Wastage of talent?

(2017) - Erzsébet Bukodi, Mollie Bourne, Bastian Betthäuser

Journal: Advances in Life Course Research
Link:: https://linkinghub.elsevier.com/retrieve/pii/S1040260817300345
DOI:: 10.1016/j.alcr.2017.09.003
Links::
Tags:: #paper #NCDS #CognitiveAbility #BCS #Attainment
Cite Key:: [@bukodiWastageTalent2017]

Abstract

The extent to which societies suffer ‘wastage of talent’ due to social inequalities in educational attainment is a longstanding issue. The present paper contributes to the relevant literature by examining how social origins and early-life cognitive ability are associated with educational success across three British birth cohorts. We address questions of over-time change, bringing current evidence up-to-date. Our findings reinforce the well-established trend that the importance of cognitive ability declined for cohorts born between 1958 and 1970, but we show that for a cohort born in the early 1990s this trend has reversed. We further show that the relative importance of family background has not seen a corresponding decline. In distinguishing between different components of social origins, we show that family economic resources have become somewhat less important for children’s educational success, while socio-cultural and educational resources have become more important. Even high ability children are unable to transcend the effects of their social origins. The problem of ‘wastage of talent’ remains; young people from disadvantaged backgrounds are still lacking the opportunity to fully realise their potential within the British educational system.

Notes

“societies suffer ‘wastage of talent’ due to social inequalities in educational attainment is a longstanding issue” (Bukodi et al., 2017, p. 34)

“Our findings reinforce the well-established trend that the importance of cognitive ability declined for cohorts born between 1958 and 1970, but we show that for a cohort born in the early 1990s this trend has reversed” (Bukodi et al., 2017, p. 34)

“relative importance of family background has not seen a corresponding decline.” (Bukodi et al., 2017, p. 34)

“problem of ‘wastage of talent’ remains; young people from disadvantaged backgrounds are still lacking the opportunity to fully realise their potential within the British educational system.” (Bukodi et al., 2017, p. 34)

“Marked social inequalities in educational attainment still show up, and over the recent past would appear to have been reduced only to a quite limited extent; and in some cases − such as Britain − scarcely at all (e.g. Breen, Luijkx, Müller, & Pollak, 2009; Bukodi & Goldthorpe, 2013).” (Bukodi et al., 2017, p. 34)

“First, although in this paper we focus on individuals’ cognitive ability, we by no means claim that cognitive ability is the only form of ‘talent’” (Bukodi et al., 2017, p. 34)

“Second, we do not seek to address in any direct way the much-debated issue of the relative importance of genetic and environmental factors in the formation of cognitive ability. In this regard, we need to say only the following: the assumption that variation in cognitive ability within a population can be simply divided up into one part that is due to environmental effects and another part that is due to genetic effects no longer appears tenable” (Bukodi et al., 2017, p. 34)

“accepting the fact that if cognitive ability is measured, by standard tests, at a relatively early stage in children’s lives, it is strongly associated with their subsequent educational attainment, regardless of their social origins (see e.g. Strenze, 2007; Deary et al., 2007).” (Bukodi et al., 2017, p. 35)

“British studies have demonstrated that the role of cognitive ability in affecting individuals’ educational attainment, net of the influence of their social origins, declined across cohorts born between 1946 and 1958 (Richards, Power, & Sacker, 2009) and between 1958 and 1970 (Galindo-Rueda & Vignoles, 2005; Schoon, 2010; Bukodi, Erikson, & Goldthorpe, 2014), but have not yet sought to determine whether this trend has continued for those born more recently” (Bukodi et al., 2017, p. 35)

“In regard to the role of social origins in individuals’ educational attainment, net of their cognitive ability, the existing literature is also inconclusive. Past research in Britain reports substantial effects of family background (e.g. Galindo-Rueda & Vignoles, 2005; Mood, Jonsson, & Bihagen, 2012) that are only moderately reduced by cognitive ability and are largely persistent over time (Bukodi et al., 2014). But conflicting evidence exists from the US and other parts of Europe.” (Bukodi et al., 2017, p. 35)

“much past research into inequalities in educational attainment in relation to individuals’ family backgrounds the implicit assumption has been made that different features of social origins can serve as ‘interchangeable indicators’; i.e. it matters little which is taken as the basis of analyses since the results obtained will be essentially the same. We believe that this assumption is an untenable one” (Bukodi et al., 2017, p. 35)

“while different features of individuals’ social origins are certainly correlated, the correlations are far from perfect, and it is therefore necessary to take a multidimensional approach and to consider how far different features of social origins − most importantly, parental class, parental status and parental education − have independent effects on individuals’” (Bukodi et al., 2017, p. 35)

“Given individuals’ cognitive ability, as measured relatively early in life, to what extent do social origins, when taken to comprise the three different components, still have separate and independent effects on individuals’ educational attainment, and are there changes in these effects across” (Bukodi et al., 2017, p. 35)

“cohorts?” (Bukodi et al., 2017, p. 35)

“we take individuals whose early-life cognitive ability is at a similar level, how far do social origins create disparities in their later educational attainment” (Bukodi et al., 2017, p. 35)

“Appendix A in” (Bukodi et al., 2017, p. 35)

“we apply multiple imputation (MI) methods using chained equations” (Bukodi et al., 2017, p. 36)

“We measure educational attainment at age 20, since this is the latest age at which we have information for the youngest cohort” (Bukodi et al., 2017, p. 36) What does this mean?

