Grad Student Dismissed

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
DynamicDidactic

DynamicDidactic

Still Kickin'
10+ Year Member
Joined
Jul 27, 2010
Messages
1,846
Reaction score
1,564
DynamicDidactic said:
IHE article, in case anyone is interested
www.insidehighered.com

Clark dismisses graduate student who complained about possible gender discrimination and research misconduct

Clark University dismisses a graduate student who complained about possible gender discrimination and research misconduct, saying she couldn't find a new adviser within 30 days.
www.insidehighered.com www.insidehighered.com
Click to expand...
Would it have made any difference if the whistleblowing student was a dude? From reading the artcle (assuming veracity of what was claimed), the scenario seems to be: grad student takes issue with post-hoc manipulation of coding scheme to turn non-results into 'results' and then prof 'retaliates.' The claim that the prof calling her 'defensive' constitutes 'gender/sexual' discrimination (because no one ever calls guys 'defensive,' right) seems odd and out of place.
 
I'm in clinic so just scanned it briefly. It sounds like crappy science was being done, but (sadly) not anything remotely out of the ordinary. It sounds like some crappy mentoring was being done. That did seem out of the ordinary, even for crappy mentors.

I think the case for gender/sexual discrimination is going to be unbelievably hard to prove based off what is written. I literally didn't see anything that suggested gender played a role in a meaningful way, though there could well be other things not in the article. She may have a better case as a whistleblower, though I don't know the intricacies of how those are treated in academic settings. Unfortunately, I'm not sure that being a bad advisor/department is necessarily something the legal system can address. Feel bad for the student...
 
Fan_of_Meehl said:
grad student takes issue with post-hoc manipulation of coding scheme to turn non-results into 'results' and then prof 'retaliates.'
Click to expand...
I'd change it to:
grad student takes issue with post-hoc manipulation of coding scheme to turn non-results into 'results' and then humiliates prof who then doesn't want to work with student and student can't find a new faculty advisor.

Unfortunately, this sort of post hoc manipulation is the standard for many researchers. Looks like this student is learning the hard way that changing the establishment is very difficult.
 
DynamicDidactic said:
I'd change it to:
grad student takes issue with post-hoc manipulation of coding scheme to turn non-results into 'results' and then humiliates prof who then doesn't want to work with student and student can't find a new faculty advisor.

Unfortunately, this sort of post hoc manipulation is the standard for many researchers. Looks like this student is learning the hard way that changing the establishment is very difficult.
Click to expand...

I think "humiliates" is a little harsh, lol -- If the methodology is defensible, then it should be able to withstand scrutiny.

If a student is dismissed for perceived "humiliation" of a professor (possibly without due process; e.g., remediation?), then that strikes me as retaliatory and possibly capricious.
 
I've been following this on twitter, too, and I'm torn, but I generally feel very bad for her.

First, I thought the gender harassment she experienced would be more than being called "defensive" (I had not context for this until this article... I don't think she's elaborated about it on twitter). This is a complaint I've heard levied against men and women, though perhaps women experience it more. Nonetheless, that in itself probably does not constitute harassment from a legal perspective, as mentioned by others.

In terms of the p-hacking that is alleged, I thought from her own twitter account that it was something more malicious. It sounds like we are talking about a Spearman rho or analysis that is measuring a simple relationship between orientation on a Likert scale (not the best method in general for complex things like sexual orientation) and harassment. Depending on what was actually wound up publishing, I do not think it is unethical to say "the scale, originally as written listed 1 as heterosexual, 5 as homosexual, and the rest as bisexual. When we ran a model like this, nothing was significant. However, when running the model such that 1/2 were heterosexual/bisexual leaning heterosexual, 3 as truly bisexual, and 4/5 as homosexual or bisexual leaning homosexual, results were significant." If it's not stated exactly like that, then in my opinion she is correct in that the professor is doing a post-hoc manipulation of the scale and committing research fraud. It sounds like from his email, though, that the professor basically did what I said regarding testing for changing in the way the scale was used, but reporting both. We can debate this post-hoc change, but it's par for the course, especially in social psych.

I think Clark is at fault if they did not follow the process of first putting her on probation, but instead unilaterally dismissing her from the program after 30 days (+21 additional days); the length of review, however, seemed thorough. Furthermore, disagreeing with someone about their approach to a statistical analysis, especially if it is blatantly wrong, should not warrant in dismissal, and her whistleblowing needs to be protected. This is honestly such a mess, and there are factors I am sure she is not tweeting about or in this article that are at place (e.g., the entire side of her mentor). I am also surprised nobody else would take her on as a mentee...... it sounds like the Clark psych department and the whole university is really circling their wagons here.

