ASA Connect

 View Only
Back to eGroups
Expand all | Collapse all

phrasing of sex/gender questions in surveys

  • 1.  phrasing of sex/gender questions in surveys

    Posted 07-08-2020 12:32

    Hi -

    Does anyone have any recommendations or references for how to best ask sex/gender related questions in surveys?  I am thinking about two different applications:

    1. Self-report to assess demographics captured.
    2. Where physicians report on a fictitious patient to assess the role of sex/gender in treatment decisions.  

     

    Thank you!

    Shannon



    ------------------------------
    Shannon Pileggi
    ------------------------------


  • 2.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 08:49
    How about "What is your gender identity?"

    1. Female
    2. Male
    3. I identify in another way.

    Comments (optional) ___________________________

    This seems like a respectful way of asking a respondent about their gender.

    ------------------------------
    Brandy Sinco, BS, MA, MS
    Statistician Senior
    Michigan Medicine
    ------------------------------

    Original Message


  • 3.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 22:08
    Thank you for the suggestion! However, from what I have read this can conflate the concepts of sex vs gender (biological sex being male/female; gender being masculine/feminine or man/woman), so I want to be sure I am very careful about this!

    ------------------------------
    Shannon Pileggi
    Assistant Professor
    ------------------------------

    Original Message


  • 4.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 09:02
    Shannon,

    I'd suggest you reach out on questions regarding SOGI (sexual orientation and gender identity) question wording to our ASA LGBTQ+ Advocacy Committee. The chair of the committee is Jack Miller, whose email can be found on the ASA website.

    ------------------------------
    David Marker
    Senior Statistician
    Westat
    ------------------------------

    Original Message


  • 5.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 22:10
    Edited by Shannon Pileggi 07-09-2020 22:26
    Thank you for the suggestion!

    ------------------------------
    Shannon Pileggi
    ------------------------------

    Original Message


  • 6.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 10:34
    One good example I have seen is used in the All of Us research project, which uses a nice two-question format with flexible response options that acknowledge the diversity of both sex and gender in humans; see "The Basics" at https://www.researchallofus.org/survey-explorer/.

    ------------------------------
    Lauren Samuels
    Research Assistant Professor
    Vanderbilt University School of Medicine
    ------------------------------

    Original Message


  • 7.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 22:12
    Thank you for the suggestion!  This looks like an all around fantastic resource.

    ------------------------------
    Shannon Pileggi
    ------------------------------

    Original Message


  • 8.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 11:42
    Shannon,

    the community that specifically deals with all aspects of survey design, including questionnaire design, is the Survey Research Methods Section. If you deal with issues of this nature regularly, you would want to join and ask questions in that group.

    I am not a questionnaire design expert, but I kinda know where to look that stuff up.

    1. Some of the Census work: https://www.census.gov/content/dam/Census/newsroom/press-kits/2018/jsm/jsm-presentation-gender-identity-cps.pdf, https://www.census.gov/content/dam/Census/library/working-papers/2018/adrm/rsm2018-06.pdf
    2. Some work by the National Center for Health Statistics: http://www.aapor.org/AAPOR_Main/media/AnnualMeetingProceedings/2015/B5-1-Miller.pdf
    3. A fairly comprehensive AAPOR presentation: http://www.aapor.org/AAPOR_Main/media/AnnualMeetingProceedings/2015/A5-3-Michaels.pdf
    4. Survey Monkey (they have excellent survey methodologists who work very hard to make it as difficult as possible for the end users to do stupid things with their surveys... they can't edit the users' questions for them, but they provide as much guidance as the existing literature allows): https://www.surveymonkey.com/curiosity/ask-survey-questions-sexual-orientation-gender-identity/

    Copying and pasting one of the slides from #3 -- basically the thoughts for you to REFORMULATE your question which is too broad as you asked it:

    • Conceptualization: What are we trying to measure?
    • Sexual orientation (LGB), Gender Identity (T), Sexual minorities, etc.
    • Sexual orientation: Behavior, Attraction, Identity
    • Gender identity: Disjunction sex/gender; Transgender identity, gender expression
    • Measurement: Can we believe respondents' answers to sensitive questions?
    • Is sexual orientation still a sensitive question?
    • Sources of error: measurement/response error (mode/context)
    • What are the research questions?
    • Prevalence (demography) of homosexuality
    • Health disparities, negative health outcomes
    • Economic & other disparities​


    ------------------------------
    Stanislav Kolenikov
    Principal Scientist
    Abt Associates
    ------------------------------

    Original Message


  • 9.  RE: phrasing of sex/gender questions in surveys

    Posted 07-09-2020 22:24
    Thatnk for the suggestion as well as the comprehensive list of resources!