“We measure individuals’ cognitive ability by capturing the common variance across a range of test items of verbal and non-verbal ability (Deary, 2001; Colom, Abad, Garcia, & Juan-Espinosa, 2002” (Bukodi et al., 2017, p. 36)

“NCDS cognitive ability was assessed at 11 years of age via a general ability test. In BCS70 scores from four subtests of the British Ability Scales (BAS), taken at age 10, are given” (Bukodi et al., 2017, p. 36)

“For each cohort, the measures were used in a principal component analysis (PCA), saving scores from the first component extracted. Previous studies have taken this approach, determining that these scores provide a comparable cross-cohort measure that adequately captures the construct of ‘g’ (e.g. Bukodi et al., 2014; Vignoles, 2005; Schoon, 2008; , 2010; Spearman, 1904)” (Bukodi et al., 2017, p. 36)

“ability in relative rather than absolute terms. We therefore allocate individuals to cohort-specific cognitive ability quintiles, in this way controlling for ‘Flynn effects’ (Flynn, 1987), and also allowing any nonlinear effects on educational attainment to show up” (Bukodi et al., 2017, p. 37)

“We take measures of social origins at around cohort members’” (Bukodi et al., 2017, p. 37)

“We conduct sets of logistic regression analyses for each cohort separately, with each educational threshold taken in turn as the dependent variable. Our independent variables are cognitive ability quintiles, parental class, status and education, with a control for gender. We present the average marginal effects (AMEs) of individuals’ cognitive ability on the likelihood of attaining each educational threshold in Fig. 1.” (Bukodi et al., 2017, p. 37)

“For each cohort and at each educational threshold, cognitive ability plays a large part in affecting individuals’” (Bukodi et al., 2017, p. 38)

“For the 1990s cohort, for example, those who demonstrate a high level of ability early in life have a 29% higher likelihood of exceeding Threshold 3 by age 20 than those whose ability level is in the middle range.” (Bukodi et al., 2017, p. 38)

“At each educational threshold, differences in the effect of cognitive ability are statistically significant between the 1958 and 1970 cohorts” (Bukodi et al., 2017, p. 38)

“Being in the top ability quintile, relative to the third, is less beneficial, and being in the bottom quintile, relative to the third, is less detrimental for individuals in the 1970 cohort than it is for those in the 1958 cohort” (Bukodi et al., 2017, p. 38)

“The question that then arises is whether this increase in the net effect of ability is met with a corresponding decrease in the net effects of social origins” (Bukodi et al., 2017, p. 38)

“Each of our three components of social origins has significant effects, over and above the effects of cognitive ability, on individuals’ likelihood of attaining the three educational thresholds, and this is the case in all cohorts” (Bukodi et al., 2017, p. 38) THIS is some real shit

“We have established that social origins have persistent effects on educational attainment, despite the increasing importance of cognitive ability between the two latest cohorts” (Bukodi et al., 2017, p. 40)

“First, taking individuals with similar ability levels, the importance of social origins for educational success apparently depends on the level of education being considered. In all cohorts, disparities by social origins, overall, decrease with increasing cognitive ability at the lowest educational threshold and increase with increasing cognitive ability at the highest educational threshold” (Bukodi et al., 2017, p. 40)

“Second, in all cohorts, parental education appears to be the most important stratifying component of social origins.” (Bukodi et al., 2017, p. 40)

“Third, disparities by social origins − the socio-cultural and educational components in particular − tend to be largest for the 1990s cohort, by and large, at all levels of cognitive ability.” (Bukodi et al., 2017, p. 40)

“we should note that the analytical strategy that we have adopted of taking early-life cognitive ability as a given in investigating its effects on educational attainment together with those of social origins will obviously underestimate the importance of the latter, insofar as they also have an effect on the formation of cognitive ability. And, as we earlier indicated, there is good evidence for supposing that this effect is a significant on” (Bukodi et al., 2017, p. 41)

“findings that we have reported on the large and persisting importance of social origins have therefore to be regarded as providing a quite conservative, even if still effective, basis for rejecting claims of the ‘decline of the social’ (Marks, 2014): that is, one that could only be strengthened with a recognition of the wider influence of social origins on children’s” (Bukodi et al., 2017, p. 41)

“normative standpoint, most disturbingly emerges from our results is that even children who demonstrate high levels of ability in early life are unable to transcend the effects of their social origins so far as their educational attainment is concerned. Although in the most recent cohort we consider there has been some increase in the importance of individuals’ cognitive ability, there has been no corresponding decline in the importance of their social origins” (Bukodi et al., 2017, p. 41)

“young people coming from disadvantaged social backgrounds are still lacking the opportunity to fully realise their potential within the British educational system.” (Bukodi et al., 2017, p. 41)