This is just such a mess... as a grad student right now, I am particularly sensitive to the power misbalance at play here. 🙁
 
From the article, it sounds like she publicly accused him of "research misconduct" in front of his colleagues. I think that goes beyond critical analysis of research methods and shows that she has no understanding of the nature of mentor relationships in the field. I can see why no one else would want to work with her.

Edit to add: Obviously if there are real concerns of research misconduct a student should speak up about them. But I think it's common sense to phrase it as "I have concerns about how post-hoc recoding of this data will affect the validity of the statistical significance" or whatever, and then discuss how to address that in the write-up, rather than jumping straight to accusing others of fraud. If I were in that situation, and felt that my concerns were being inappropriately dismissed, I would start with discrete conversations with other department leaders before making public accusations. Pointing fingers in public and posting about it on Twitter before it's been resolved within the research team, all before lining up a new advisor, seems like shooting yourself in the foot and really makes me question her judgment.
 
Last edited:
jdawg2017 said:
I am also surprised nobody else would take her on as a mentee......
Click to expand...
I've seen students and mentors not get along. I've seen numerous students change mentors after their master's. I've seen students change mentors within the same department but in a different program. In my experience, small sample, I've never seen a student not find a new mentor at all. This seems very telling of both the faculty and the student.
 
I'll probably get slammed for this but I can't be the only one thinking of the ouroboros as a metaphor here with the mythical snake of grievance studies eating it's own (tail).

Also, not taking the side of the prof (without more info) but, what if an actual male psychology supervisor has a female supervisee who is--you know--*actually* exhibiting problematic (say, 'defensive') behavior and, say, that same supervisor is being tasked with giving that feedback via an appropriate formal channel (like a supervisory rating form)? From the original article and some of the perspectives here it seems like there may be a de facto assumption that a male supervisor labeling a female supervisee's behavior 'defensive' is potentially 'sexual harrassment' merely due to the makeup of the dyad (male/female)? And we wonder why accurate feedback/gatekkeeping isn't always responsibly practiced?
 
DynamicDidactic said:
how about embarrasses?
Click to expand...

Maybe, but I guess I prefer criticizes? Or even critiques?

Embarrass or humiliate would seem more appropriate to me if the feedback provided by the student was super ad hominem -- We don't say that faculty are embarrassing or humiliating students when they voice methodological concerns during a dissertation defense, so I don't see why we would change the language in this scenario. To insinuate that faculty are unique from students in having to defend their methodologies, even if the request for a defense is made by a student, seems fallacious to me.

I can't recall if the student even shared these concerns with anyone other than her advisor at first -- It sounds like her first course of action was to apply elsewhere for graduate school, and her advisor "embarrassed" her by allegedly writing terrible letters of recommendation.

FWIW, my department ensures that all trainees are guaranteed an advisor regardless of politics, so long as the trainee adheres to all policy carefully outlined within the department's handbook. IMO, poor advisor fit is not an excuse for unilateral expulsion from a program.
 
Last edited:
I didn't get the impression he would have reported the other way of running the analysis. Its really not good to do, but its also par for the course. The line is also super-fuzzy. Doesn't sound like it was the case here, but what if he had collapsed them because of how the distributions lined up? What if another paper had just come out doing it a different way?

These things are easy to pick at for likert scales just because its simple/obvious and I think is the main reason we're getting more replication crisis there, but believe me they are 100x worse in neuroscience and other areas. I encounter dozens of choice points in even the simplest MRI analysis. I try to strive for model fit and accounting for all sources of variance over confirming hypotheses, but this is messy messy stuff.

If this was meant as a call-out/public twitter-shaming, I can see why no one would want to work with her. Its becoming a part of our culture and that is especially true among the younger cohorts now in college/grad school, but is not widely accepted. That said, I am in no position to say with any degree of certainty who was more out of line here. It does sound like the school was overly quick to usher her out the door, which doesn't look good for them. Writing bad letters of rec is tricky. We don't know what it said. It could be "slightly less glowing then a current-advisor recommendation should be because someone who admitted to personal problems wrote it in a rush" bad or "this student is an evil witch out to destroy my career" bad. That said, if the former and the 30 days is an institutional policy and has been enforced in the past...she may not have a leg to stand on.

Its still weird. And fishy.
 
