Teaching Research Methods in Psychology

Research Methods can be a bit marmite for some – both teachers and students alike. But love it or hate it we can’t ignore the fact that it’s the backbone of Psychology – and other social sciences for that matter. You can’t really discuss an explanation, theory or model without considering the evidence, and you can’t critically evaluate evidence without a solid grounding in research methodology. Experimental research is often held up as the gold standard of research, the one type of study that allows you to draw causal inferences, to conclude that X does actually lead to Y. 

Therefore it’s right that we devote considerable time ensuring students have sound knowledge of experimental methods, and in this post I want to consider how my approach to teaching this has been influenced by my growing understanding of principles of cognitive science. I think a lot of discussion around research methods teaching revolves around making it ‘fun’ or ‘engaging’ rather than prioritizing the learning (I’ve written more on this here) and I think Psychology in particular has a problem in this regard as I discuss here.

I’ve outlined here a lesson on hypotheses, but the format and underlying principles could be (and are) applied to any part of research methods. This sequence takes me about 60-80 minutes depending on the class (our lessons are 40 mins but majority of 6th form lessons are doubles); if there’s time at the end there’s good opportunity for some additional application examples and practice questions, retrieval practice, or moving on to the next bit. 

Retrieval practice – activating prior knowledge

The lesson starts with some basic questions on the board recapping content from the previous lesson on IVs and DVs. At this point I just want to check they can explain the difference (IV what you manipulate, DV what you measure) and they understand how each should be operationalised. These are answered on mini-whiteboards (MWB) and this means I can quickly scan responses to gauge knowledge. If there are significant gaps at this point I’ll have to take some time to go back and re-teach as they simply won’t get hypotheses without the basics first.

This is then followed by an example scenario they have to analyse (usually a typical memory experiment eg effects of listening to either rock or classical music on recall of a list of words) and write down the IV and DV, fully operationalised, on MWBs. I’m really pedantic on this point and will make sure that their answers really are properly operationalised, sharing good examples from individuals to highlight this, and getting them to self-correct if necessary.

Then I’ll ask them to come up with a list of possible predictions about the outcome of the experiment. At this point I make clear that I’m not asking what they actually think will happen, just to identify all the possible permutations that they could come up with. There should be four:

  • Rock music leads to better recall than classical music
  • Classical music leads to better recall than rock music
  • There is a difference in recall between classical & rock music but we don’t know which way
  • There is no difference between them

Explanations and examples

I don’t want to worry about telling them the names of different types/forms of hypotheses yet, it’s too abstract and I want them to start with something more concrete; explanations work best when moving from concrete to abstract. Once we’ve got all four (which I write on the board in simplified versions) I show them one version of a fully operationalised hypothesis. I state the ‘rules’ of hypothesis writing (must make a clear prediction, both IV and DV must be fully operationalised) and then annotate the example to show how this has been done (image below shows annotation on the slide, this will be animated to appear as I’m explaining each bit. My board writing isn’t what it should be so I prefer this approach sometimes!) 

I’ll ask them to match up my formal version with one from the list I just wrote on the board so they can see the difference in wording and emphasise the level of precision we’re looking for. Then we’ll go through matching this up with the proper terms (alternative vs null, directional vs non-directional) and I give examples of each of the other versions so they can see what they look like. Then we look at another example of each type from a different context which they are already familiar with so they can see the structural differences between them without being overloaded by understanding a novel context.

Modelling and independent practice

Now they’re almost ready to have a go at writing some for themselves. There is a handout to work through with a bunch of examples using a range of research contexts. I want this knowledge to be flexible so they can apply it to any experiment, not just a few limited contexts. Before I give them the sheet I show it on screen. I know the moment they have a piece of paper in their hands some of them will start reading and trying to complete the task but I want their attention on me first (Adam Boxer put this best – “Teach like an attention megalomaniac!”). They read through the first example and then I will then model precisely how I want them to engage with it – underlining key parts and annotating them. 

