A Multiple Approach Analysis of Inclusion in Deliberation Exercises

• In Event: Elite Cues, Policy Information, and Discourse

Abstract

As deliberation exercises have moved from the realm of political theory to practice, one of the major obstacles has been the production of events in which “all of those who are possibly affected by a decision have equal chances to enter and take part” (Habermas 1996: 305). Studies have found significant disparities by both gender (e.g., Karpowitz and Mendelberg 2014) and race (e.g., Kennedy et al. 2021) in both the recruitment and participation portions of deliberation exercises. The approach of these studies, however, has generally been unitary, with race and gender treated as separate variables, and the confluence of these attributes treated as additive (Wojciechowska 2019). This formulation is problematic, given that much of the theoretical literature in the past 30 years has moved to an intersectional frame for understanding political disparities, where race and gender are treated as co-constitutive of political oppression (see e.g., Hancock 2016). Much of the reason for this omission stems from the lack of statistical power in previous studies of deliberation (Kennedy et al. 2021).

This study takes a multiple approach (Hancock 2007) to understanding the implications of race and gender in deliberative participation. Using a uniquely diverse deliberation setting with more than 4,000 participants, we analyze how the interaction of race and gender influence multiple aspects of deliberative behavior. First, we analyze the scale of participation, finding that, while race is the predominant factor in differences, particular intersections are particularly disadvantaged (Black men and Latina/Hispanic women). Second, we utilize Bidirectional Encoder Representations from Transformers (BERT) to code and analyze the rate of particular deliberative behaviors associated with deliberative quality (Himmelroos 2017), noting substantial differences in such behavior between gender/race groups. Finally, we utilize a series of dictionary and machine learning models to detect differences in topics discussed by respondents, again noting important differences in the style of discussion and topics addressed in the gender/race intersection. These findings have important implications for the design of future deliberation exercises and the study of these events.

Habermas, J., 1996. Between facts and norms. Oxford: Polity Press.

Hancock, A.M., 2016. Intersectionality: An intellectual history. Oxford University Press.

Hancock, A.M., 2007. When multiplication doesn't equal quick addition: Examining intersectionality as a research paradigm. Perspectives on politics, 5(1), pp.63-79.

Himmelroos, S., 2017. Discourse quality in deliberative citizen forums–A comparison of four deliberative mini-publics. Journal of Deliberative Democracy, 13(1).

Karpowitz, C.F. and Mendelberg, T., 2014. The silent sex: Gender, deliberation, and institutions. Princeton University Press.

Kennedy, R., Sokhey, A.E., Abernathy, C., Esterling, K.M., Lazer, D.M., Lee, A., Minozzi, W. and Neblo, M.A., 2021. Demographics and (equal?) voice: Assessing participation in online deliberative sessions. Political Studies, 69(1), pp.66-88.

Wojciechowska, M., 2019. Towards intersectional democratic innovations. Political Studies, 67(4), pp.895-911.

Authors

• Ryan P. Kennedy, Ohio State University

• Amanda Talbott Austin, University of Houston

• William Minozzi, Ohio State University