Last edited:
acclivity said:
To insinuate that faculty are unique from students in having to defend their methodologies, even if the request for a defense is made by a student, seems fallacious to me.
Click to expand...
I guess I can say it more directly, without insinuating, traditionally it is the role of the faculty to question their students' methods rather than the opposite. I am not saying this is correct (conversely, I believe it should be a two-way road). But I can understand how a student's critique of their mentor's methodology, in front of the mentor's colleagues (this is how I read the article) can be embarrassing (one does not need ad hominem attacks to feel embarrassed or humiliated).

I think most (if not all) people on this forum had to learn how to navigate the interpersonal waters of grad school (and internship and post doc). I am highlighting that this student did a very courageous thing (something I did not have the courage to do when I saw similar or worse practices) but may have compromised her training in return. I feel for her but I can also understand the other side (damn this dialectical philosophy!!!).

it just seems like their could have been more effective methods to deal with this situation that would have worked out better for each party. Note: I don;t know all the details, just basing this on the article.
 
DynamicDidactic said:
I guess I can say it more directly, without insinuating, traditionally it is the role of the faculty to question their students' methods rather than the opposite. I am not saying this is correct (conversely, I believe it should be a two-way road). But I can understand how a student's critique of their mentor's methodology, in front of the mentor's colleagues (this is how I read the article) can be embarrassing (one does not need ad hominem attacks to feel embarrassed or humiliated).

I think most (if not all) people on this forum had to learn how to navigate the interpersonal waters of grad school (and internship and post doc). I am highlighting that this student did a very courageous thing (something I did not have the courage to do when I saw similar or worse practices) but may have compromised her training in return. I feel for her but I can also understand the other side (damn this dialectical philosophy!!!).

it just seems like their could have been more effective methods to deal with this situation that would have worked out better for each party. Note: I don;t know all the details, just basing this on the article.
Click to expand...

I agree that it seems like both trainee and advisor could have dealt with this differently and possibly more effectively -- Like you mention, I place greater responsibility on the advisor in this situation, because the trainee is still acquiring the interpersonal skills necessary to navigate graduate school and an academic department.

I also believe that, from what is made available in the article, the advisor's behavior seems retaliatory (e.g., in response to being criticized by an underling), defensive (e.g., in response to a trainee expressing concern regarding a decision he made), and possibly capricious (e.g., contingent on his experience of perceived humiliation or embarrassment) -- When I am evaluating the legitimacy of a grievance raised by a trainee against a professor, the presence of retaliation and capriciousness are often the two criteria I am most concerned with first establishing.

Were there concerns from other faculty regarding this student's behavior? Was the student provided with an opportunity to remediate these perceived weaknesses?

I also agree that if the Department and University complied with their own policies and procedures, then the trainee likely does not have a leg to stand on legally.

...... also why trifurcate responses on a 5-point Likert scale? It's almost as if sexuality should be conceptualized as existing on a continuum rather than within orthogonal categories.
 
Last edited:
Not sure who is at fault here and likely never will. That said, this highlights a rather large complaint I have had about the field since in graduate school and that is the lack of standardized training and procedures. You would think a field dedicated to training scientists and researchers would do a better job of setting up operationalized standards and procedures across training sites and schools. I would put money on the fact that the dismissal from her supervisor that was 'mutual' was never tendered in writing by both parties. This is still very much a go along to get along field.
 
Last edited:
Sanman said:
Not sure who is at fault here and likely never will. That said, this highlights a rather large complaint I have had about the field since in graduate school and that is the lack of standardized training and procedures. You would think a field dedicated to training scientists and researchers would do a better job of setting up operationalized standards and procedures across training sites and schools. I would put money on the fact that the dismissal from her supervisor that was 'mutual' was never tendered in writing by both parties. This is still very much a go along to get along field.
Click to expand...

From the student's tweets, it's clear from her side that the decision to leave his lab was unilateral on the side of the professor. I do think programs need to be better about supporting students who leave labs (this has happened in a handful of cases in my clinical program, and it's absolutely fine and people can figure out how to move on professionally).

As an aside, it's interesting (but maybe good?) that in our field trainees cannot be let go from most institutions without length processes being followed, whereas in industry and other settings one can be terminated even after short periods of poor performance.
 
None of this should be on the Tweeters in a perfect world. This makes it ripe for misinterpration, bandwagon effects, etc.

Any/all major universities should have a series of procedures for what seems to have occurred here. Its not really clear what fully occurred here...and you can't get the full story in tweeter bites, nor should you trust it (from either side).
 
