. In conclusion, we consider some theoretical and policy implications of our findings.

Notes

“Primary effects are all those, whether of a genetic or socio-cultural kind, that are expressed via the association between children’s class backgrounds and their actual levels of academic performance. Secondary effects are those that are expressed via the educational choices that children from differing class backgrounds make within the range of choice that their previous performance allows them.” (Jackson et al., 2007, p. 211)

“e show that secondary effects reinforce primary effects to a substantial extent, accounting for at least one quarter, and possibly up to one-half, of class differentials as measured by odds ratios.” (Jackson et al., 2007, p. 211)

“Boudon (1974: 29–31) introduced an important, but still surprisingly neglected, distinction between ‘primary’ and ‘secondary’ effects in the creation of class differentials in educational attainment” (Jackson et al., 2007, p. 212)

“That is to say, children from more advantaged class backgrounds tend on average to take up more ambitious educational options than do children from less advantaged backgrounds, even when level of previous academic performance is held constant (for early results, see Boalt, 1947, and for a review of more recent research, see Goldthorpe, 2007, vol. II: ch. 2).” (Jackson et al., 2007, p. 212)

“As earlier indicated, we work with the standard threefold collapse of the schema: salariat or service class (Classes I and II), intermediate classes (Classes III, IV and V) and working class (Classes VI and VII). A more refined treatment of class background would not, we believe, be advisable in view of a possible lack of comparability across our data sources” (Jackson et al., 2007, p. 213) This is pretty lazy.

“However, we focus on grades in just two subjects: mathematics and English. The main reason for this is that these count as almost obligatory subjects in the examinations in question and ones in which, therefore, nearly all students will have grades.” (Jackson et al., 2007, p. 214)

“To obtain a single measure of performance for each student, we use the ‘scores’ that are officially attached – for purposes of aggregation – to ordinal grades and, for each student, simply sum the scores corresponding to the grades that he or she obtained in mathematics and English.” (Jackson et al., 2007, p. 214)

“deal with problems resulting from changes over time in the form and nomenclature of the public examinations, we take over the equivalencies in the scoring of grades under different examination systems that have become conventionally accepted in educational administration and research, although we appreciate that the detail of these has in some respects been questioned.7 The resulting scores are inverted (so that high scores indicate high ability) and then standardized to z-scores with a mean of zero and standard deviation of unity.8” (Jackson et al., 2007, p. 214)

“ver the period we cover, 1974 to 2001, a strong increase occurred in the total numbers of students taking up such courses. However, class differentials, as measured by odds ratios, reveal no consistent decline. We have been able to show that in understanding this situation both primary and secondary effects must be given an important part” (Jackson et al., 2007, p. 223)

“class differentials persist in academic performance: i.e. in grades obtained in public examinations taken at the end of the period of compulsory schooling. On the other hand, class differentials also persist in transition propensities at all levels of academic performance, though most markedly at intermediate levels” (Jackson et al., 2007, p. 224)