Looking Back at White Working-Class Racial Attitudes and Behaviors in the Late 1940s: An Updated Analysis

ABSTRACT

We re-analyze quantitative and qualitative data from a study of a white, working-class neighborhood in mid-twentieth century Chicago to pursue two research goals. First, we test how attitudes toward blacks as coworkers and neighbors are correlated, and determine whether those attitudes are associated with commitment to the union and neighborhood attachment. Second, we address three timely issues in urban sociology and race relations: (1) how community efficacy is related to attitudes toward racial exclusion, (2) whether white homeowners are likely to oppose racial minorities as neighbors due to fears over property value decline, and (3) the “whitening” process by which people of southern and eastern European ancestry came to embrace a white racial identity.

KEYWORDS:

• Community and urban sociology
• social psychology
• racial and ethnic minorities

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Arnold Hirsch (1983) mentions that South Deering residents wanted to create a “white” organization that would advocate their interests like the NAACP supported those of blacks, and they used the term “white” more often than ethnic labels like “Polish,” “Serbian,” or “Italian.” But most of Hirsch’s evidence on South Deering came from local newspaper articles and third-party reports, and is drawn from years during and after the Trumbull Park (public housing project) crisis over racial integration (mid-1950s). Reitzes’s data on Grainville are from the late 1940s and come from personal interviews with residents.

2. The president of Local 1 helped identify 102 union members living in Grainville. He sent letters on union stationary (signed by him) requesting their cooperation in the study. These contacts yielded 80 completed interviews. A union steward provided names and addresses of 51 members of Local 2 living in Grainville, which generated 38 completed interviews. A supplementary list of 59 names and addresses led to 33 additional completed interviews.

3. This conclusion runs a risk of making a type II error (accepting a “no difference” or .00 correlation result, when actually a significant correlation exists between these two attitudes). A widely accepted standard for guarding against type II error is that the statistical test’s power should be .80 or higher power (Cohen 1988, 1992). With a sample size of 150, and if there is actually a “medium” (.30) or “strong” (> .50) correlation (“effect size”), the power of our statistical test is over .95, which is far above the conventional standard (.80). If a weaker actual correlation exists, the power is lower: .87 for an actual correlation of .25, and .70 for an actual correlation of .20. We think, therefore, that the risk of a type II error here is acceptably small, especially for non-trivial correlations that would be substantively or practically important.

4. Letters and numbers in parentheses refer to interviewee’s case number. Letters identify the interviewer, first number identifies interviewee, and number after colon is the page in Reitzes’s dissertation.

Additional information

Notes on contributors

Donald C. Reitzes

Donald C. Reitzes is a Professor of Sociology at Georgia State University. His research interests are in symbolic interaction theory, identity, and urban social psychology.

Charles Jaret

Charles Jaret is a professor emeritus in the Georgia State University Department of Sociology. His research interests are in urban sociology, migration, and race/ethnicity.

Ashley Rockwell

Ashley Rockwell is a doctoral student in the Georgia State University Department of Sociology. Her research interests center on inequality, gender, work, race, empathy, and social psychology.