One of my secondary research interests in LGBTQ+ issues, and I honestly would have little issue as a reviewer with someone saying that they coded a sexual orientation scale like that, with a note saying that a straight (heh) linear coding didn't yield significant results, To me, this doesn't equal p-hacking so much as seeing that, for this particular issue "mostly hetero" and "mostly gay" participants fall seem to fall in line with their "completely hetero" and "completely gay" counterparts. Honestly, our understanding of people with sexual orientations that aren't 100% straight and heteroromantic or 100% gay and homoromantic is murky at best, so we honestly need exploratory analyses in this area to try to understand how people who don't fall into these two camps differ or don't differ from those that do. Of course, fully explaining your methods is absolutely critical for these reasons, as well as basic research integrity.
 
Unipsychler said:
From the article, it sounds like she publicly accused him of "research misconduct" in front of his colleagues. I think that goes beyond critical analysis of research methods and shows that she has no understanding of the nature of mentor relationships in the field. I can see why no one else would want to work with her.

Edit to add: Obviously if there are real concerns of research misconduct a student should speak up about them. But I think it's common sense to phrase it as "I have concerns about how post-hoc recoding of this data will affect the validity of the statistical significance" or whatever, and then discuss how to address that in the write-up, rather than jumping straight to accusing others of fraud. If I were in that situation, and felt that my concerns were being inappropriately dismissed, I would start with discrete conversations with other department leaders before making public accusations. Pointing fingers in public and posting about it on Twitter before it's been resolved within the research team, all before lining up a new advisor, seems like shooting yourself in the foot and really makes me question her judgment.
Click to expand...

Just following up on this now that I've had a chance to read through her extensive account on Twitter (before I had just read the linked article). It does seem that she addressed this in research meetings with a more "critical analysis", less "outright accusation" tone at first. So what I said about going public before addressing it collaboratively was not accurate.

That being said, the more I read about it, the more it seems like there were longstanding issues. E.g., the advisor seems to think they had an understanding that she wasn't continuing the PhD after completing the MA. And doesn't quote any text stating such, but does cite an email from 2017 stating that it might be best for her to switch labs *if* she continues in the program after the MA. She refutes this as essentially BS, showing the title of a thesis that was submitted 2 years later that seems incompatible with his statements about her research interests. But no info about his opinion/approval of this thesis, how they agreed on it, or any other conversations that must have happened in the 2 years between his email and the current issue. I'm not going to even begin to guess what is really going on here, but it is apparent that her status in that lab was contentious waaaay before this alleged research misconduct issue. So to echo others, I have to assume there's a lot of info going into the decisions of the student and the school/PI that isn't being reported.
 
While her situation sucks, I don't get how she expected to get a positive LOR from a mentor who clearly disliked her or a helpful response from the department when it seems rare that programs actually have protections for students or any oversight for mentors behaving badly. My program has issue worse than this every semester and students that are blatantly discriminated against are just told to stop whining.
 
Unipsychler said:
Just following up on this now that I've had a chance to read through her extensive account on Twitter (before I had just read the linked article). It does seem that she addressed this in research meetings with a more "critical analysis", less "outright accusation" tone at first. So what I said about going public before addressing it collaboratively was not accurate.

That being said, the more I read about it, the more it seems like there were longstanding issues. E.g., the advisor seems to think they had an understanding that she wasn't continuing the PhD after completing the MA. And doesn't quote any text stating such, but does cite an email from 2017 stating that it might be best for her to switch labs *if* she continues in the program after the MA. She refutes this as essentially BS, showing the title of a thesis that was submitted 2 years later that seems incompatible with his statements about her research interests. But no info about his opinion/approval of this thesis, how they agreed on it, or any other conversations that must have happened in the 2 years between his email and the current issue. I'm not going to even begin to guess what is really going on here, but it is apparent that her status in that lab was contentious waaaay before this alleged research misconduct issue. So to echo others, I have to assume there's a lot of info going into the decisions of the student and the school/PI that isn't being reported.
Click to expand...
Can you quote or link some of this stuff?

After just a cursory look through her recent tweets doesn't leave me very confident in her side of the story.
 
psych.meout said:
Can you quote or link some of this stuff?
Click to expand...




She says she wasn't aware anyone had decided she wouldn't continue in the program. His email from 2017 says "If you end up staying at Clark, I would recommend that you identify research areas that you'd like to pursue, assess whether I'm the best mentor for you, and if not, identify other faculty members with whom you have more compatible research interests."

