Multigenerational effects on educational outcomes: a systematic - PDF document

Multigenerational effects on educational outcomes: a systematic review [Multi-generational Social Mobility Workshop version, September 2017] Lewis Anderson, Paula Sheppard & Christiaan Monden Department of Sociology, University of Oxford lewis.anderson@trinity.ox.ac.uk Abstract Is the intergenerational transmission of educational attainment a Markov process? Interest in this question has swelled recently and results from a variety of studies have tended to suggest that grandparent s’ socioeconomic characteristics are associated wi th children’s educational outcomes, independently of the resources of the parental generation – a finding with important implications for the estimation of populations’ social mobility. But how robust is this association? In this paper we critically and systematically review the extant literature on grandparent – or multigenerational – associations with children’s educational outcomes. By comprehensively surveying the field, we explore whether findings vary substantially by study features such as country, data availability, the operationalisation of grandparental resources, and controls for parental characteristics. Results indicate a striking lack of any link between such variables and the likelihood of an analysis finding a significant association. We then discuss what light the literature taken as a whole can shed on recurrent issues such as the validity, interactivity, size, distribution, and mechanisms of grandparent effects. This research is supported by European Research Council consolidator Grant for the FAMSIZEMATTERS project 1

1. INTRODUCTION A long and fruitful tradition in sociology has examined the intergenerational persistence of socioeconomic characteristics. A wide variety of analytical approaches and datasets have been deployed in this enterprise, but the great majority of studies have in common a two-generation approach: for the purposes of determining the influence of family background on an individual’s life chances, the former is effectively equated with the characteristics of that individual’s parents. This approach makes an implicit assumption about the intergenerational transmission of advantage over successive generations, namely that this long-run process can be adequately described as a series of independent associations between adjacent generations. Thus, insofar as grandparents affect the outcomes of their grandchildren, this effect is indirect, or Markovian, which is to say that the grandparental effect is fully mediated through the parental generation. An important recent development in the social sciences has been the increasing availability of data on more than the usual two generations (Song & Campbell, 2017). A recurrent finding from these data is that the Markov process envisaged by the two-generation literature overestimates social mobility. That is, taking estimates of the persistence of social status based on two observed generations and extrapolating over three or more tends to yield an estimate of this persistence over three or more generations which is lower than an estimate based on three or more actually observed generations. Whereas Becker & Tomes concluded that ‘ [a]lmost all earnings advantages and disadvantages of ancestors are wiped out in three generations’ (Becker & Tomes, 1986: S1), more recent empirical findings do not bear this out (e.g. Kroeger & Thompson, 2016; Pfeffer & Killewald, 2016; Lindahl, Palme, Massih, & Sjögren, 2015; Clark, 2014). Socioeconomic status does not appear to decay geometrically across generations. Two types of explanation have been proposed to account for this. Clark (2014) posits an underlying latent factor determining socioeconomic status, which is transmitted between adjacent generations (in a fully Markovian process) at a higher rate than any observed measure, with an intergenerational correlation coefficient around 0.75. According to this argument, intergenerational correlations in observed measures such as income or education are lower because the latent factor does not translate perfectly into observed measures of socioeconomic status. This is both because individuals are subject to random disturbances (‘market luck’), and because they may choose to trade off dimensions of status, for instance forgoing a high-income occupation for one with more prestige. Clark holds that the intergenerational correlation of this latent factor is constant across a range of contexts, resistant to macroeconomic conditions or policies aiming to increase social mobility, and hence further that it is for the most part genetically transmitted. A major limitation of Clar k’s evidence is that it depends on extrapolating from the top (e.g. US physicians, the Swedish nobility, the extremely wealthy) to the whole of the status distribution. Also, Breen observes that genetic relatedness ‘disappears quickly’ over generations, such that an ‘truly remarkable degree of genetic assortative mating’ (2015: 302) would be necessary for the long-term persistence of social status to be driven primarily by genetics. The more common explanation for the non-geometric decay of socioeconomic status across multiple generations is that ancestors more distant than the parental generation directly affect a child’ s outcomes. That is, their influence is not mediated through intervening generations as in the traditional two-generation approach or approaches like Clark’s . As figure 1 illustrates, research exploring whether this is the case has proliferated recently, especially since Mare (2011) called for a ‘multigenerational view of inequality’, arguing that the processes generating social stratification are unlikely to be fully described by the usual approach of beginning with individuals and looking at their parents as the determinants of their relative positions. Not only does this usual approach miss the question of how successfully individuals are able to reproduce social advantage – a question which 2

requires attention to demography – it misses the possible channels beyond one’s most immediate ancestors through which family background influences life chances. Figure 1 Studies testing a direct grandparent effect on educational outcomes (N=40) With a focus on educational outcomes, this paper examines efforts to answer Mare’s call, as well as earlier relevant evidence. That is, we critically and systematically review the evidence for a direct effect of grandparental resources or characteristics (Generation 1 or G1) on grand children’s educational outcomes (G3). Within this broad survey, we address several specific issues. First, given that the majority of studies report some form of significant G1-G3 association, how robust is this association? It may be, for instance, that apparent direct grandparent effects are due to measurement error in parental socioeconomic status. We examine whether study characteristics such as the level of information available on parental socioeconomic status are correlated with finding a grandparent effect. Second, through what mechanisms might a grandparent effect act, and what is the evidence for these? Third, we examine evidence on the distribution of the grandparent effect, since there are reasons to expect heterogeneity in its operation. We use a mix of quantitative and narrative methods to give a broad overview of the evidence whilst also paying attention to consequential particulars. The paper is organised as follows. Section 2 describes the methods of our systematic literature search and review. Section 3 presents results, both quantitative and narrative. In section 4 we discuss our findings and offer conclusions and directions for future research. 2. METHODS 2.1 Literature search We restrict the scope of this review to educational outcomes as education is arguably the central stratifying characteristic in modern societies. Moreover, this restriction keeps the review 3