    ------------------------------
    Shannon Pileggi
    ------------------------------

    Original Message


  • 10.  RE: phrasing of sex/gender questions in surveys

    Posted 07-10-2020 07:16

    Thanks to Shannon Pileggi for the question and everyone for their replies, especially to Stanislav Kolenikov for the citations. I have recently begun to serve on the Data Users Advisory Committee for the Bureau of Labor Statistics and am very interested in addressing this very question in that area. The comments and references given here will assist that work.

    Please allow me to make two comments, one technical and one ethical:

    1. As there are different uses of gender, an understanding of how the data will be used could affect the phrasing of the question to solicit answers consistent with the purpose. For example, in some circumstances, biological gender may be preferred, what ever the person's identify. For example, the survey administered to blood donors by the Red Cross asks women the number of past pregnancies; this question is not administered to men. In other surveys, identity rather than XX / XY / others biology is required. Different surveys could ask the question differently depending on how they data are to be used.    
    2. Ethically, I will suggest survey questions be parsimonious: not collecting any fields not required. This is especially true of factors related to identity. While gender may be asked pro forma, because past surveys always asked for it, this is not an adequate reason. Survey questions should be asked with the survey design in mind, and every question should have a purpose. Without a clear, a priori reason to collect identify information such as gender, we should not collect it. I am most certainly not asking people to refrain from collecting gender in surveys: I am asking people to consider the design of the survey at hand and have a clear purpose for the questions it contains. We collect data because it is needed and have a plan for how it will be used, not merely because we have always collected it in the past.  
    Disclaimer: these comments are those of the author only and expressly do not represent the views of any company, agency, organization, or other entity.

    ------------------------------
    David J Corliss, PhD
    Director, Peace-Work www.peace-work.org
    davidjcorliss@peace-work.org
    ------------------------------

    Original Message


  • 11.  RE: phrasing of sex/gender questions in surveys

    Posted 07-10-2020 12:08
    Hi Shannon,

    I think it is first important for researchers to have a good reason to ask about sex or gender on a survey, such that these variables may moderate an effect.  I think at a minimum, an "other" option should be provided with a fill-in-the-blank text box to be more inclusive.  If a researcher is focusing on a gender\sexual minority group, then the researcher may want to have more than just the "other" option. For instance, some literature (see the manuscript from the Williams Institute below) has suggested that asking about sex assigned at birth and current gender identity is a more reliable way to measure cis and transgender identification than just using the terms cis and trans gender (since those terms are not widely understood by the general public). Since many people use the terms sex and gender interchangeably, you could provide definitions of such terms in a survey to make it clear to the participant. 

    These resources may be useful to you:

    1. The Federal Committee on Statistical Methodology has some useful documents on measuring SOGI: https://nces.ed.gov/FCSM/SOGI.asp
    2. An article titled: Transgender-inclusive measures of sex/gender for population surveys: Mixed-methods evaluation and recommendations
    3. This manuscript from the Williams institute on Best Practices for Asking Questions to Identify Transgender and Other Gender Minority Respondents on Population-Based Surveys


    ------------------------------
    --------------------------------
    Stephen Parry
    Statistical Consultant
    Cornell Statistical Consulting Unit
    Cornell University
    --------------------------------
    ------------------------------

    Original Message


  • 12.  RE: phrasing of sex/gender questions in surveys

    Posted 07-10-2020 16:10
    Hi, I just wanted to mention that the Resource Center for Minority Data at ICPSR has an inventory of the questions used in surveys in our collection that address sexual orientation and gender identity (not yet complete).

    ------------------------------
    Margaret Levenstein
    ------------------------------

    Original Message


Related Content