Draft of Observer Column Sparks Strong Social Media Response

A set of abstract watercolor textured speech bubbles in various sizes and colors all overlapping symbolizing gossip, social media networking and conversation.

The Observer, APS’s membership magazine, found itself in an unusual position this week when a draft of an upcoming Presidential Guest Column began circulating on Facebook and Twitter.  The opinion piece, written by APS Past President Susan T. Fiske at the invitation of current APS President Susan Goldin-Meadow, decries the unmoderated criticism of researchers on social media.

The piece, submitted for publication in the magazine’s November issue and still in the editing phase, has generated a strong response on Twitter and Facebook, with many criticizing both Fiske and the Observer for its tone and content.

APS encourages its members to air differing viewpoints on issues of importance to the field of psychological science, and the Observer provides a forum for those deliberations, Goldin-Meadow notes.

“Susan Fiske is a distinguished leader in the field and I invited her to share her opinion for an upcoming edition of the magazine,” she says. “It’s unfortunate that many on social media view her remarks as an attack on open science, when her goal is simply to remind us that scientists sometimes use social media in destructive ways. APS fully expects and welcomes discussion around the issues she raises.”

Those wishing to share their opinions on this particular matter are invited to submit comments in the space below. Alternatively, letters can be sent to apsobserver@psychologicalscience.org. We ask that your comments include your full name and affiliation.

 

Comments

I would invite APS to consider hosting a town hall on these issues at a forthcoming conference. Clearly, many people are impassioned about these issues – perhaps more than for any other matter in the field right now.

Just as some scientists use social media destructively, it is perhaps worth remembering that some also use traditional peer-reviewed manuscripts or op-ed pieces destructively. Thus, it doesn’t seem the medium (social media) or degree of peer review are strong predictors of behavior here. The APS is in a unique position to facilitate positive discussions around issues of psychological science and best practices! I look forward to seeing a variety of opinions on these topics in coming issues.

Goldin-Meadows writes: “It’s unfortunate that many on social media view her remarks as an attack on open science, when her goal is simply to remind us that scientists sometimes use social media in destructive ways.”

It is unfortunate, but understandable. Some of the terms used in Fiske’s piece include:
-Mob
-Online vigilantes
-Self-appointed data police
-Personal ferocity
-[they are] crashing people
-Unmoderated attacks
-Unaccountable bullies
-Adversarial viciousness
-Methodological terrorists
-Dangerous minority
-Destructo-critics
-They attack the person (oops)
-Self-appointed critics

In our blog DataColada.org we do not allow comments from readers to avoid vicious language as that use by Fiske to be broadcasted to the masses. I guess APS is less concerned with that.

Uri

This Observations note manages to both misrepresent the content of Fiske’s column AND the backlash.

No, the column was NOT about destructive use of social media; it was about critics of a particular *sort*. Who are “self-appointed data police”? That phrase implies a particular kind of researcher. Pretending otherwise is counter to a plain reading of the text, and can only be meant to protect Fiske from (much deserved) critique.

No, the backlash was NOT because Fiske attacked “open science”. It was because she attacked particular kinds of scientists – that is, *people*, not an abstraction. Her “methodological terrorist” is a person. She made this attack using innuendo and generalisations (“they do X”). This is wrong, period, and ironically exactly the sort of thing she was supposedly decrying.

Finally, APS has noted that there is already plenty of discussion elsewhere in the social media universe. This particular comment section was not needed to generate discussion. It is already happening.

It is therefore unclear what is added by this “Observations” note. It has negative information value because it misrepresents the situation, and the comment space is redundant.

I hope that psychologists will strive to be civil in their discourse, be it via social media posts, manuscript reviews, or newsletter columns. Transitioning to a more transparent, open, replicable psychological science is likely to be a long and challenging process. Polarization is the last thing we need.

Steve

P.S. Watch for a forthcoming Observer piece on the virtues of preregistering research.

I suggest that the APS pull this column from publication and issue an apology to its readers. To me as an early-career researcher, the column seems to be sending two dangerous messages:

1) That the publication of an article in a peer-reviewed journal should limit our skepticism of it to the point where that skepticism can only be conveyed in a very limited range of communication avenues (e.g., via letters to the editor, which are rarely published). In general, this anti-criticism attitude is fundamentally incoherent with the critical and skeptical nature of science.

2) That engaging in free and open discourse and criticism is a privilege for those in power: That it is *not* ok for me as a young researcher to criticise the studies of a powerful academic like Dr Fiske in non- peer-reviewed media, but that it *is* ok for her to use non- peer-reviewed media to call people like me “methodological terrorists”.

