Lack of data to track scientific society membership demographic composition
Lack of data to track scientific society membership demographic composition
A long-term goal of ACCESS societies is for their PD programming for HEP to increase the diversity of their membership and ultimately their disciplines. However, without robust demographic data, ACCESS societies cannot assess the long-term effects of decades of diversity-focused programming to diversify their membership, and in turn, build a more inclusive STEM workforce. Low response rates on membership surveys is one factor that renders scientific societies currently unable to accurately collect data about the demographic composition of members. For example, at the end of 2019, 46% of ASPET members and 19% of ASCB members declined to answer, or provided no answer, to questions related to ethnic/race background (Segarra et al., 2020b). This incomplete data prevents scientific societies from accurately assessing demographic diversity in their membership at any given time, much less longitudinally.
Lack of demographic information impacts other aspects of scientific society dynamics. Recently, ACCESS showed that data on the demographic composition of conference speakers and award winners is often not collected, and therefore not available as a resource to assess inclusivity when considering who is and is not publicly recognized as a disciplinary thought leader (Segarra et al., 2020c). Organizations that have a larger scope in STEM, including the American Association for the Advancement of Science (AAAS), have reported similar challenges. For example, in October 2020, the AAAS reported in its Baseline Assessment of Demographic Representation in AAAS/Science Functions document that no data were available on race/ethnicity for 46.9% (N=13,480) of AAAS/Science associated memberships or for 87.8% (N=49,316) of the authors/reviewers in the Science family (AAAS, 2020).
Much of what the field knows about demographic survey nonresponse rates comes from studies focused on information related to income. Evidence shows that demographic survey nonresponse rates are higher for women, as well as individuals with low educational attainment (Riphaln and Serfling 2005; Lor et al., 2017). Moreover, higher nonresponse rates have been associated with individuals who identify as African American (Ross and Reynolds, 1996). Two different dimensions – cognitive and motivational – are thought to contribute to nonresponse in surveys (Beatty and Herman, 2002; Lor et al., 2017). The cognitive dimension includes an individual’s degree of knowledge (or perception of knowledge) about the subject. The motivational dimension includes an individual’s disposition to provide requested information. It has also been recognized that concerns about confidentiality can negatively affect demographic survey response rates (Singer et al., 1992). This might be particularly true for scientists from UR backgrounds if they have concerns that confidentiality may be compromised.
There are empirically-tested approaches, including increased categorical and messaging granularity, that yield higher response rates from individuals, including respondents from racial and ethnic minorities (Lor et al., 2017). One of the outcomes we hope to achieve through the LED-BIO Think Tank series is a set of community standards for demographic data collection, developed with input from experts in this area. We also plan to generate training activities to empower scientific societies to implement these more inclusive practices. These materials will include guidance on messaging as well as categories to include in demographic surveys, so that individuals from a wide range of backgrounds feel welcome and counted. Training materials will also highlight the cultural shifts needed in order for society members to realize that we can all contribute to inclusivity in our scientific CoPs through small actions, including disclosure of demographic information.
In addressing this challenge, we plan to discuss barriers, resources, and strategies regarding three major sub-questions:
Discussion Topic #1: What kind of data should professional societies be collecting about their membership? How should this information be managed?
Barriers: Clearly there are reasons why scientific professional societies do not collect demographic data on their membership. We hope to identify some of these reasons by asking Town Hall participants to discuss their own experiences with attempting to collect membership demographic data or being asked to provide such data about themselves.
Resources: Successful data collection requires an input of resources. Professional society leadership should be aware of what is needed to carry through on a commitment to track the demographic makeup of society membership, including what is needed to satisfy any concerns members have about the confidentiality of information they provide. It is also crucial to consider how to leverage activities that are currently established to facilitate data collection.
Strategies: A sure way to ensure that demographic data collection is successful is to ensure that the types of data collected and the mechanisms used are acceptable to their membership. Inclusive data collection should help all members feel that they are being seen and that their sense of belonging is important to professional society leadership. We will utilize both the lived experiences of our Town Hall participants and the academic expertise of our Think Tank participants to identify practices that can help achieve this goal
Discussion Topic #2: How should professional societies make use of membership demographic data? What uses are inappropriate?
Barriers: There are certainly legal and ethical issues that must be considered regarding the use of demographic data. Although United States federal law (Title VII) prohibits discrimination on the basis of race, color, religion, national origin, or sex, the question of what kinds of actions in reality constitute discrimination is a thorny one that is still being debated in litigation brought before judges up to and including the Supreme Court. The fear of winding up on the wrong side of the law may prevent any organization from taking action on any demographic data in their possession. On the other hand, fear of having personally identifiable information broadcast for all to see may be a reason that members of HEP groups often decline to respond to demographic surveys, and these concerns must be addressed. Any organization that plans to collect and act on demographic data must consider legal and ethical limits, both real and perceived.
Resources: Under regulations set by the United States Food and Drug Authority, all human subjects research is overseen by an Institutional Review Board. While membership demographic data collection and implementation of programming in response to the results may not fall under the category of human subjects research, the principles that govern appropriate design and implementation of human subjects research protocols can be informative. Likewise, attorneys who specialize in workplace and employment discrimination regularly advise businesses on their practices even before lawsuits are filed. There are likely other resources that professional society leadership can make use of to guide their appropriate use of membership demographic data. We hope to identify some of these through our Think Tank and Town Hall discussions.
Strategies: Scientific societies should only be using demographic data in ways that align with the mission of their organization and the values of their members. We will ask Think Tank participants to help us identify the most appropriate uses of demographic data toward the goal of increasing equity and inclusion in professional societies specifically and STEM in general. We will also ask our Town Hall participants to share what is most important to them regarding the use of membership demographic data in decision making.
Discussion Topic #3: How can professional societies increase response rates when carrying out demographic surveys of their membership?
Barriers: Fear of identification, negative emotions around being labeled, and a perception the collection of demographic will not lead to lasting change are all potential reasons so many demographic and climate surveys see such low response rates. We hope to identify others through Town Hall discussions and to draw upon the expertise of our Think Tank participants to determine the major drivers that make people decline to disclose their demographic information.
Resources: There are known strategies that increase survey completion. Some are centered on the design of the survey. Others may have to do with the messaging that professional society members receive regarding the motivation for collecting the data. We hope that our discussions will help uncover additional resources and ideas for how to shift members' responses to demographic surveys in both the cognitive and the motivational dimensions.
Strategies: We will ask Town Hall participants to share what would motivate them to complete a demographic survey. We will also attempt to gather promising practices with the help of our Think Tank participants, many of whom collect and work with sensitive data regularly.