Now that in no way demonstrates that they reached a mutual agreement that she wouldn't continue the program. But that "if" statement really makes it sound like this was discussed in some way, and can't be a complete surprise. To have any idea what is going on here, I'd want to know what kind of conversation led up to that 2017 email and how they resolved that at the time. Like, did they resolve their research mismatch and she ended up picking a topic that fit in within his research scope, with his blessing to move forward, and then 18 months later he used these past things as an excuse to retaliate? Or was he telling her all along that it wasn't a good fit and she wasn't listening?

Especially since her cited reason for expulsion was not finding a new advisor within 30 days of her formal dismissal from the lab, it's relevant to know if she was blindsided by this with only 30 days lead time, or if she was aware of this issue for 2 years but failed to take appropriate action. There just isn't enough info to know.
 
Unipsychler said:



She says she wasn't aware anyone had decided she wouldn't continue in the program. His email from 2017 says "If you end up staying at Clark, I would recommend that you identify research areas that you'd like to pursue, assess whether I'm the best mentor for you, and if not, identify other faculty members with whom you have more compatible research interests."

Now that in no way demonstrates that they reached a mutual agreement that she wouldn't continue the program. But that "if" statement really makes it sound like this was discussed in some way, and can't be a complete surprise. To have any idea what is going on here, I'd want to know what kind of conversation led up to that 2017 email and how they resolved that at the time. Like, did they resolve their research mismatch and she ended up picking a topic that fit in within his research scope, with his blessing to move forward, and then 18 months later he used these past things as an excuse to retaliate? Or was he telling her all along that it wasn't a good fit and she wasn't listening?

Especially since her cited reason for expulsion was not finding a new advisor within 30 days of her formal dismissal from the lab, it's relevant to know if she was blindsided by this with only 30 days lead time, or if she was aware of this issue for 2 years but failed to take appropriate action. There just isn't enough info to know.
Click to expand...

Thanks!

It's such a weird situation and there's so much that is odd, confusing, contradictory, or otherwise doesn't make much sense.

Do you have any links or screenshots for this one as well?
Just following up on this now that I've had a chance to read through her extensive account on Twitter (before I had just read the linked article). It does seem that she addressed this in research meetings with a more "critical analysis", less "outright accusation" tone at first. So what I said about going public before addressing it collaboratively was not accurate.
Click to expand...
 
Having read the Twitter thread, this reads to me like "student takes issue with how professor conceptualizes sexual orientation -> student accuses professor of research misconduct -> professor disagrees -> student throws a fit." Should a manuscript include a note that a linear analysis of the scale didn't yield significant results? Yes, definitely. In a study looking at other's behavior especially, is it indefensible or even questionable to group "mostly het" and "mostly gay" participants with their "completely het" and "completely gay" counterparts? Imo, no. There is a lot of in-fighting in lesbian and queer female communities about who's "really a lesbian", and it seems that the student basically decided that not agreeing with her on that equaled research misconduct.
 
Ollie123 said:
I'm in clinic so just scanned it briefly. It sounds like crappy science was being done, but (sadly) not anything remotely out of the ordinary. It sounds like some crappy mentoring was being done. That did seem out of the ordinary, even for crappy mentors.

I think the case for gender/sexual discrimination is going to be unbelievably hard to prove based off what is written. I literally didn't see anything that suggested gender played a role in a meaningful way, though there could well be other things not in the article. She may have a better case as a whistleblower, though I don't know the intricacies of how those are treated in academic settings. Unfortunately, I'm not sure that being a bad advisor/department is necessarily something the legal system can address. Feel bad for the student...
Click to expand...

The gender discrimination will be tough to prove however she attached a retaliation claim on there as well. She had an adverse action(termination) which was contemporaneous with the complaint. So all she has to do is connect the dots.

The gender discrimination allegations don't have to be true to prove retaliation.
 
DynamicDidactic said:
I hope the faculty at Clark (and across departments) starts to realize that if you are going to teach students about p-hacking and HARKing you might have to stop modeling it as well (or start hiding it better).
Click to expand...
What better way to teach them than to show them that it's how a large portion of research gets done and ultimately published? 😉
 
DynamicDidactic said:
I hope the faculty at Clark (and across departments) starts to realize that if you are going to teach students about p-hacking and HARKing you might have to stop modeling it as well (or start hiding it better).
Click to expand...
This isn’t that, though. I’m honestly not seeing much, if any, ethical grounding for this complaint at all.
 
