Forecasting Undergraduate Majors Using Academic Transcript Data

Committing to a major is a fateful step in an undergraduate education, yet the relationship between courses taken early in an academic career and ultimate major selection remains little studied at scale. Using transcript data capturing the academic careers of 26,892 undergraduates enrolled at a private university between 2000 and 2020, we describe enrollment histories using natural-language methods and vector embeddings to forecast terminal major on the basis of course sequences beginning at college entry. We find (I) a student's very first enrolled course predicts major thirty times better than random guessing and more than a third better than majority-class voting, (II) modeling strategies substantially influence forecasting accuracy, and (III) course portfolios varies substantially within majors, raising novel questions what majors mean or signify in relation to undergraduate course histories.

The Realignment of Class Politics and Class Voting

Since the early days of the study of political behavior, class politics has been a key component. Initially researchers focused on simple manual versus nonmanual occupations and left versus right parties, and found consistent evidence of a strong effect of class on support for left-wing parties. This finding was assumed to be simply a matter of the redistributive preferences of the poor, an expression of the “democratic class struggle.” However, as the world became more complex, many established democracies developed more nuanced class structures and multidimensional party systems. How has this affected class politics? From the simple, but not deterministic pattern of left-voting workers, the early 21st century witnessed substantial realignment processes. Many remain faithful to social democratic (and to a lesser extent radical left) parties, but plenty of workers support radical right parties or have left the electoral arena entirely. To account for these changes, political scientists and sociologists have identified two mechanisms through which class voting occurs. The most frequently studied mechanism behind class voting is that classes have different attitudes, values, and ideologies, and political parties supply policies that appeal to different classes’ preferences. These ideologies are related not only to redistribution but also to newer issues such as immigration, which appear to some degree to have replaced competition over class-related inequality and the redistribution of wealth as the primary axis of class politics. A secondary mechanism is that members of different classes hold different social identities, and parties can connect to these identities by making symbolic class appeals or by descriptively representing a class. It follows that class realignment can occur either because the classes have changed their ideologies or identities, because the parties have changed their policies, class appeals, or personnel, or both. Early explanations focused on the classes themselves, arguing that they had become more similar in terms of living conditions, ideologies, and identities. However, later longitudinal studies failed to find such convergences taking place. The workers still have poorer, more uncertain, and shorter lives than their middle-class counterparts, identify more with the working class, and are more in favor of redistribution and opposed to immigration. While the classes are still distinctive, it seems that the parties have changed. Several social democratic parties have become less representative of working-class voters in terms of policies, rhetorical appeals, or the changing social composition of their activists and leaders. This representational defection is a response to the declining size of the working class, but not to the changing character or extent of class divisions in preferences. It is also connected to the exogeneous rise of new issues, on which these parties tend not to align with working-class preferences. By failing to represent the preferences or identities of many of their former core supporters, social democratic parties have initiated a supply-side driven process of realignment. This has primarily taken two forms; class–party realignments on both left and right and growing class inequalities in participation and representation.

The intergenerational foundations of class voting: social mobility and electoral choice in Western Europe

Scholarly explanations of the survival of left parties and the upsurge in mainstream politics discontent often refer to voters' intergenerational mobility resulting from the post-industrial transition. As the occupational structure evolves, voters across generations are exposed to heterogenous life chances, and the social elevator progressively alters class voting patterns. Yet empirical evidence for the electoral implications of social ascent and decline as well as their reasons is mixed at best – likely because most empirical studies seek for homogenous average mobility effects. To address this limitation, we analyse the diverse consequences of mobility across social groups in a quasi-descriptive fashion by applying a cutting-edge ANOVA-based OLS model. Contrarily to prior studies, this approach allows us to identify class-specific mobility effects on voting (ceteris paribus), consistently with theory. Our analyses draw on individual-level detailed information on both intergenerational social mobility and political behaviour from the European Social Survey (rounds 1-9) across 19 Western European countries. Although scholarly accounts on the consequences of social mobility averagely find little to no support in our analyses, we do observe some significant and substantial class-specific effects of both social ascent and descent on voting choice.

We have been left behind, haven't we? Economic status loss, class voting and the populist radical right

A growing body of research attempts to reconcile economic and cultural explanations of populist radical right (PRR) voting by highlighting citizens' resentment against their gradual marginalisation within society. Nonetheless, widespread speculations about the deteriorating relative economic position of PRR voters are not supported by proper empirical evidence. To address this shortage, the present study first provides a theoretical discussion of the electoral consequences of economic status loss by bridging multidisciplinary literature on relative economic inequality and group deprivation; subsequently, it assesses such consequences empirically, by means of a novel measure of economic status loss. Our multilevel analysis on ESS and EU-SILC data on 19 elections (2008-2017) across 9 Western European countries demonstrates that PRR parties are most successful among social classes facing a collective decrease in economic status – rather than material deprivation per se. This result is consequential for scholarly debates on the reasons for class PRR alignment and on the electoral repercussions of economic inequalities.

Understanding electoral change

This chapter examines the contributions of realignment and dealignment to electoral change. Realignment is particularly apparent in the emergence of sociopolitical centre-periphery cleavages. Post-materialism and attitudes to European integration have clearly served to realign electoral support, while there are also growing signs of electoral differences based on age, education, and gender. Dealignment is evident principally in the weakening of partisan attachments, which in turn can be traced largely to the factors which underlie the decline of class voting, such as the strategic behaviour of parties themselves and also, to some extent, the erosion of distinctive and cohesive social classes, which is itself partly a consequence the changing shape of the socio-economic structure.

Class voting or economic voting? Electoral support for chavismo (1998–2015)

This article analyses electoral support for chavismo in Venezuela from 1998 to 2015, comprising five presidential elections (1998, 2000, 2006, 2012, and 2013) and the legislative election of 2015. Drawing on a comprehensive historical series, the findings contradict an influential body of literature on Venezuelan politics and show that economic voting prevailed during the analysed period. In relation to class voting, the analysis does not find a monotonic vote, in which the poor supported Hugo Chávez and his allies, whereas the rich rejected them, at each election. The direction of associations between these classes and voting for chavismo varied over the investigated period. The findings have important implications for Latin American politics, showing the relevance of economic factors for the left turn in regional politics and helping explain the recent losses of leftist parties in presidential elections.

Conservative Hegemony?

This chapter concerns the British general elections of 1983, 1987, and 1992. All three were won by the Conservatives. In 1983, the party was returned with a majority of 144 seats, despite having been deeply unpopular for much of its term of office thanks to economic recession and an unprecedented post-war level of unemployment. The scale of the victory was partly due to the 1982 Falklands conflict, in which the prime minister, Margaret Thatcher, had seemingly fulfilled her promise to restore British pride and prestige. However, the Conservatives also benefited from a divided opposition, with Labour and the newly formed Liberal/SDP Alliance winning a plurality of votes between them. In 1987 the situation was similar, although by this time the economic outlook had improved and the Conservatives benefited from a ‘feel-good factor’. By 1990 Mrs Thatcher had once again become deeply unpopular, and was replaced by the less controversial (but uncharismatic) John Major. Under his leadership the Conservatives secured a record tally of votes in the 1992 election, but their overall parliamentary majority was greatly reduced thanks to the operation of the electoral system. In each case, relevant developments in the preceding inter-election period are described (including trends in party popularity) and an account of the campaign provided. In addition, the election results themselves—patterns of party support and of turnout—are extensively analysed. The chapter also discusses the academic controversy over the extent of class voting in Britain, which emerged at the time, as well as the growing North–South regional divide in party support. Finally, the suggestion that this period was one of ‘Conservative hegemony’ is considered.