Orthodox academic biases and heterodox rectification? What?

Is Heterodoxy in itself a good thing? How strong has orthodoxy to be to make heterodoxy valuable.

I have to admit I am a bit uncertain what the name of the new blog-in-town “Heterodox Academy” implies: Is it meant to be the “Academy of Heterodox views” or just promoting for a less homogeneous academia? I ask, because I more or less interpreted the Academies title in the former way which made me rather unsympathetic to their cause. In this interpretation their mission appears self-serving from my biased left-liberal point of view.

I put the question from the beginning to twitter. Only Richard Tol answered,

which is not the worst possible quote.

You wonder what the Heterodox Academy is? In the end, I guess, it all refers back to this recently published paper which attracted a number of peer-commentaries (also to be found at the above link). The Heterodox Academy summarises the paper here. The text is not too long, thus read it.

But what it’s about? It’s about the following [supposed] problem

Psychologists have demonstrated the value of diversity—particularly diversity of viewpoints—for enhancing creativity, discovery, and problem solving. But one key type of viewpoint diversity is lacking in most of the social sciences (other than economics) as well as in the legal academy and the humanities: political diversity.

Online-comments on the HA appear to commonly (mis)understand the argument of the authors to be as follows: to introduce a new bias to oppose the current “liberal” bias in academia so that in the long term an unbiased view prevails. I think that is not what the Heterodox Academy has in mind.

Their argument is rather for a research environment that allows for and includes multiple diverse and possibly opposing viewpoints. Such an environment would, following their logic, reduce/prevent possibly existing biases in the short term, i.e. in the best case for each individual paper. To be more specific, my (mis)understanding of the argument is, that if a methodology is biased against conservative viewpoints, involving a conservative researcher in the development of the study could prevent this bias, or the opinion of a conservative reviewer could result in more thorough discussions of the implications of this potential methodological bias.

The Heterodox-summary at their web-page gives some examples how the current academic environment may prevent diversity of political viewpoints including, but not only, due to obvious hostility towards certain political leanings. However, I think there are more general arguments to be made about the problem.

I am slightly surprised by the form of the opposition to the Heterodoxy. I would assume that it is common sense that diversity of opinions helps to prevent biases. Furthermore, I would think that we all can imagine environments that discourage certain viewpoints to be voiced – for whatever reasons.

Some questions:

Would a post-growth economist want to work at an academic research institute that is clearly oriented towards a non-intervention school of thinking?

More extreme: Would a post-growth economist want to work at an academic research institute whose head publicly calls post-growth economists “Hippies”?

And one more: we can agree that “progressives” are not inherently “better”?

To be clear, I don’t favour opposition for opposition’s sake, and neither balance for balance’s sake – especially not if the opposing viewpoint is ‘objectively’ bollocks – where objectively is either obviously impossible or likely a value-judgement.

Words do matter. Framing a question, framing a discussion depends on our personal subjective perspectives. Therefore, the way we interpret and discuss the results of a study is at least slightly coloured by our assumptions. Awareness of our personal biases and subsequent attempts to minimise them are unlikely to be 100% successful. Peer discussions and review can contribute to reducing subjectivity.

If the space of opinions is one dimensional, discussions and reviews of different views may result in relevant reduction of biases. If the different views are located in the lower two quadrants of a two-dimensional space, a bias is going to remain.

Incidentally Machin (2015) also caught my attention this week and linked to Reider (2014, pdf), who states: “Knowledge is largely directed by the questions we, not merely as individuals, ask, but far more importantly, the questions a given society is willing to invest its resources to discover.” Machin (2015) takes this further: “what a given society counts as a ‘discovery’ will affect the direction that knowledge takes.”

This implies that if a society considers a certain perspective not valid or, less extreme, not valuable, this perspective is in danger not to be pursued. In a democratic society, it is likely that all perspectives within the general [political] spectrum of this society are valued high enough to give input on the questions pursued.

The implication for a society is also valid for different social communities within a society. If a community considers a certain perspective not valid, or less valuable, this perspective is in danger not to be pursued within this social community. The implication applies for academic communities as well. Again, one may hope that single academic branches have enough pluralism of opinions to correct biases. If that is not the case one may assume interdisciplinarity or at least interdisciplinary perspectives may provide a corrective force.