Last edited:
DynamicDidactic said:
I hope the faculty at Clark (and across departments) starts to realize that if you are going to teach students about p-hacking and HARKing you might have to stop modeling it as well (or start hiding it better).
Click to expand...
I was thinking, "Why does it take a student (rather than other faculty members) to police the intellectual/scientific integrity of the professor's research?"

Like, if there are credible (and rather obvious) criticisms of the methodology, why aren't other faculty members agreeing with (or otherwise addressing) the substance of the critique rather than the source of the critique (a student)?

My guess is that 'looking the other way' reduces cognitive dissonance ('cause everyone does it, too') and finding 'positive results' or support for one's theory 'pays the bills' whereas finding 'negative results' means you 'lose' in the academic game. From a certain philosophy of science perspective (e.g., falsificationist perspective where one, in theory, should be choosing the most stringent empirical tests of one's theory in order that they may be more and more corroborated (by empirical non-falsification) over time) this is completely inverted. But it's modern 'science' in the social sciences. Almost like your purpose is to 'find results to publish' rather than pursue the truth.
 
Fan_of_Meehl said:
I was thinking, "Why does it take a student (rather than other faculty members) to police the intellectual/scientific integrity of the professor's research?"

Like, if there are credible (and rather obvious) criticisms of the methodology, why aren't other faculty members agreeing with (or otherwise addressing) the substance of the critique rather than the source of the critique (a student)?

My guess is that 'looking the other way' reduces cognitive dissonance ('cause everyone does it, too') and finding 'positive results' or support for one's theory 'pays the bills' whereas finding 'negative results' means you 'lose' in the academic game. From a certain philosophy of science perspective (e.g., falsificationist perspective where one, in theory, should be choosing the most stringent empirical tests of one's theory in order that they may be more and more corroborated (by empirical non-falsification) over time) this is completely inverted. But it's modern 'science' in the social sciences. Almost like your purpose is to 'find results to publish' rather than pursue the truth.
Click to expand...
Again, I’m not seeing how this is that. I’m really not seeing how the professor did anything unethical here. The decision made theoretical, conceptual, and methodological sense.
 
futureapppsy2 said:
Again, I’m not seeing how this is that. I’m really not seeing how the professor did anything unethical here. The decision made theoretical, conceptual, and methodological sense.
Click to expand...
Unethical, no. But I think that science (in general) and maybe slippery sciences (social science) in particular have been a bit warped by the consequences (reward and punishment) for setting up systems of 'investigation' that tend to give certain types of results (e.g., confirmation of theory rather than refutation).

And, to be clear, if the investigator: a) is utilizing methodology that is in common use among his/her peers and is b) transparent about it (i.e., makes clear in the research article that the first manner of coding didn't find the results they were looking for so they played with the coding scheme until they found one that 'discovered' positive results, then fine. The context of discovery can be meaningfully distinguished from the context of empirically testing that which was discovered earlier. I think it's important that investigators be transparent about what they're engaging in and be honest about calling the results. For example, if this 'new' method of coding that has been found to be 'successful' is cross-validated by trying it on new samples or developing new methods of coding (and these new methods interlace with prevailing theories) and these are themselves critically examined empirically, then fine.

From a philosophy of science perspective, the choice to revise the coding scheme (rather than revise the theory) when the empirical results didn't line up is an interesting one. It's reminiscent of the concept of 'protective belts' of auxiliary (ad hoc and post hoc) assumptions that 'protect' the theory from refutation rather than accept the negative empirical result and go back to the drawing board with respect to the theory.
 
Last edited:
@futureapppsy2 maybe we aren't looking at the same data/information. Based on student's report:

the a priori analysis was non-significant. A post hoc analysis was conducted. If that post hoc analyses was presented as the a prior hypothesis, then it is the textbook example of HARKing. http://www.pitt.edu/~peterb/papers/Kerr-HARKing-PersAndSocialPsychReview98.pdf

Now, I have no idea if this was published or will be published. But the student reports in her Tweets that this was already presented at a conference. I have not seen this poster but I am fairly certain that this was presented as an a priori hypothesis.

If one were to test this hypothesis in a manner more aligned with honest (I am purposefully avoiding the term ethical and maybe honest isn't the best term) research principles, a new sample needs to be collected and the question re-asked (maybe using revised language).

This also doesn't take into consideration that the analysis of this ordinal data seems to be inappropriate. Are they taking an ordinal scale and creating three separate categories (making it nominal)?
 