This is really important for me when teaching students how to deal with application questions – too many students will highlight but not annotate. This means when they get to answering the actual question they have to go back to the stem and waste time & effort working out again what they’ve highlighted and which bits they need. Annotation at the beginning takes that away, essentially ‘outsourcing’ working memory to the page. This is particularly important as methods questions in exams often entail a scenario with multiple questions relating to a range of different features of the stem. They could be asked to identify the IV & DV, write a hypothesis, comment on the experimental design, or any number of other features. By training them to annotate properly even when we have a single focus they are developing valuable skills for the exam.

Once I’ve modelled my annotation I will then write out the hypothesis for them, narrating my thinking and questioning the class as I go:

Only then do I give out the paper version and ask them to copy down my model first – annotation and all – before starting to work through the rest on their own. I’ll let them settle for a minute and then start circulating. If more than a couple are not getting it right on the next one I’ll pause the whole class to bring them together. The most common thing here is they immediately start writing without annotating so I remind them again. Then I let them get on with it, circulating and giving feedback as necessary. I don’t usually need to pause them all again at this point, they will be working at different paces and have individual issues. For those that finish quicker there’s some easy extension work to convert hypotheses from one form to another (which I think is good practice for them), or some additional examples to look at.

Assessing understanding using multiple choice questions

After all that practice, I can be reasonably confident that they’ve grasped the gist and they are usually getting quite fluent in writing their own hypotheses correctly by the end of the task. Now I want to double check some conceptual knowledge – do they *really* understand? I have a multiple choice question which presents a new research scenario followed by four possible hypotheses. One is correct and the other options represent different ways that students typically get this wrong:

After reading a post by Blake Harvard on MCQ design (here) I added an extra level to this. Students next have to explain why their chosen option is correct, and why each incorrect answer is incorrect:

This really allows me to properly assess their understanding and identifies any misconceptions or if they were just guessing – or misread the question. It takes only a few minutes but I think it gives me some really useful feedback on my students developing schema for hypotheses (In case you’re interested / don’t know: D is correct; A is null, B is non-directional and C is poorly operationalised). 

What about practicals? Don’t they get to do an experiment?!

If you ask a bunch of Psychology teachers how they teach methods, invariably someone will mention doing lots of practical activities eg in-class experiments. It’s frequently been suggested at INSET I’ve attended and often accepted uncritically (by me for many years)  as the *best* way to teach the topic. I think there’s definitely a place for them, and do use them in my teaching, but am much more selective about the types of practicals I use and where they fit in the learning sequence. 

Research suggests that disciplinary knowledge isn’t best acquired through disciplinary methods (Kirschner, 2009) and so, somewhat counterintuitively for many, learning about how research methods work is not best achieved by teaching through practicals. In the past for this topic I might have started my lesson with a class experiment and then used it to draw out elements of the method I wanted my class to learn. I found that typically students struggle to dissociate the content of an experiment (eg what it was testing) and the process (ie the different features of the experimental method such as the IV and DV). This led to a degree of cognitive overload meaning they didn’t acquire the disciplinary knowledge I wanted them to as well as I’d have liked. It also invariably takes time to create the resources as well as to actually run in class; while I do acknowledge the other benefits of doing practicals in class, I’m not convinced the opportunity-cost makes it worth it many instances. So the approach outlined above means I am much more confident that they really understand hypotheses before I let them loose on an in-class experiment. 

Let’s summarise the lesson sequence:

Retrieval practice → Explanation → Modelling → Independent practice → MCQ → Application.

As a learning sequence it’s pretty straightforward and I’m not claiming to have reinvented teaching here. I think in the past I’ve used all of these strategies to some extent and I’m sure many others do too. The key for me is that I’ve thought much harder about the how and why of each phase of the lesson, and I’ve thought more carefully about the best way to sequence this in a coherent fashion to best help my students learn. If anyone has any thoughts on how anything could be improved further, or alternative approaches that they’ve found successful then please do share.

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 )

Connecting to %s

Create your website with WordPress.com
Get started
%d bloggers like this: