John A. Bernau

PhD Candidate, Sociology

github twitter linkedin email
AJS - Vol. 123 No. 4 - January 2018
Jul 24, 2018

Craciun, Mariana. 2017. “Emotions and Knowledge in Expert Work: A Comparison of Two Psychotherapies.” American Journal of Sociology 123(4):959–1003.

Summary: The author uses ethnographic data to examine the role of emotions in the work of psychoanalysis (PA) and cognitive-behavioral (CB) therapy. She proposes a model of emotions as supportive, didactic, and inductive (an “affective-relational framework”). While both therapies use emotions, PA knowledge and/or findings don’t translate well outside their discipline (i.e. to insurers). CB uses scales and tests – a mechanism much more familiar to outsiders. This is one of the reasons for its current dominance in the field - the ability to communicate results and credibility to outsiders. If the outside world is playing the evidence-based-medicine, Weberian-rationalization game, you have to play along.

Quote: “By shunning the methods of evidence-based practice, PA therapists have lost ground to their better-positioned colleagues in pharmacology and CB therapy who draw on institutional sources of legitimacy.” p991


Heerwig, Jennifer A. 2017. “Money in the Middle: Contribution Strategies among Affluent Donors to Federal Elections, 1980–2008.” American Journal of Sociology 123(4):1004–63.

Summary: Who gives money to political candidates and why? How have their strategies changed over the last few decades? Individual donors have always been (and continue to be) the most significant source of funding for candidates in federal elections. The author assembled data on over 15 million itemized contributions from the FEC disclosure records. To examine longitudinal patterns, each itemized donation had to be matched to individual donors. While each contribution entry recorded donor name, address, and occupation, these responses were not standardized over the years, presenting “a daunting number of missing values and discrepancies”: addresses changed, names were misspelled, occupations had varying descriptions, etc. The author employed a probabilistic record-matching procedure to calculate a match score based on the similarity of any two records. A clever way to get around this messy within-person variability! In short: more frequent donors are more likely to be bipartisan givers, but this contribution strategy declined over the past few decades. Previous literature has assumed individual donors to be ideologically motivated and thus more partisan, but these results suggest a more diffused, bipartisan, “access-oriented”, quid pro quo strategy. However, this strategy has declined recently, shedding light on a new angle of the oft-discussed political polarization.

Quote: “Relative to the 1984 cohort, the odds of the 1990 entry year cohort ever giving to both parties decreased by 23%. The odds declined by 40% for the 1994 cohort, by 49%, for the 1996 cohort, and by 59% for the 2000 cohort.” p1025


VanHeuvelen, Tom. 2017. “Recovering the Missing Middle: A Mesocomparative Analysis of Within-Group Inequality, 1970–2011.” American Journal of Sociology 123(4):1064–1116.

Summary: Reading generalist journals can push you out of your comfort zone. VanHeuvelen’s formidable article did this for me both substantively and methodologically. He starts with a discussion of within-group inequality (WGI). Despite popular attention to between-group inequality (especially in the social sciences), most observable inequality occurs within demographically similar groups. The article also provides a brief review of the Kuznets curve and the Great U-Turn: two popular heuristics for understanding the relationship between economic development and inequality. Constructing an original longitudinal dataset of 722 geographic commuting zones (CZs) in the US over 41 years, the author examines the relationship between economic development and BGI and WGI. Results indicate that WGI, but not BGI, follows a U-shaped pattern. In other words, economic development not only affects the level of inequality at the local level, but also the type and form. By using a meso-level approach, social scientists are afforded a better understanding of how WGI depends on local features of economic development.

Quote: “Cross-sectionally, WGI accounts for 62% to 72% of total inequality, which simply illustrates that the bulk of inequality in any period of time occurs within, rather than between, groups.” p1065


Hirsh, Elizabeth and Youngjoo Cha. 2017. “For Law and Markets: Employment Discrimination Lawsuits, Market Performance, and Managerial Diversity.” American Journal of Sociology 123(4):1117–60.

Summary: How do civil rights lawsuits affect the race and sex composition of managerial teams? Coming out only a few months after the #MeToo movement broke (and likely in press much earlier), Hirsch and Cha’s article provides information on 171 high-profile lawsuits against publically traded companies from 1997 – 2007 to identify important variables giving rise to larger patterns of workplace diversity. In short, the authors found four factors that significantly increased managerial sex and race diversity in the aftermath of a discrimination lawsuit: 1) an immediate drop in stock price, 2) national media attention, 3) court-mandated accountability policies, and 4) EEOC support. More surprising is the lack of effect of costly monetary payouts on increasing minority representation. The authors thus discount simple rational-economic approaches to organizational change in favor of a (neo)institutional approach. I especially liked their measure of “cumulative abnormal return” (CAR) which they calculated by 1) using a regression to predict daily stock price in an eleven-day event window, 2) summing the residuals, and 3) dummy coding to 1 if negative (suggesting a lower overall return than otherwise expected). The article does an excellent job presenting results graphically while explaining coefficients in odds ratios or percent change in odds.

Quote: “Employers facing extensive fines may assume that the monetary penalty is response enough and, even worse, curb the hiring, promotion, and retention of protected groups as a litigation prevention strategy.” p1153


Logan, John R. and Matthew J. Martinez. 2017. “The Spatial Scale and Spatial Configuration of Residential Settlement: Measuring Segregation in the Postbellum South.” American Journal of Sociology 123(4):1161–1203.

Summary: I wish every article was structured as cleanly as Logan and Martinez’. Without mincing words, the introduction sets out a clear three-tiered agenda while their conclusion reviews respective findings. Broadly, this article encourages a more careful operationalization of segregation. Using census data, mainstream research often reports low levels of residential segregation in post-bellum south. However, these aggregate summaries often miss lower-level detail and confuse geographic proximity for symbolic proximity. The authors use geo-coded street level data from the 1880s to show high levels of intra-block segregation. They also rely on qualitative accounts to understand the spatial configuration of residential life. In this way, we see new spaces like backyards, alleys, and side streets serving as the important symbolic boundary between an otherwise “integrated” city block. Researchers today should take this level of detail into account when possible, while also noting important features like mobility, nearest neighbor enumeration, and other qualitative factors that change the experience of geographic space.

Quote: “When calculated from data even as fine as census block counts, [distance-based measures] cannot capture common forms of segregation such as the alley or side street configuration.” p1195





Back to posts