Fan_of_Meehl said:
My guess is that 'looking the other way' reduces cognitive dissonance ('cause everyone does it, too') and finding 'positive results' or support for one's theory 'pays the bills' whereas finding 'negative results' means you 'lose' in the academic game. From a certain philosophy of science perspective (e.g., falsificationist perspective where one, in theory, should be choosing the most stringent empirical tests of one's theory in order that they may be more and more corroborated (by empirical non-falsification) over time) this is completely inverted. But it's modern 'science' in the social sciences. Almost like your purpose is to 'find results to publish' rather than pursue the truth.
Click to expand...
Almost definitely. I am fairly certain that every faculty in that program has been guilty of HARKing or p-hacking. It is the standard in a lot of social psychology research (and common in all sub-fields of psychology).
 
DynamicDidactic said:
@futureapppsy2 maybe we aren't looking at the same data/information. Based on student's report:

the a priori analysis was non-significant. A post hoc analysis was conducted. If that post hoc analyses was presented as the a prior hypothesis, then it is the textbook example of HARKing. http://www.pitt.edu/~peterb/papers/Kerr-HARKing-PersAndSocialPsychReview98.pdf

Now, I have no idea if this was published or will be published. But the student reports in her Tweets that this was already presented at a conference. I have not seen this poster but I am fairly certain that this was presented as an a priori hypothesis.

If one were to test this hypothesis in a manner more aligned with honest (I am purposefully avoiding the term ethical and maybe honest isn't the best term) research principles, a new sample needs to be collected and the question re-asked (maybe using revised language).

This also doesn't take into consideration that the analysis of this ordinal data seems to be inappropriate. Are they taking an ordinal scale and creating three separate categories (making it nominal)?
Click to expand...
I used to joke (by adopting the inflection and tone of voice of a major professor at the university I trained at (and in whose lab I trained)) about her data being good and 'obedient' data. At one point, her research assistant came into the room with a stack/sheaf of printouts of an intercorrelation matrix with statistically-significant correlations highlighted and then we proceeded to 'brainstorm' about what the results meant in theoretical terms and in terms that were consonant with the mainstream theories in that area of science (so that these findings might be written about and published). It turned me off to the idea of a career in academia completely. It is this spirit that I think is corrupt. I have no idea if this spirit actually exists in the professor's lab (in question). I'm just pointing it out as one mindset that I've witnessed personally and that I think is fairly prevalent in social science research.
 
DynamicDidactic said:
@futureapppsy2 maybe we aren't looking at the same data/information. Based on student's report:

the a priori analysis was non-significant. A post hoc analysis was conducted. If that post hoc analyses was presented as the a prior hypothesis, then it is the textbook example of HARKing. http://www.pitt.edu/~peterb/papers/Kerr-HARKing-PersAndSocialPsychReview98.pdf

Now, I have no idea if this was published or will be published. But the student reports in her Tweets that this was already presented at a conference. I have not seen this poster but I am fairly certain that this was presented as an a priori hypothesis.

If one were to test this hypothesis in a manner more aligned with honest (I am purposefully avoiding the term ethical and maybe honest isn't the best term) research principles, a new sample needs to be collected and the question re-asked (maybe using revised language).

This also doesn't take into consideration that the analysis of this ordinal data seems to be inappropriate. Are they taking an ordinal scale and creating three separate categories (making it nominal)?
Click to expand...
As someone who does work in the murkier areas of sexual orientation, we really don’t know that much about the 2s and 4s on that scale, and if or how they differ or not from the 1s, 3s, and 5. For the hypothesis of the study in question—which has a strong grounding in how they are viewed by others—it makes theoretical sense to see if grouping “mostly gay” and “mostly straight” participants with “completely straight” and “completely gay” participants changes results. Yes, definitely say that you looked at the scale both ways, for sure, but I don’t see that as HARKing, because we don’t have a standardized way to talk about 2s and 4s or categorize them yet. So we need exploratory analyses like this.
 
This is one of those situations where there is a power structure. Maybe it makes sense, maybe it doesn’t. But trying to fight isn’t going get you what you want. Pretty ballsy to call shenanigans on it. I like it.

I thought everyone knew that lab time is a place where you’re your PIs “b”.
 
futureapppsy2 said:
Yes, definitely say that you looked at the scale both ways, for sure, but I don’t see that as HARKing... So we need exploratory analyses like this.
Click to expand...
Then it should be presented as exploratory, which this wasn't (apparently). Like I said, this is the textbook case of HARKing: if you are doing post hoc analyses after your originally planned analysis is nonsignificant without describing them as exploratory.

futureapppsy2 said:
because we don’t have a standardized way to talk about 2s and 4s or categorize them yet. So we need exploratory analyses like this.
Click to expand...
Also sounds like there is a need for a better validated measure of sexual orientation. Or a general better idea of the construct of sexual orientation (not on my area of knowledge). If these are not well understood in the field then I wouldn't be working on the OMTH or the double jeopardy hypothesis.