Dr Fiske claims that “unmoderated attacks” are driving researchers out of our field. I cannot speak for everyone, but for me as an early career researcher, it is comments like those of Dr Fiske that make me hesitant to pursue a career in psychology – *not* those of the methodologists desperately trying to improve our research practices.

Pick your sides and go to war in the Scientific Revolution 2.0, or engage in thoughtful debate about the best practices to advance science. I’m glad my early drafts of opinion pieces are not published before peer review, though I’m disappointed in the extreme language here. I’m sad that there are actually some replication scientists who are cruel and deserving of harsh criticism, but surprised at the resistance to truly better methods, even if they are still developing. APS leads its peers as an organization in bringing forth scientific transparency ever since the Replication Crisis emerged. When all the dust is cleared, History will recognize Psychology as a leader in bringing forth this paradigm shift. Be patient as Science churns.

“Comments go live after a short delay.” Godot

Dear colleagues,

Susan Fiske’s column does not name names. It is therefore not clear which online activities she considers to be constructive criticisms and which online activities she considers to be destructo-criticisms that hurt psychology as a science.

As a blogger and a moderator of a Facebook Discussion group, I value feedback and peer-comments.

Everybody is welcome to read my blogs (e.g., my replicability report on ego-depletion or my replicability rankings of psychology journals) and provide comments.

google search: replicability index

I am pleased to read that “not all self-appointed critics behave unethically, and some do so more than others. One hopes that these critics aim to improve the field, not harm people.

The Psychological Methods Discussion Group follows the US American rules of free speech.

Limitations to free speech are (a) infringe on copyright laws, (b) incite violence against a particular person, or (c) use untruths to harm others (slander).

Members are encouraged to bring posts that violate these norms to the attention of the administrator. With over 2000 members, no post was flagged for violating these rules.

You can see for yourself. Look for Psychological Methods Discussion Group (or group 853552931365745)

final post.

I tried to submit a single post, but it was not accepted.

I then tested various hypotheses (no swear words, no hyperlinks, etc.), but I still don’t understand the behavior of this system as it just accepted the hyperlink to the Facebook discussion group.

Let’s see whether this post gets accepted.

Dr. Fiske should consider herself lucky that many scientists critical of her work voluntarily forgo the anonymity afforded peer reviewers. I shudder to think what those comments would look like..

Susan Fiske’s editorial correctly describes my experience in a Facebook discussion group where I have been put on a “list of shame” together with fraudster Diederick Stapel and where it was claimed that I had “fabricated” results. I can only hope that people don’t let themselves be terrorized by such invectives.

I am deeply saddened by Susan Fiske’s commentary. I had hoped that the copy Fritz Strack posted on Facebook was an early draft, and that once she received some feedback she might come up with a version that is more including and more understanding, but it appears that she is standing fast in her over-the-top portrayal of us who are working on reforming science.

She is also misdiagnosing the problem. The worries she is describing are not due to the reform-movement. It is, as many have pointed out, due to the structures and incentives. I synthesized those thoughts in this blog post (yes blog post https://asehelene.wordpress.com/2013/11/19/too-many/) a couple of years ago, but two recent papers have modeled how the current structures are selecting for bad science (Smaldino, McElreath, 2016), and how the propensity for file-drawering non-significant results leads to the canonization of false facts (Nissen, Magidson, Gross & Bergstrom, 2016). Ed Yong does a compentent (as always) overview of the Smaldino paper in his Atlantic Article “The inevitable evolution of bad science.

The way the Scientific endeavor has used up and discarded talent is a long standing problem. Paul Meehl talked about it in his last series of lectures. Paula Stephan outlines it in her book “How Economics Shapes Science.”

That some people have behaved unconscionably is, to say the least, regrettable, but is also, unfortunately, a long standing issue in science. Anonymous Peer Review is notorious. That doesn’t excuse the behavior.

I want a science that, in the long run, produces good, reliable information. I want a science that takes care of and nuture its talents. I got into this because I’m curious about what is, and I like to teach.

Nissen, Silas B., Magidson, Tali, Gross, Kevin, Bergstrom, Carl T (2016). Publication bias and the canonization of false facts. https://arxiv.org/abs/1609.00494

Smaldino, Paul E. & McElreath, Richard (2016). The natural selection of bad science. Royal Society, Open Science (http://rsos.royalsocietypublishing.org/content/3/9/160384)

Ed Yong (2016). The inevitable evolution of bad science. The Atlantic. http://www.theatlantic.com/science/archive/2016/09/the-inevitable-evolution-of-bad-science/500609/

These sentences are funny.