Obviously, there are viewpoints a society may deem unacceptable: far-right and/or far-left, questionable ethics, or more generally, certain value-sets may be off-limits. Could these nevertheless contribute interesting insights, valuable knowledge? Possibly. Nevertheless, shunning them is probably the right thing if they contribute to anti-constitutional or anti-democratic tendencies.

If an academic discipline, a social group, a society limits the ways of looking at the problems it deals with, if it restricts the questions it asks, it narrows the amount of knowledge it can work with, the sources of its progress. Thus, if there are indications, that certain perspectives are deemed less worthy to pursue, the community and the society are well advised to clarify whether this is really the case. Similarly, if potentially shunned points of view are within the valid spectrum, disciplines, communities and society are well advised to ensure that biases are reduced or at least measures implemented to compensate for the effects of the biases.

Advertisements

16 thoughts on “Orthodox academic biases and heterodox rectification? What?

  1. Very thought provoking. I have two basic thoughts about this whole issue. I think diversity is very good and I would certainly agree that having a larger diversity of viewpoints in academia would be good. However, I do think we have to be very careful about blurring the lines between diversity associated with what in the UK would be called “protected characteristics” (gender, race, ….) and diversity associated with differing opinions. The latter – in my view at least – is very different to the former. That’s not to say that more diverse viewpoints wouldn’t be good for academia, but how we address this has – in my opinion – to be very different to how we might address gender imbalance (for example). An obvious issue seems to be that political viewpoints can easily change on a timescale much shorter than a typical academic career.

    As far as bias goes, I’m certainly one who has perceived this as being an attempt to introduce a new bias. Maybe that is wrong, and this is more about trying to introduce viewpoints that would lead to more diverse ideas being addressed. Okay, that’s possible, but even your own example seems to imply that they’re talking about bias in a study itself

    To be more specific, my (mis)understanding of the argument is, that if a methodology is biased against conservative viewpoints, involving a conservative researcher in the development of the study could prevent this bias, or the opinion of a conservative reviewer could result in more thorough discussions of the implications of this potential methodological bias.

    not some kind of bias against addressing certain issues. Also, if there are areas of research that could be interesting, have impact, and generate funding, it’s rather surprising if people are not doing so simply because of some inherent bias. That’s not an attitude I’ve experienced in academia.

    To be fair, I’m fairly new to this whole idea myself and so I’m sure there is more to it than I’ve yet realised.

  2. On your 1. Yes, there is a difference between diversity of viewpoints and diversity of race/gender/…. However, I am not sure how to define the difference. If we put religion to race/gender we already have included different value-sets/ethics in the categories of diversity.

    What does it mean, there is a difference? Do methods to increase gender/ethnicity-diversity not work? Are these methods taking a sledgehammer to crack a nut? I cringe a bit when reading some of the Heterodox Academy’s proposed solutions. But I think I am wrong to do so. Is political diversity less important than the other kinds? If we talk about equal opportunities, is it really so obscure to include political viewpoint? Let’s leave aside the topic of hiring policies, not least because political-de-biasing of hiring depends on self-declaration.

    Someone commented recently on a different type of lack of diversity that it is not so much about enabling the diversity in the first place but about making the diverse set of [anything] welcome and really include them. These are also the solutions which don’t make me cringe. Political attitude may very well remain in the closet. It should be common sense not to disrespect whole sectors of the political spectrum by the language we use. Argue with a (under-/post-grad) student about the evidence and about your viewpoints, take them serious not only even though but because they are conservative. Confront your potential biases. “Liberal” and open-minded should be … at least related. Besides acknowledging potential biases it likely requires to search for different viewpoints on methodological implementations, research-proposals, etc., which leads us to …

    2. With respect to bias. The point is, and I think all the climate-psychology studies point also in that direction, we have inherent biases. These may slant how we view a topic. A differently biased observer, collaborateur, internal/external peer-reviewer may ensure that this bias is reduced – once more, the topic is not introducing a new bias, but reducing a potentially existing one. (My example may have been less than optimal)

    Yes, maybe it is not so much a bias against certain topics. But it is about a bias in how we ask questions and what we consider interesting, and this may very well differ between different viewpoints. “If it provides funding, it is studied” works only under two conditions: you have to have the idea to ask the question, the reviewer of the proposal agrees it is an interesting one. If, potentially, either or both is/are not the case, it is unlikely that the question is pursued (in the short term).

