21 June, 2016

A word (or many) about the survey

Since we've opened the new Queer in STEM survey yesterday, we've been overwhelmed by the initial response. Thousands of people have visited the study website, and hundreds of folks have answered the survey already. We've also had folks raise concerns with a number of specific questions on the survey — either their wording, or the nature of their enquiry, or their suitability for different identities, populations, and career paths. We appreciate this feedback, and will do everything we can to take concerns into consideration and make decisions that allow us to accommodate unforeseen issues but also ensure validity of the survey data. We also want to provide the following more general statement:

First, we acknowledge that a survey is a blunt instrument to examine issues as complex as genders and sexualities. This is why the Queer in STEM project overall employs a mixed methods design, with the survey as stage one of data collection, followed by an open-response questionnaire and individual interviews. These modes allow for participants to provide more detailed explanations of their experiences and identities. We also know that responding to questions that ask about personal identities can cause discomfort, and we apologize for any distress that was caused by our survey.

We selected terms for questions about identity that aim to distinguish biological sex, gender identity, gender expression, and sexual orientation as separately defined categories, in keeping with other research in the field of gender and sexuality studies. The goal in separating these items is not to provide additional scrutiny for people's identities, but rather to avoid conflating categories. The original 2013 survey asked only about gender identity and sexual orientation; we received many requests to add the other categories to more accurately capture trans, nonbinary, and nonconforming identities, and to include intersex identities. We also responded to many requests from members of the Asexuality Visibility Education Network to allow for reporting of sexual v. romantic attraction.

The use of data collection techniques such as surveys are heavily debated in social science fields, particularly for populations who have been marginalized or undercounted. In conducting this study, we have decided to err on the side of trying to reach out to as many people as possible, while acknowledging the limitations of generalizability and the difficulty in capturing nuanced and complex categories of identity. This is why we allow respondents to select multiple options in lists of choices, and include "fill in other term" options for identity categories. We know that for many people identities are fluid and do not fit easily into existing categories, and that answering questions that appear to force a choice to define oneself can be extremely frustrating.

We spent about six months drafting the wording of the new survey, and went through several rounds of piloting and member checking with colleagues, particularly aiming to capture identities and perspectives not shared by members of our research team. Different iterations were tested and the final version was reviewed by several people, including individuals who identity as trans. We also complied with Institutional Review Board requests to adjust certain items, removed several questions that we deemed as better answered via interview, and followed ethical guidelines for social science research.

This said, we are by no means infallible. As with any new research project, mistakes are likely and we assume we will make them—we strive to be open to criticism and responsive to suggestions. About 10% of the original study participants identified as trans, nonbinary or gender nonconforming; this is significantly higher than much research that claims to be "LGBTQ" but really only captures LGB perspectives. Many researchers have dismissed trans identities in survey research as "too complex"—we reject this approach as exclusive, but entirely acknowledge the complexity of identities and do not intend to oversimplify.

Our sincere goal is to contribute to the visibility of queer (used as an umbrella term for a diversity of genders and sexualities) individuals in STEM fields, who have traditionally been ignored at best, or stigmatized and discriminated against at worst. There are many ongoing advocacy efforts in particular fields (such as LGBT+ Physicists) and organizations such as oSTEM and NOGLSTP that have been engaged in such work for many years. Our goal is to support, not supplant, these efforts.