The opinion piece, written by APS Past President Susan T. Fiske at the invitation of current APS President Susan Goldin-Meadow, **decries the unmoderated criticism of researchers on social media.**

The piece, submitted for publication in the magazine’s November issue and still in the editing phase, has generated a strong response on Twitter and Facebook, with many criticizing both Fiske and the Observer for its tone and content.

being strongly criticized for decrying strong criticism 🙂 😀

She did actually, really get attacked and shamed somehow, like at a conference in front of 300 people

APS created the Registered Replication Reports.

A recent RRR published in the Perspectives Series showed that an original finding from Strack et al. (1988) could not be replicated, even with a sample size more than 10 times larger than the original sample. The lead author of the original article, Fritz Strack, was given the opportunity to comment on this result. It is not clear whether this comment was peer-reviewed or not.

In this commentary, FS presents some secondary analyses of the RRR results. Again it is not clear whether these analyses were peer-reviewed.

FS examined the correlation between effect size and sample size, which is a standard method to examine publication bias in meta-analysis. If there is publication bias, effect sizes tend to be negatively correlated with sample sizes.

FS reports a correlation of r(17) = .45, p =
.069.

We see that the correlation is not negative, which is not surprising because not a single study of the 17 independent replication studies was significant. So there is no evidence that publication bias created too many significant results.

Second, the positive correlation is not significant by conventional standards and the most plausible explanation for the positive correlation is sampling error. Yes, the correlation looks strong, but in small samples sampling error can produce strong correlations.

However, FS rejects the null-hypothesis and claims that he found a meaningful pattern that requires interpretation.

He writes, and I quote, “without insinuating the possibility of a reverse p-hacking, the current anomaly needs to be further explored.”

Again, it is not clear whether this statement passed peer-review. I wonder how readers of the esteemed Perspectives series published by APS are supposed to interpret this sentence, but as a social psychologists I have a pretty good idea how the human brain processes a sentence “without insinuating the possibility of X.”

Here are a few examples.

“Without insinuating that registered replication reports were created to undermine the reputation of prominent psychologists.”

“Without insinuating that attacks by social media critics of famous psychologists are fueled by envy.”

“Without insinuating that failed replications are the result of massive use of questionable research practices by original authors.”

“Without insinuating that original authors’ responses to failed replications are a defense mechanism to protect their own self-esteem.”

I guess we all get the picture. It is possible that the warning had the unintended, ironic effect to create the impression that the RRR results cannot be trusted because reverse p-hacking occurred.

– It is time to fix this website as I am not the only one who encounters problems posting commentaries.

This is unfortunate because larger effects in larger samples would imply a greater chance to obtain significant results. Thus, even a significant positive correlation between sample size and effect size would not support the hypothesis that the authors of the RRR were engaging in reverse p-hacking.

I suggest that APS submits commentaries published in its prestigious journals to the same rigorous review process as original articles.

This comment thread has drifted off topic. I’d like to bring it back to the Fiske piece.

Is psychological science about reputations and careers, or it is about discovering facts about the world?

I am far more concerned about the quality of the science that we produce than the character of the social relationships between the scientists.

In the absence of a functional and cumulative model of scientific practice and dissemination, the professional status becomes disconnected from merit. In the absence of meaningful scientific progress, status becomes merely a means of perpetuating status.

It’s easy for a field to lose its soul.

Without a functioning model, we run the risk that scientific psychology becomes a kind of make-work.

Speaking for myself alone, I do not want to be involved in a field that serves as little more than economic stimulus for academics. I believe that some humanities fields are already there. Psychology is not, but we have been perilously close. It is time to change course.

Change is difficult. Change is scary. But in this case, change is necessary. And it is necessary now.

I’m not clear on the status of this statement. Is it a draft as the title of this post suggests?

The version that was published online by Fritz Strack on September 19th was labeled ‘in press’ and the only file name I have for this column includes “copy edited”.

Linda Sitka has stated she believed the column had been edited recently so the “most unfortunate phrase removed”. I asked Prof. Sitka and others for confirmation regarding whether the work was peer reviewed and if this or other inflammatory remarks were removed. I received no response.

So I’ll ask again:

Did the column undergo peer review?
Was the most unfortunate phrase (hopefully others) removed?
If so, when and why?

many thanks.

The Observer is an association membership magazine, not a peer-reviewed journal. The opinion piece in question was, without our prior knowledge or approval, posted online. We did not apply the “In Press” designation to the draft. We won’t discuss what wording remained or was excised from the piece since it was still in the midst of internal copywriting and had not even entered production.

That said, the column in question was submitted as an opinion piece, invited as a guest contribution by the APS President. All APS members are welcome to share opinions through the Observer Forum or by submitting story proposals to the editorial staff, at apsobserver@psychologicalscience.org.

