Working in interdisciplinary teams: Some stumbling blocks and how to deal with them

phd071705s(c) “Piled Higher and Deeper” by Jorge Cham
www.phdcomics.com

I have been working in interdisciplinary teams for a while now — interacting with computer scientists, museum curators, pedagogues, managers, etc. pp. Personally I am a psychologist (Ph.D.) with a strong interest in computers/technology (I can, for example, program iOS apps and have an intuitive connection to technology, yeah, I’m a little geeky 😉 ).

This gives me the possibility to communicate quite efficiently with people with a computer science and technology background, I think. But I also made the experience that a lot of communication problems are very, very difficult to address because you cannot see them easily.

False Agreement/False Disagreement

One problem is that different disciplines use terms quite differently. Take “model” for example, or “learning”. Depending on your discipline the meanings of these terms differ, which can lead to major discussions about them, or be so subtle that you don’t quite get why you don’t understand each other clearly. Disciplines also differ in the way they assign meaning to to concepts, some use definitions, others will want to have an operational definition: you specify how to assess it and this defines the concept. In psychology there’s the additional problem that many terms psychologist use are also used in everyday life — only that psychologists usually mean something quite different when they talk about “identity”, “self”, “learning”, etc. pp. No wonder that some psychologists have started creating their own terms to avoid this, although with rather limited success.

In practice this can lead to false agreement (you think you are talking about the same thing but aren’t) and false disagreement (you disagree, sometimes very emotionally, but you essentially mean the same thing).

So, you need to make your terms clear early in the conversation/project and you have to remind people again and again of the agreed upon definition, because in most cases, people will work more in their disciplines than in the project and they will forget that the term had another meaning there (which can be very frustrating). A shared glossary can help here if it’s easy accessible — don’t overdo this with a specialized solution to dig into conceptual work for a large project. It’s a quick reference guide, not a way to restructure the disciplines. You should explicitly mention if the definition differs from the way it is used in certain disciplines to remind these disciplines that they have to be careful here.

No knowing when they don’t know

I had training in statistics during my studies (and acquired much of it on my own during my diploma and dissertation work) and there are things that I now do almost automatically. There are also things I simply wouldn’t write and that I wouldn’t even consider. For example, in analyzing whether items can be aggregated to a scale I recode items which are reversed (e.g., 7/10 items are of the kind “the higher the value, the more X the person is”, 3 are in the opposite direction, “the higher the value, the less X the person is”, the 3 items are recoded so that the direction of all items is the same). Similarly, when dealing with correlations I wouldn’t think of writing (I think): “X and Y correlate positively, so X leads to Y” (correlation doesn’t imply causality). Problem is that because I do these things automatically, I often do not remember to tell others about it when I am giving them information on how to analyze their data. This isn’t their fault, they cannot know it because they weren’t trained in it. I was, that’s why I am in the project.

But in practice this leads to the situation where things go wrong and you do not notice it, because the people dealing with the data do not know that they do not know something, and you do not know that they do not know. It’s just something you would never expect someone handling data to do and failing to see that this can be a problem for people not trained in data analysis this can be a serious problem.

So, make sure that you know not only what another person from a different discipline doesn’t know, but also what they cannot know that they do not know. If you ever instructed first year students or student research assistants it helps you to get an impression of what people cannot know that they do not know. Get to know them, remind yourself that they were trained differently and handle things differently, according to their discipline.

Of course, this also applies in another direction — there are many things I cannot know, for example, having contact with museum personnel, I was in a room with some exhibits and one person asked me what I thought of an exhibit. I picked it up and played around with it, not knowing that curators usually use gloves to handle exhibits. Luckily it wasn’t anything important but do this in a museum with valuable objects and you make a curator’s heart jump in the wrong direction.

Why do it?

This posting might sound a little negative regarding interdisciplinary research, but if I would think that it was a waste of time, I wouldn’t have bothered writing a posting about it. In truth, I think that interdisciplinary research is in many (but not all) cases the way to go. I mean, a lot of research is highly interdisciplinary if you look at the way art influences science. But even (or rather: especially) with different scientific disciplines interdisciplinary work leads to superior results. You can do things no discipline could do on its own because they lack knowledge, skills, methods, etc. Different disciplines stimulate each other, you see concepts differently and talk about things you take for granted. You have access to tools, methods and expertise you couldn’t have otherwise. Research is less likely to be so specialized that it’s not usable by anyone but experts in that specific discipline. In short: Every discipline has blind spots, strengths and weaknesses and working together interdisciplinary can produce superior results  — if the communication works out.

Explicitly work on the communication

Just putting people from different disciplines into the same room doesn’t work — there’s the contact hypothesis in social psychology which gives helpful information about the conditions that must be met for people in a way to resolve conflict (or in this case: prevent it). For example, researchers from different disciplines must respect each others expertise, accept others as being of equal status/worth. The must have common goals they can only reach together in the desired quality. They need time to get to know each other as persons, not only as “that educator” or “this computer geek who programs my experiments”. And they need the support of the context (authorities, supervisors) for their work.

You can image that in many cases this is difficult to achieve. Finding out what computer scientists consider worthwhile (and it isn’t writing your experimental environment) is very interesting, as is hearing the viewpoints of curators or pedagogues. Different disciplines have different criteria for scientific work, different methods, different standards, but if you respect that and put together all the strengths of the different disciplines you can achieve things that are truly more than the sums of their parts.

Categories: Community Aspects, Doing Science, Science


Post Navigation
next older posting:
next newer posting:


2 Trackbacks & Pingbacks

  1. Perfect Memory and Polymaths in the 21st Century | ORGANIZING CREATIVITY
  2. Method Monoculture in Science | ORGANIZING CREATIVITY

Leave a comment

Your email address will not be published.


*


css.php