What constitutes a declarable conflict of interest in gender studies?

Usually, these relationships are financial.

For example:

• In a paper talking about how Twitter has failed to police harassment, a professor who's an investor in an early-stage social network competitor would have to declare that.

• Similarly, board memberships, speaking gigs, etc. are things that would have to be declared. For example, while writing a paper about abortion access rights, a professor should disclose if she was a paid speaker for a pro-choice advocacy group.

Generally, "I really care about this" for personal reasons doesn't so much count as a conflict of interest. I cannot imagine, for example, a journal intending for those writing in gender studies to disclose if they had been victims of intimate partner violence.

As a matter of nomenclature, I would say that a conflict of interest is related to material gain (aka financial gains) that hinge on the results of your work going a certain way. @Fomite does a good job of describing this.

What you're talking about is probably better described as bias. All researchers have some bias for their own work, this is normal and expected. For example, at the low-end you have "I spent two years on this work and don't want it to go to waste", and at the high end you have "I have a deep personal connection to this work" or even "I really think this is the right theory, even if I can't quite find the right evidence to support it". Having a strong connection to your work is a good thing! As long as it doesn't affect your objectivity...

Conflicts of interest are easy to identify and disclose (do you own shares of a competing company? Yes / No?), while biases are more subjective and much harder to spot - and in many cases would be silly or a violation of personal privacy to declare. Good researchers / authors are expected to recognize and control their own bias in their writing, but sometimes it leaks through and this is where editorial review and critical reading come in - don't take everything in a paper at face value!

You are confused between a conflict of interest and bias. A conflict of interest is when it is in your personal best interest if the results are interpreted in a particular way, regardless of whether that way is correct or not or supported by the data you have gathered. As others have said, that is most frequently financial, but there are other consididerations. Bias is more or less having an opinion.

What would non-financial conflict of interest be? Say you were doing a study on recidivism rates of convicts that earn degrees while in prison, at the same time you were about to come up for parole...clearly it is better for you personally for people to believe that you are safe to let out.

Bias would be if your childhood best friend was studying for his degree, while you do the above study -- you may unintentionaly bias the data in favor of convict-students, but presumably there's no danger of you deliberately corrupting the data to get the result you want.

You are expected to recognize your own biases and guard against them unduly influencing your results. Bias is too pervasive and too nebulous to require reporting, there is simply no end to the things that could conceivably cause you to lean one way or another when gathering and interpreting the data.

A conflict of interest is much more concrete, and you are required to disclose a potential conflict of interest so others can judge whether you are cooking the books and/or are likely to be doing so. If you don't disclose, and the conflict is discovered, the assumption is going to be that you were.