The Book of Na'Lon

or rather, Inane Ramblings of an Expatriot

Previous Entry Share Next Entry
Project talk
mesmall
na_lon
I am trying to get my project written up tonight and tomorrow. It's not gone as planned, and I have had a few shocking lessons in what happens when you use a methodological technique you don't have the faintest clue about. I like what my data seem to suggest, and I think I can write a decent report, but I am suffering from near-deadline inertia. Must overcome this before panic sets in. So, in an attempt to make myself get started, I am writing this entry.

The whole project started with a few observations that I had made during my time teaching and learning statistics, which I then threw together with a few research findings from the teaching and learning literature which doing a post-graduate certificate in Learning and Teaching in Higher Education had brought me into contact with.

That teaching stats is not particular bed-of-roses is a well-known phenomenon in the teaching community. No matter how well you teach the courses tend to be evaluated badly, because there is a tendency for students not to differentiate between mode of delivery and material covered. Few students get the chance to compare one mode of delivery with another for the same material, and as a result they may not recognise good teaching even if it came and bit them on the nose. This is something that occurred to me today as a student who’d failed the second level methods module last year and is repeating it this year commented how she thought my teaching style was better than that of the lecturer who’d taught the course the previous year. I am not sure it would be safe to assume that what she says is so. My teaching may just seem better because on her second attempt she is better able to understand the tasks in question because she is learning something that is already in part familiar.

But back to the initial problem of teaching stats. One of the thoughts I had about this in summer 2003 when I had to redesign the first-level methods module at my previous place of employment was that many of the moans that I remembered hearing about stats both as student and as teacher were related to the fact that it all seemed little to do with psychology, the subject that the students wanted to study. It also came hand in hand with number – often a scary proposition. Even giving examples to try and show students how the material related to serious psychology did not seem to make any difference. It was all alien and difficult and frustrating. They sit through the modules because they have to to progress; they try to pass their assignments and all the while wish they didn’t have to.

For me the breakthrough with statistics came when I started analysing my own data. When the stats and methods related to something I wanted to figure out an answer for. Maybe if I could give students that experience? Motivate them by getting them interested in the outcomes of what we were studying? Maybe if we could investigate something they were interested in anyway?

Motivation is one of those things. Elusive and so important. What makes people motivated? In the literature, there exists a distinction between extrinsic motivation – like money, grades, approval from others – and intrinsic motivation – motivation that comes from inside us, because we have an interest, or because we have something at stake.
In redesigning my modules in summer 2003, I tried to take some of these considerations into account. I decided to bring an element of ‘ownership’ of the material studied into the course from the very first session. For five session during which the students were introduced to some of the basic issues in experimental design and data analysis, this element was related to each students own research idea, but then I thought it would be a good idea to cast off the safety-net and to venture out into uncharted territory with my first year group.

In total we carried out 5 projects from start to finish, from design through data-gathering and analysis. I also used the projects to hang on the study of some basic study skills like referencing and literature searches, so students would experience these aspects of the research process in context. The assessments for this part of the course were primarily research reports – a traditional method of assessment for methods courses. As much as possible I used the students’ own ideas to illustrate different types of research and statistical analysis and I guided them through the research process.

The grades for the course weren’t dramatically better than previous cohorts’ grades had been, but I had experienced the course as a relatively successful experiment. But the real litmus test of the success of the venture would be found only in the students’ experiences of what they had been through.

I decided to gather some qualitative data – as opposed to my usually preferred quantitative approach – because responses to open questionnaires might provide me with richer data giving and insight into the students’ experience. Maybe I should have interviewed them for even richer data, but I wanted them to be able to complain about the course, if they felt they needed to. But since we – the students and I – got on together rather well, I feel that maybe they ended up being less critical than they might have been if they didn’t like me. But I can only go with what they have written.

I used some of the basic techniques underlying grounded theory, which takes qualitative data and ‘mines’ it for possible meaning, trying to create a theory from the data to explain what might be going on. The first step was to read and re-read the data to get a feel for what the students were expressing, then I engaged in a process of ‘open coding’ in which different propositions and utterances made by the questionnaire respondent are given categories which attempt to capture fundamental aspects of what is being said. For example, I might take a sentence like (N.B. not an actual example)

“I like carrying out the small projects from start to finish, it made you feel you were discovering something new.”

And code it like this:

“I like carrying out the small projects [doing the research] from start to finish [the whole process], it made you feel you were discovering something new [discovery].”

Once I had carried out this coding process for all six questions on the six questionnaires that were completed, I took different code headings and gathered together related statements from across all questions and questionnaires to help identify the themes that kept reoccurring.

Some of the main themes fitted in with my initial ideas about motivation and the course, some findings were unexpected and didn’t quite fit in with what I thought might crop up and I had to go back to the literature to try and make sense of what I had found.

But the description of the results and the discussion will have to wait for another lj entry. Now I have warmed up a little, I’ll try and write some of the actual report.

  • 1
Oh, please write it up soon, that was really interesting and thought provoking!

This sounds really interesting. :)

Stats was always something I found a nightmare. I still do - I will never have a natural feel for it - but I would have tried harder and got a lot further if I'd understood why I should have bothered and how useful it would be later when I was working as a scientist.

Making teaching relevant to peoples personal goals (whilst bearing in mind they may not be aware of everything they are going to need!) should be the basis of teaching. Otherwise, whats the point.

Qualitiative data is I think often as revealing as quantitative when you are dealing with something as complex as motivation. It gives you a feel for whats really going on in peoples minds - which may not fit into a preexisting theory.

  • 1
?

Log in

No account? Create an account