  3. What does it mean, there is a difference? Do methods to increase gender/ethnicity-diversity not work? Are these methods taking a sledgehammer to crack a nut? I cringe a bit when reading some of the Heterodox Academy’s proposed solutions. But I think I am wrong to do so. Is political diversity less important than the other kinds?

    IMO, it is very different. It’s the difference between a characteristic over which you have no control, and one over which you do have control. If anything our whole democratic system is based on the idea that you are allowed to be biased politically; to have a viewpoints that disagree (sometimes strongly) with views held by others. However, we currently live in societies (well those of us who are discussing this topic) where being biased against women, homosexuals, people from different race groups, …. is regarded as unacceptable. That’s why I don’t think we should be looking to address the issue of political diversity in the same way as we might try and address something like gender balance.

    I’m not arguing against political diversity as such, but suggesting that we shouldn’t be suggesting that there is an equivalence between this type of diversity and others.

    Yes, maybe it is not so much a bias against certain topics. But it is about a bias in how we ask questions and what we consider interesting, and this may very well differ between different viewpoints.

    But isn’t this the point? This implies a fundamental bias in how a particular piece of research may be carried out. Addressing this by adding different viewpoints then seems to be suggesting that we resolve this bias issue by adding a different type of bias, rather than by encouraging people to minimise how bias might influence how they carry out (and report) their research.

  4. I should have added that I largely agree with the second part of your response to 1. There is a difference between being open to different viewpoints and trying to be inclusive, and actively trying to redress some potential lack of diversity. My one comment would be that encouraging such behaviour would be good; actively banning the alternative though might be counterproductive. Of course, if academics are bullying more junior staff members, or students, that should not be simply discouraged, but should be actively dealt with.

  5. It is different.

    What about sexuality? What about religion? Is there a “hard” separation? And, even if the characteristics are very different, would it cost us so much effort to address it?

    Yes, we should be careful to differentiate between the different categories, but that shouldn’t hinder us to address it.

    I try hard to avoid the word “discrimination”, but I fail. Let’s follow the UN (taken from wikipedia:Discrimination): “Everyone is entitled to all the rights and freedoms set forth in this Declaration, without distinction of any kind, such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.” In turn, if there is implicit or explicit discrimination of political viewpoints, stop it? But we agree on that.

    On the other point, I wonder why/whether we write at cross purposes. I’ll try once more and hope I don’t move the goal: It is not about adding a different bias, it is about trying to identify and to remove a potentially existing bias, it is about formulating research questions in way so that they do not bias the results in a certain direction. Rewording: It is about debiasing research as far as possible.

    The Heterodox-paper-summary gives examples [disclaimer: I am not necessarily able to follow all of their arguments]. I think their argument is not for balance for balance’s sake, the proposed solutions are generally in favor of openness, not in favor of perfect political balance or even reverse discrimination.

    My thought from the “target”-post is that from debiasing likely follows an increase in our knowledge.

  6. What about sexuality? What about religion? Is there a “hard” separation? And, even if the characteristics are very different, would it cost us so much effort to address it?

    Yes, in the UK I think there is a hard seperation. I’m no expert at this, but I think that gender, race, sexual orientation, and religion are legally regarded as “protected characteristics”. Political ideology is not. Therefore I do not think there is a legal requirement to reduce political bias in the workplace. I could be wrong about that.

    On the other point, I wonder why/whether we write at cross purposes. I’ll try once more and hope I don’t move the goal: It is not about adding a different bias, it is about trying to identify and to remove a potentially existing bias, it is about formulating research questions in way so that they do not bias the results in a certain direction. Rewording: It is about debiasing research as far as possible.