Scott Sleek
Editor

I think Andrew Gelman has written the best article about Fiske’s piece, and he also situates it within the greater debate: http://andrewgelman.com/2016/09/21/what-has-happened-down-here-is-the-winds-have-changed/

I just want to highlight a paragraph right at the end of his post:

“The thing that saddens me is Fiske’s characterization of critics as “adversaries.” I’m not an adversary of pscyhological science! I’m not even an adversary of low-quality psychological science: we often learn from our mistakes and, indeed, in many cases it seems that we can’t really learn without first making errors of different sorts. What I am an adversary of, is people not admitting error and studiously looking away from mistakes that have been pointed out to them.”

This is what bothers me about the debate as well; that suddenly we are “adversaries” of each other. An unfortunate “us vs. them” mentality seems to have developed, with many on both sides (hah!) contributing to the polarization.

But who is “us”, and who is “them”? Am I one of the people Fiske is worried about, because I am skeptical of many common research practices and well-established ways of doing things (like anonymous peer review) and see the need for changes to the way we do (psychological) science? Or am I part of the establishment perpetuating the status quo, because despite my desire for change, my actual behaviour is pretty conventional (I use SPSS, run simple 2×2 social psych studies with cute manipulations, have yet to replicate anyone else’s findings)?

The human propensity to form groups and develop ingroup biases is pretty well established (surely the minimal groups paradigm would be robust to an RRR??), so it’s not surprising that an “old” vs “new” heuristic has developed in this context as well, with some people clearly being on one “side” or another. But I’m sure there are lots of people, like me, who don’t feel comfortable “picking sides”, and would rather focus on our superordinate identity as, oh, I dunno, psychological scientists.

As a psychological scientist, who’s planning to stick around in this field for a while, I will be aiming for truth *and* civility. I’m happy to talk about trade-offs and compromises, but I *won’t* make it an “either/or”.

This comment section needs like buttons.

Like comments by Hanne M Watkins, Andrew K. Przybylski, Matt McBee, y Åse Innes-Ker, Edward Ester, Matt Williams,
D. Stephen Lindsay, y Jon Grahe, Richard D. Morey, y Uri Simonsohn, y Jonathan Peelle.

Is the author going to respond to these comments?

With her column Susan Fiske points at a serious problem in our discipline. As a young colleague wrote to me recently: “The direction in which the discourse in our field is going is scary.” Free speech should not involve insulting language and it is sad that some colleagues seem to be unable to voice criticism without using such language. It is interesting that hardly any comment in this section denies that this is happening; instead they claim that this is part of parcel of free speech. I think that Susan Fiske did the field a service by raising this issue.

Dear Wolfgang Stroebe,

there is substance and there is style. We can quibble about style, but the real issue is substance.

I have commented on several of your peer-reviewed articles that comment on the replicability crisis in social psychology. The OSC reproducibility project replicated only 25% of original studies and I have presented objective and scientific evidence that one reason for this very low replication rate is that original studies have low power and published results were selected from a larger pool of results because they were significant.

Your publications often seem to suggest that social psychology does not have a replication problem and that there is no conclusive evidence that publication bias contributed to the low replication rate in social psychology.

I would like to take this opportunity to ask you directly to comment on the presence of publication bias in social psychology and what you think the field should do to improve its reputation.

Sincerely,
Ulrich Schimmack

Wolfgang Stroebe, the way I read it people are not defending their right to behave unethically by invoking free speech, people are defending free speech from the accusation that it will inevitably lead to unethical behaviour. Which is a claim from Fiske’s column, not from the commenters:

‘The destructo-critics are ignoring ethical rules of conduct because they circumvent peer-review: They attack the person, not just the work; […], they have implicated targets’ family members and advisors’.
Just to reiterate, people act unethically “because they circumvent peer-review”. I strongly disagree. Blogging and personal attacks, and especially blogging and harrassing family and colleagues of other scientists are very separate things.

As for the issue of harrassment happening and being widespread, how could one deny it? Fiske cites her own experience to support her claim, which is impossible to verify. Many people have probably not experienced the same, but that is rather meaningless, especially if these are people who are very actively engaging in the kind of dialogue she decries (by which I mean unmoderated dialogue, not harrassment). But ultimately, the question of how widespread harrassment is may not even be so important. Even if it is rare we should work to avoid it, but not by blaming it on a certain group, or shunning what many perceive as valuable tools.

Mr. Schimmack,
as you called me a stupid liar in one of your critiques, I am not entering into any discussion with you. My general impression is that in times of Trump, insult and slander has become defined as free speech.

Leave a Comment

Your email address will not be published.
In the interest of transparency, we do not accept anonymous comments.
Required fields are marked*