Creation and Development of the Quantitative Subtle Gender Bias Index (SGBI) to Better Understand Micro-inequities within the Academic Setting Participants
Because UMass Lowell has 203 and UMass Worcester has 129 eligible women and men STEM faculty, purposeful sampling was utilized in obtaining 19 total interviewees. Nine STEM faculty women (Lowell - 4, Worcester - 5) were interviewed in the pilot study in 2011. Ten STEM faculty women (Lowell - 5, Worcester -5) were interviewed in summer and fall of 2013. Participants all possessed:
- a current faculty affiliation at either UMass Lowell or UMass Worchester in a STEM department,
- had a Ph.D. or equivalent degree, and
- had worked as faculty (e.g., Assistant, Associate, Professor) at their respective institution for at least 6 months.
- Interviews - Each 60-minute in-depth semi-structured interview was recorded, transcribed, and de-identified by interviewers unaffiliated with either UMASS campus.
- Coding - Each interview was coded according to thematic conceptualizations of bias and micro-inequities based on recommendations within qualitative research methodology (Miles & Huberman, 1994). Word documents and NVivo 10 were used to code qualitative interviews. Specifically, the team of 4 psychologists began by coding one interview, discussing codes, and developing a shared understanding of the codes to be used in the interview. After achieving at least 80% consistency in codes in the first 2 interviews, the rest of the interviews were coded in pairs. Each interview was independently by 2 coders. Coding inconsistencies were discussed until a consensus was achieved for all codes.
- Item Development - Coded segments were then extracted for discussion among the larger interview team until consensus was reached regarding each coded item. Items were then organized into charts which were used to create survey items that captured the essence of each interview statement or segment. Once items were created they again entered the discussion process were the larger team reviewed each item for accuracy, wording, and comprehension. Items were then categorized into domains, subdomains, and corresponding literature domains. Once repetitive and redundant items were removed from each item set, the full set of items were sent for consultation with the External Advisory Committee.
- Expert Consultation - Items were then shared with the larger research team of scholars affiliated with the UML Center for Women and Work who are national experts on gender issues in the workplace. These experts provided feedback to the items developed by the research team. Revisions have been made to refine items. Literature searches were also conducted to gather existing empirical research and scales focused on subtle gender bias to ensure the current measure includes all necessary domains. The next step in the consultation phase will involve 4 STEM faculty women (Lowell = 2, Worcester = 2) to participate in a Read Aloud Pilot Study in which the participant provides feedback on the items for clarity. Accuracy of interpretation will also be assessed by having participants describe how they are interpreting the items as they read through the measure. The Read Aloud Pilot Study will be complete by the end of June.
- Next Steps - The last year of the grant is devoted to quantitative validation of the SGBI measure. The finalized items will be uploaded to Qualtrics along with other measures intended to assess the validity of the SGBI. Active recruitment of academic women across disciplines and the country will take place in Fall 2014. Statistical analysis will begin in Spring 2015.