    My argument is that even if this is the purpose (and this is a purpose that I would agree with) I don’t see how this is addressed in the manner in which they’re suggesting. I’ll see if I can explain, also without moving the goal posts. To me there is a difference between an explicit bias in a specific piece of research, and a bias in terms of what research is actually undertaken. The difference being that one might be biased in terms of what you choose to research without being biased in how your present what you do research. If the argument was that we should add viewpoint diversity so as to consider a broader range of research topics, then that might make sense. However, they do seem to be implying that the current bias introduces a bias in how a particular piece of research is carried out and hence that this bias becomes evident in the research that is presented. That’s why I don’t quite see what they’re suggesting as a solution to reducing bias, because it sounds much more like they’re saying that the current makeup of these fields introduces a bias that we can counter by adding more viewpoints. Unless they think that conservative are less likely to be biased (for which I don’t think there is any evidence) then it sounds very much like they’re arguing for adding to the bias in order to average it out, rather than arguing that we should be finding ways to minimise the existing bias.

  7. one might be biased in terms of what you choose to research without being biased in how your present what you do research.

    Sure.

    they do seem to be implying that the current bias introduces a bias in how a particular piece of research is carried out and hence that this bias becomes evident in the research that is presented

    Yes, in my understanding, one of their arguments is that their fields tend to formulate their research topics, methodologies and results in a biased way because of the political bias in the community. My interpretation of their solution is that a more diverse set (i.e. beyond a simple con-lib-divide) of researchers and peers in individual institutes and throughout review-processes would lead to (a) more thorough discussing of the potential biases in the research, (b) a less biased formulation of the research topics and the research methodology (e.g., questionnaires). A hypothetical example from a field closer to our normal-topic would be the following: someone “biased” towards thinking the slightly colder mid-last-millennium was mainly due to volcanic forcing and someone thinking it was mainly due to solar forcing try to optimise the research strategy to solve the question. If either alone tried the optimisation, their inherent biases in weighing the importance of evidence might bias their results towards an outcome favoring their individual hypotheses. Together they may be able to compensate their individual biases. Even if there are no individual biases, the increased diversity likely strengthen the quality of the outcome.

  8. I won’t continue this further, but I’ll just comment on this

    Yes, in my understanding, one of their arguments is that their fields tend to formulate their research topics, methodologies and results in a biased way because of the political bias in the community.

    If this is true, then it indicates poor methodology and practices. This is something that should be discouraged/addressed. My issue would be that you do so by actually addressing the underlying methodologies and practices, not by suggesting that adding people with a different set of political bias would somehow rectify this. I don’t think they would. I might be wrong.

  9. The starting point of the Heterodox Academy is simple: The vast majority of social psychologists are, in US terms, liberals; and this probably affects what research is done, what proposals are funded, and what papers get published. The founders of the HA, although liberals themselves, recognize this as a problem and what to do something about it.

    The situation in economics is different. The orthodoxy is not defined along political lines, but along methodological ones.

    The problem in the environmental and geosciences is, I believe, much like the problem in social psychology. Professor Ken Rice is one example. He clearly despises people who disagree with his political views, which are to the left/green end of the political spectrum. While that does not affect his research on astrophysics, it does mean that a young astrophysicist who is not so left and green, would feel uncomfortable working at the University of Edinburgh — after all, we share a lot of social functions with our colleagues, and our political views have a direct impact on how we treat our students. And Professor Rice may well be appointed to the hiring and promotion committees in the environmental sciences.

    The first step towards solving a problem is recognizing that you have one.

  10. The situation in economics is different. The orthodoxy is not defined along political lines, but along methodological ones.

    Brilliant, in Economics it’s just methodological rather than an indication of a lack of diversity. Nonsense.

    Professor Ken Rice is one example. He clearly despises people who disagree with his political views

    What utter bollocks. Please don’t confuse my dislike of you with a dislike of people who might hold similar views to you. That a Professor of Economics could make such a rookie mistake is utterly bizarre, but in your case, no great surprise given that you seem to revel in going around the social media making things up about other people and their work. Some people might be reluctant to actually call that fundamentally dishonest, but not me. I’m more than happy to publicly state that I regard you as one of the most dishonest people I’ve ever had the misfortune to encounter.

    it does mean that a young astrophysicist who is not so left and green, would feel uncomfortable working at the University of Edinburgh

    This is utter nonsense, of course. For numerous reasons, one of which is that you’re of course being entirely dishonest about my actual position. As I mentioned above, though, this is no great surprise given that you are one of the most dishonest individuals I’ve ever encountered. Additionally, we live in a democratic society in which expressing one’s opinions in public is allowed. This is a good thing. If it happens to put off someone who is too narrow-minded to work somewhere where people hold different opinions to them is a particularly poor argument for curtailing free speech.

