Studies in Social Issues, Individual Differences, New Brunswikian Resources
Research in the Brunswikian tradition continues at the University of Connecticut. An annotated bibliography of all published cue probability learning studies was prepared along with my contribution to The Essential Brunswik. Access to this Microsoft Word 6.0/95 document file (MCPL bib.doc) is available through a link on the Brunswik Society web page, or may be accessed directly at http://www.sp.uconn.edu/~holz/MCPLbib.doc.
My colleagues Steven Mellor, Dan O'Shea, and I are investigating judgments concerning becoming a replacement worker and crossing a picket line. Four situational cues (number of vacated positions filled, publicity for the strike, number of striking workers on the picket line, and threat of violence by striking workers) and an individual difference factor (financial need) were included in a multilevel judgment model.
Regression parameters from within-person judgment analyses indicated that situational factors did influence judgment polices about willingness to cross a picket line to accept a position. Using parameters as outcome variables, between-person analyses indicated that financial need did not impact on situational influences.
Two Brunswikian studies were presented as posters at the SJDM 1999 Meeting in Los Angeles. The first is a study of sexual harassment in the workplace. Lisa Kath, Carrie Bulger, and I compared court rulings with lay persons' judgments of sexual harassment. We identified important facts (cues) from actual court cases, and asked raters to judge impact levels (scale values) of each cue. Achieving consensus on cue level ratings proved difficult, so cues were coded dichotomously by two subject matter experts.
We conducted judgment analyses to capture individual policies concerning severity and pervasiveness of harassment cases and compared three methods of determining cue values (personal scale values, group mean scale values, and dichotomous).
We accounted for significant judgment variance, and the three methods of cue scaling produced roughly equivalent results. Research on sexual harassment will continue with new and different tasks.
The second poster concerns my attempts to relate biographical data (biodata) to styles of inductive reasoning. A biodata questionnaire was developed for assessing demographic and biographical variables potentially related to analytical and intuitive styles of cognition. The 195 items assess individual differences concerning exposure to science, math, and fine arts, tolerance for uncertainty, tolerance for ambiguity, and decision style.
Biodata questionnaires were completed by 378 students representing more than 34 majors within several colleges. Four criterion measures (essay items) known to induce different modes of cognition were also completed.
Results indicate individual differences in styles of cognition for the study participants. These individual differences were related to biodata variables.
At the Brunswik Society Meeting in Boulder this past summer, Tom Stewart and I presented a proposal for creation of a Social Judgment Data Network (SJDN). This SJDN will be sort of a co-op for judgment and decision making (J/DM) researchers. Researchers from around the world will be able to access the SJDN over the World Wide Web, borrowing task materials for research and instructional purposes. In turn, hopefully these researchers will contribute their data sets to the SJDN for use by other researchers and their students.
Tom and I are now working out details for recommended formatting and cataloging of task materials and data sets. Anyone interested in receiving a copy of our proposal, and helping with this effort, should contact one of us.