But what the hell do I know, I am not productive enough or creative enough to be on the tenure track at Clark. I also tried to avoid p-hacking and HARKing in grad school (I have a lot of null findings that never got published).

PsyDr said:
Pretty ballsy to call shenanigans on it.
Click to expand...
And I didn't have the balls to call out the faculty that routinely used these practices (fortunately, not my advisor).
 
DynamicDidactic said:
Almost definitely. I am fairly certain that every faculty in that program has been guilty of HARKing or p-hacking. It is the standard in a lot of social psychology research (and common in all sub-fields of psychology).
Click to expand...

Common in all sub-fields of psychology that use hypothetical-deductive research methodologies. P-hacking is only an issue if you use p
 
DynamicDidactic said:
Then it should be presented as exploratory, which this wasn't (apparently). Like I said, this is the textbook case of HARKing: if you are doing post hoc analyses after your originally planned analysis is nonsignificant without describing them as exploratory.
Click to expand...

The problem is that most of the information is coming from the student and her supporters, who are getting their information from her.

Yes, she child be entirely correct and there are inappropriate things going on, like p jacking and HARKing, but it reminds me of that recent case of the clinical student dismissed from Idaho State University.

If you take the word of his wife at face value, based on her many posts on her website, the university capriciously dismissed him for racially prejudiced reasons. The problem is that this didn't line up with the actual facts, but the university couldn't publicly post about it like she did due to FERPA.
 
psych.meout said:
The problem is that most of the information is coming from the student and her supporters, who are getting their information from her.

Yes, she child be entirely correct and there are inappropriate things going on, like p jacking and HARKing, but it reminds me of that recent case of the clinical student dismissed from Idaho State University.

If you take the word of his wife at face value, based on her many posts on her website, the university capriciously dismissed him for racially prejudiced reasons. The problem is that this didn't line up with the actual facts, but the university couldn't publicly post about it like she did due to FERPA.
Click to expand...

Totally agree. And even from the evidence that she is posting herself, trying to rip her former advisor and school apart, I believe it paints a very ambiguous picture. Makes you wonder what email screenshots THEY are using to support their side. But of course she's the only one airing this all on Twitter, so we'll never know.
 
P hacking is weird. If your model says it should work, and it doesn’t, you can publish null results. I’ve done it.

She seems to see herself as a whistle blower. Her twitter also has stuff like non replication of mortality salience. Which I can appreciate on some levels. I used that manipulation gives times and got nothing, and I have several colleagues w the same experience.
 
Unipsychler said:
Totally agree. And even from the evidence that she is posting herself, trying to rip her former advisor and school apart, I believe it paints a very ambiguous picture. Makes you wonder what email screenshots THEY are using to support their side. But of course she's the only one airing this all on Twitter, so we'll never know.
Click to expand...
This is why it's important to find out what actually happened in terms of how the issue was discussed in the department, e.g., did she discuss these issues individually with her advisor and then escalate them when he didn't provide satisfactory explanations or solutions vs. did she just call him out in front of his colleagues?

If it was some version of the latter, can you really blame the other faculty for not wanting to take her on as an advisee?
 
ClinicalABA said:
Common in all sub-fields of psychology that use hypothetical-deductive research methodologies.
Click to expand...
Question is how it is presented. Do whatever you want with your data as long as you aren’t obfuscating the analyses/results. It seems convenient to report only the analyses that support the hypotheses. Again, based on the one-sided facts here. I understand that it may not be the full picture. But we know this is not an uncommon practice.
 
You must log in or register to reply here.

Similar threads

singasongofjoy
international student for practicum between universities?
Replies
2
Views
452
PsyDr
PsyDr
T
PhD/PsyD Formal postdoc vs. informal vs. staff post-grad?
Replies
10
Views
2K
PsychPsychBaby
PsychPsychBaby
PsyDr
Student Funding Opportunities (including internship travel)
Replies
0
Views
828
PsyDr
PsyDr
J
Deciding where to go to Grad School
Replies
13
Views
2K
MAClinician
M
S
  • Question Question
Nontraditional student -- best options?
Replies
14
Views
1K
summerbabe
summerbabe
Top