    And Professor Rice may well be appointed to the hiring and promotion committees in the environmental sciences.

    Unlikely, but irrelevant – obviously.

    The first step towards solving a problem is recognizing that you have one.

    Words to live by, Richard, words to live by.

    Oliver, feel free to post this, or not. It’s your blog. It doesn’t need to degenerate into a stupid tit-for-tat with Tol and his idiocy. I don’t particularly care if he goes around smearing me on the internet since I seriously doubt that anyone credible takes him seriously anymore.

  11. 1. Heterodoxy in econ is clearly defined. Wiki helps.

    2. If you two want to take your personal issues here, no thank you. I am willing to make a new blog where only you two can comment on an empty post. Please say so if that is needed.

    3. Recognizing a problematic bias is an important step in any field.

    4. A lot of people take Richard serious in general though they may disagree on specifics.

    5. Leaning of institute putting off potential employees. See my post-growth example.

  12. 1. Yes, but how does that illustrate that their are issues with heterodoxy in some areas of Social Science but not in economics? Given that one of the claims in the Heterodox academy is that political bias influences methodolgies, it would seem to be as prevalent in economics as elsewhere.

    2. Not interested. My only reason for commenting was because Richard is stating things about me that are not true. If you’re happy to have people do such things on your blog, that’s your choice. I would typically moderate such things on my blog, but then that’s me.

    3. Yes, and I’ve never said that diversity wouldn’t be good are that recognising a bias isn’t an important thing to do. That’s not my argument.

    4. Yes, but many do not.

    5. What’s that got to do with me?

  13. 1. As far as I can see nothing besides that the words mean something different in economics, i.e. wordplay. In my understanding noone did comment in this thread on whether there are political biases in economics or not.

    3. The environmental (in contrast to the industrial) part of the Earth-sciences may have a liberal bias, and some appear to prefer not to ask whether that is the case.

    5. That was the largest part of your previous comment as far as I understood your comment.

    2. IMO there is one critical statement in Richard’s comment. Due to me not moderating Richard’s, there are IMO three in your comment. Each of these statements says more about its author than about its target. I am not going to snip either, but if you want to go on, do it elsewhere.

    ————————
    Could you two ask Edinburgh’s climate community to moderate a pub-meeting between you two.
    ————————

    Before anyone comments: please think thrice whether your comment contributes anything to the topic of the main post.

  14. 1. Richard did, at least implicitly.

    2. Facts versus opinions. In fairness, I could have made it clearer that my second reference was an opinion, but since that was referring to the first, I thought it was obvious. Since Richard stated someone as true, that he cannot know to be true, I’m happy to stand by it. Also, what you allow people to say on your blog also reflects on you.

    3. Okay. My point is not that there isn’t a bias, but that if this bias is influencing how people approach their research, adding new biases is unlikely to be the solution. This isn’t, however, an argument againsy diversity, but an argument that diversity alone is not a solution.

    5. Huh? My comment was a response to Richard suggesting that someone may choose not to work at my institution because of what I may have said. Given that I’m neither an institution, nor representing one, this seems irrelevant. What someone may choose to do because of something I may have said in public, in my private capacity, has nothing to do with me. In fact, the only reason you know the institution with which I’m associated is because Richard has told you (you don’t know it from me other than I haven’t denied it).

    Could you two ask Edinburgh’s climate community to moderate a pub-meeting between you two.

    If I ever meet Richard in person, I will say to his face what I’ve said on your blog. A pub-meeting between us is unlikely to go well. I certainly wouldn’t waste my time doing so.

    Before anyone comments: please think thrice whether your comment contributes anything to the topic of the main post.

    Feel free to moderate, delete, anything I’ve said. It’s your blog.

  15. Economics has long lived with political heterogeneity. We celebrate Marx and Hayek, Malthus and Ricaro, Friedman and Stiglitz, Tinbergen and Becker, Krugman and Mankiw.

    Economists, however, have very strong views on how research is done, and people who think differently are not welcome.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s