- Women are 26% more likely to feel unwelcome, discriminated against, or harmed in the secular movement.
- Participants get the most blame from women for problematic behavior and unwanted advances. Still, unwanted advances by group leaders and employees are claimed by more than 15% of the women who indicate they’ve had negative experiences.
- When asked about the disadvantages of participating in secular organizations, “problems within the group itself” are cited 157% more often by women than overall.
A commenter by the name of C. Folk responded that because participants self-select, the data may not be representative of the population, to which Zvan claimed that participation in surveys is always self-selecting to some extent.
Presumably addressing C. Folk’s criticism, the American Secular Census clarified that the census is “not necessarily the most scientific way to delve into the details of sexual harassment at secular events,” though the “data holds a great deal of potential for deeper investigation of a demographic or a viewpoint.”
In the discussion that followed between C. Folk and Zvan, I noticed the use of the term hyper-skepticism. Zvan, not accepting C. Folk’s criticisms wrote the following:
… would you like to show me where you lecture people with degrees in psychology who write about scientific methodology who have just told you about self-selection in survey data–about self-selection in survey data?
I ask because what you’re doing here is swerving into the realm of hyper-skepticism, and I suspect you want to give that some thought.
C. Folk noted that “pointing out the limitations of the ASC data is not hyper-skepticism” and then adopted a deferential attitude and removed himself from the discussion.
Before I discuss hyper-skepticism, was C. Folk correct to point out the limitations of the ASC data? External validity is always an issue when selecting and sampling, but that doesn’t mean that a random sample is just as compromised as a nonrandom sample. Suggesting otherwise, as Zvan did, is embarrassing.
Surveys have levels of bias, but they can produce generalizable results if care is taken to minimize error and adjust for potential bias. I am not suggesting that the data produced by nonprobabilistic sampling methods are useless, but I urge extreme caution about making inferences from non-probabilistic samples.
I also urge caution with respect to the use of the term hyper-skepticism. Jason Thibeault writes that there is a “subset of skeptics that are skeptical well past reasonable levels.” These hyper-skeptics will “pore over every detail of a paper or news article hoping to find (or spin) some part of it into something supporting their overarching worldview that the paper doesn’t actually do by itself.”
Alonzo Fyfe argues that the hyper-skeptic’s goal is to “block progress – to be an obstructionist – and he uses skepticism as a tool for building obstructions.” This type of skepticism is not a virtue, but is instead a “form of manipulation, much like lying” and those who use this technique “seek to hijack the will of others and to take the products of their efforts for one’s own, by manipulating them into acting in ways that are not in their benefit.”
I accept these definitions and the problems associated with hyper-skepticism. I also accept that sexism and misogyny are enduring and common occurrences that permeate our culture. And, just so there’s no confusion, I support the adoption of human rights, anti-harassment, and non-discrimination policies.
Here’s the problem: I don’t think this ongoing controversy is just about sexism, misogyny, and the implementation of anti-harassment policies. Zvan’s response to C. Folk is a case-study in why some people in the atheist and skeptic communities are opposed to Rebecca Watson and her supporters’ framing of this “debate.”
C. Folk’s caution regarding the ASC data’s representativeness is an appropriate and reasonable response. Even if he was wrong, it would still be an appropriate and reasonable response because he has the right to be wrong without conjuring the specter of hyper-skepticism. One can be wrong without being a hyper-skeptic. It just so happens that he was right.
How do we distinguish between hyper-skepticism and more virtuous forms of skepticism? How do we avoid dismissing people like C. Folk as hyper-skeptics?
Whether Watson and her supporters agree with the perception or not, there is a perception that certain prominent members of the community, in their efforts to support the worthy goal of eliminating sexism, misogyny, and harassment from conferences and other gatherings, make exaggerated claims and have a bias against disconfirmatory evidence. Given my last few posts about this issue, it’s fair to say that I share this perception, while also being in favor of anti-harassment policies.
I can consider myself a pro-feminist without buying into every facet of feminist epistemology. I can recognize problems regarding the validity and accuracy of women’s subjective feelings of empowerment or loss of empowerment without dismissing the value of anecdotes. I can also think that sexism is a problem in the community without seeing it as ubiquitous. Most of you can do the same without experiencing cognitive dissonance.
Fyfe has a suggestion for how to deal with genuine hyper-skepticism:
This form of skeptic needs to be met with the claim that, ‘We have information enough on which to make a decision. Unless you can come up with positive reason to believe that we should not commit the act, then it is perfectly reasonable for us to act.’
We do have enough information that sexual harassment and sexism exist, even if we disagree about the particulars. I think it is reasonable for us to act, even if there isn’t much data about the prevalence of inappropriate gendered behavior within the atheist and skeptic communities themselves. However, that does not mean that we should stifle debate or help stir a moral panic.