Current and Future Directions

The results we present here are only a first step towards a data-based approach to understanding bias in Linguistics. Ultimately, we would like to use such results to come up with meaningful solutions for the issue.

We have been busy growing the network of people working on this project. In the early stages of the project, the work was presented at Michigan State University’s Undergraduate Research and Arts Forum 2017 and at the Michigan State Undergraduate Linguistics Conference 2017.

In the past few months, we presented and lead workshops at UMD’s Winter Storm 2018 and formed a working group made up of graduate students and faculty involved in UMD’s Language Science Center. We are presenting at the LSC’s lunch talk series, and are working to connect with other like-minded groups addressing gender and other biases in the field of linguistics.

Broadly, these are our big goals for this project:

  1. Conduct further descriptive work to identify where and why the bias exists, including looking at things like race, native language, sexual orientation, etc.
  2. Collectively come up with solutions to such problems of bias. We are particularly interested in working with departments and institutions to think about the policy implications of our data.

Here are some specific goals we are currently addressing:

  1. Are there differences in publishing rates? These can be counted in various ways: by number of publications, by first/last authorships, or by single-authored papers.
    1.  Do individual journals show gender biases? This may manifest in the amount of time it takes to get an article published or the proportion of acceptances vs. rejections, and is particularly important with regard to reviewer blindness.
    2. Do women or men publish more at similar stages in their career, particularly as graduate students?
  2. Are there disparities in the proportion of men/women in the audience vs. the proportion who ask questions? Work on this has been undertaken by several linguists (for example, here and here), so we are interested in creating an online app where people can go and enter in their own data to have it included in a database. The results will be presented on the webpage much like our current information is, with plots and data available for download.
  3. What are the differences in salaries at similar stages of the career at public institutions?
  4. What policies may mitigate the leaky pipeline (for example, regarding parental leave)? How can we measure the effects? How can we encourage schools to adopt these policies?
  5. Publicizing our results more!

We invite you to help contribute to these efforts. If you’d like to know more or get involved, please contact Bethany or sign up for our listserv!

2 Comments

  1. Borja Borja
    February 7, 2018    

    The word ‘bias’ has ‘prejudice’ and ‘injustice’ as part of its meaning. If you want to check whether there is a bias against some gender, race etc. what you need to prove is therefore that the group was treated unjustly. Simply saying that they represent a certain percentage of the total (even if the percentage declines as one goes up the hierarchy) does not prove this in my opinion.

    I would suggest to check whether merits (e.g. publications) and achievements (e.g. position) pattern in the same way in the two genders. That is, if women do need to publish more than men, on average, to reach tenure, for example, then there is bias. If they get promoted less simply because they publish less, then there is no bias.

    One could say there is discrimination more widely if women end up publishing less because they are burdened with other responsibilities (e.g. pregnancy, childcare etc.) but then the bias would not be in Linguistics or academia, which judging by the content of this website is what you are interested in. Although this is an unpopular thing to say these days, certain biases are biologically determined by the fact that women and not men get pregnant or by the fact that men can postpone paternity until a later age (after achieving their carreer goals).

    • February 13, 2018    

      Hi Borja, thanks for the comment and suggestion. Regarding publication rates, it looks like we were thinking the same thing. We’re actually currently in the process of collecting data on this, and we hope to have some results that we can share in the coming months. Though I disagree with your suggestion that women publishing less would be an indicator of there being no bias. If it’s true that women publish less, that suggests that there is bias, since, all things being equal, you would expect similar publishing rates between men and women in the field, controlling for stage of career and type of position.

      As for our use of the word ‘bias’, we think that a decline in the percentage of women in the field as one progresses up the academic food chain indicates that women and men are confronted with different obstacles and this deserves to be called ‘bias’. I think the baseline expectation is that, all things being equal, women and men should leave the field at the same rate. So whatever is pushing women out of the field at a greater rate is a source of concern that should be investigated. For example, we haven’t yet looked at this for linguistics specifically (though we plan to), but there is a pretty significant pay gap in academia, at least in the United States.

      Also, I’m not sure that I really understand your suggestion that pregnancy is a biologically determined bias. We agree that women are often disproportionately burdened with childbirth and childcare, and we are interested in all biases that affect the field of linguistics, including this one. If parenthood disproportionately affects women, this itself is a source of concern. If one truly values unbiased representation in the field—which we think should be valued—then one can make policies that promote that value. Women should not be penalized for having and raising children, and men should be encouraged to take greater responsibility in caregiving.

Leave a Reply

Your email address will not be published. Required fields are marked *