Tuesday, April 22, 2014

More research

I doubt very much that this series of blogs about writing my research proposal makes for scintillating reading. I readily admit that they are written primarily for myself. It is interesting for me to read which paths I took, where I made progress and where I made mistakes. I read somewhere that it's good for a researcher to keep a journal of her progress, so this is my journal.

First of all, I want to correct a mistake that I wrote a few days ago. The Myers-Briggs Test Inventory (MBTI) is not the same thing as the 'Big Five'; I confused the two because the Occupational Psychologist (OP) added a fifth scale to her version of the MBTI.

Somewhere along the way, I obtained the idea that one's cognitive style can be obtained from the MBTI solely by looking  at the S-N continuum. I had to go over several papers before I found the one which referenced the idea. Actually, I think that I got the idea from the Wikipedia page on MBTI because the given paper actually said that the cognitive style was derived from S-N and T-F.

It's all well and good making this statement, but I needed a way to include suitable questions in my research questionnaire in order to establish what the respondent's cognitive style is. I retrieved the questions which appear in the OP's version of the MBTI, but these questions were about which job the respondent would like, and seemed totally unsuitable. 

There are many questionnaires purporting to be the MBTI, but only one is the official MBTI. The end result of the test is assigning one of sixteen values (four characters from each continuum) to the person who took the test. Any set of questions can theoretically do so, but presumably the 'official' MBTI is the only one which has been validated - checking by other means to see that the results obtained from the test actually match the respondent's personality. Off-hand, I don't remember the degree of validity for this test - I think it's somewhere around 60%. 

Anyone can string together an instrument containing 60 questions, fifteen to each dimension, then on the basis of that compute an end value. But this value is worthless unless it has been validated.

Looking for MBTI questions on the Internet was a frustrating experience. There are several sites which offer an online test, but each one offers a different set of questions and there is no way of knowing which questions comprise the S-N continuum as each site performs its calculations "offline". Eventually, at 6:30pm yesterday, I struck gold when I found a document meant for "future chief residents" at the Dartmouth Hitchcock Medical Center. The document contains seventy questions, a score card and then a series of explanations for each of the sixteen types. I needed only four pages out of a sixty page document, but those pages are priceless.

Looking at the scorecard, I can see that the S-N continuum is based on twenty questions, as is the T-F continuum. So I may have to add forty questions to the questionnaire, something which I very much do not want to do, or maybe only twenty. I suppose I could add ten questions from one scale and ten questions from the other scale and to hell with the validity. In any case, I don't know what the validity for this complete exam is, and as I write in the research proposal, I am not interested in making a psychological determination for each respondent. I'm only interested in the general sense of cognitive style. 

I need now to add a paragraph or two about "self efficacy", whose definition I am taking to be "the belief in one's capabilities of using a computer in the accomplishment of specific tasks". One might term this "computer confidence".  The paper which I have is very interesting and even shows a fascinating model, but it's completely out of date! The paper was written in 1995 (19 years ago) which might be a short time in psychology but a huge time in terms of computers. In 1995, the average person didn't know what the Internet was; now it is ubiquitous. There were mobile phones in 1995, but they were large and generally only business people had them, whereas today, even five year olds have mobile phones (and possibly even smart phones as opposed to the telephones of yesteryear). I shall have to see how suitable the research questions in the paper are, and maybe update them for my use. Presumably, this may be a factor only for the over 40s: anyone younger than will have grown up with computers and will feel comfortable with them.

I improved a few other parts of the research proposal: I added a much clearer structure to the section on the literature (which is the main part) and I finally figured out what my aims and objectives are. I was unwittingly helped in this by reading someone's PhD thesis; I had hoped that I would get some questions about spreadsheet competency from this thesis. The questions turned out to be of no use as they basically asked each respondent to rank herself (I was hoping for an objective ranking), but the section of the thesis which was concerned with aims and objectives was priceless. I don't mean to say that I copied the aims and objectives from the thesis; I couldn't as they are about a different subject. What was important to me was the linguistic construction of the objectives and their derivation from the aims.

My mentor underwent an operation today so he'll be out of action for a few days. I want to add some more material then send him a new, updated version next week. I also have to address some of the points which he has raised (some of the points arise from his misunderstanding, so I don't have to change anything in my work in order to answer these points).

No comments: