Happy birthday!

[altariel]

Happy 70th birthday Q methodology, the method for investigating intersubjectivity for the discerning if marginalized social scientist! It is elegant, powerful, and bloody good fun if you like that sort of thing.


Q methodology is a method of statistical analysis which is chiefly concerned with human subjectivity. A Q study aims to elicit the multiple accounts or understandings that surround a particular theme, issue, or topic, and investigate the shared nature of these accounts by revealing these patterns (factors) in an interpretable and structured manner.

Participants in a Q study are invited to consider a number of 'items' [1], and to rank them according to a matrix which indicates the extent to which they agree or disagree. These 'Q sorts' [2] are taken from a variety of respondents, and are then correlated and factor-analyzed. The resulting factors represent clusters of people who have ranked the same statements in essentially the same fashion, i.e. that have a common understanding of the topic under investigation. These factors are then interpreted in terms of commonly shared perspectives.

For good explications of Q see Brown (1980), McKeown and Thomas (1988), and Stainton Rogers (1995).

The fabled Blake's 7 fan Q study.

Notes

[1] An 'item' is a statement about the issue under study.
[2] A 'Q sort' is the full collection of items related to the topic of study; a completed Q sort is the ranked ordering of all statements as carried out by an individual respondent.


References
Brown, S.R. (1980) Political Subjectivity: Applications of Q Methodology in Political Science. New Haven: Yale University Press.
McKeown, B. and Thomas, D. (1988) Q Methodology. Beverly Hills: Sage Publications.
Stainton Rogers, R. (1995) 'Q Methodology.' In Smith, J. A., Harré, R. and Van Langenhove, L. (Eds.) Rethinking Methods in Psychology. London: Sage.





On June 28, 1935, William Stephenson penned the following letter to the Editor of the British science journal Nature, thus initiating the development that has come to be known as Q methodology. The letter eventually appeared in the 24 August 1935 issue of Nature (p. 297).

Technique of Factor Analysis

Factor analysis is a subject upon which Prof. G. H. Thomson, Dr. Wm. Brown and others have frequently written letters to Nature. This analysis is concerned with a selected population of n individuals each of whom has been measured in m tests. The (m)(m-1)/2 intercorrelations for these m variables are subjected to either a Spearman or other factor analysis.

The technique, however, can also be inverted. We begin with a population of n different tests (or essays, pictures, traits or other measurable material), each of which is measured or scaled by m individuals. The (m)(m-1)/2 intercorrelations are then factorised in the usual way.

This inversion has interesting practical applications. It brings the factor technique from group and field work into the laboratory, and reaches into spheres of work hitherto untouched or not amendable to factorisation. It is especially valuable in experimental aesthetics and in educational psychology, no less than in pure psychology.

It allows a completely new series of studies to be made on the Spearman 'central intellective factor' (g), and also allows tests to be made of the Two Factor Theorem under greatly improved experimental conditions. Data on these and other points are to be published in due course in the British Journal of Psychology.

W. Stephenson
Psychological Laboratory,
University College,
Gower Street,
London, W.C.1.
June 28.

Long may we continue to correlate persons instead of tests.

ETA: iainjcoleman's quick and dirty B7 Q study analysis.

Comments

[(Anonymous)]

Happy birthday.

*clunk* (that was my tea-mug engaging in toasting activity)

Mr A.

[kalypso-v.livejournal.com]

This sort of Q beats Star Trek's any day!

I enjoyed reading the Blake's 7 Q study again. Though, tying up with the current discussion on Freedom City, I have absolutely no recollection of the title of Account 2, despite recognising at least three quotes of mine further down the page. This probably means that they weren't mine at all.

What happened to Kiwi Separatism? Did it turn into Carry On?

She loves Q yeah, yeah, yeah

[communicator.livejournal.com]

Handily 'Q' can be readily substituted for 'You' in all celebratory songs: 'Happy birthday to Q', 'We wish Q a merry christmas', 'For Q's a jolly good fellow... ' er, back to the drawing board.

[altariel.livejournal.com]

*clunks with coffee mug*

[altariel.livejournal.com]

Glad you enjoyed re-reading the study. I did a presentation on it only last summer, one of my most successful pieces of work.

My recollection was that you came up with the title of Account 2! (Not famed for my memory, however.) I changed the title of Kiwi Separatism to Carry on Up the Rebels for the wider audience, but it is the same one.

I absolutely love that icon.

Re: She loves Q yeah, yeah, yeah

[altariel.livejournal.com]

Thank Q for the music...

[kalypso-v.livejournal.com]

I just dug up my actual reply from April 1998, and I'm right about the three quotes! Still can't remember the title, which I see wasn't your original. If it had been a few years later I'd have sworn it was Executrix.

Thank you, I'll do it again, then!

[altariel.livejournal.com]

What was my original title for that account? (My hard drive with all my posts on about it died a sorry death a few years back and I stupidly hadn't backed up my email archive.)

[kalypso-v.livejournal.com]

Boys will be Boys. I'll send you my copy in case it's any use.

[altariel.livejournal.com]

Thank you, that would be great.

Excellent new icon. I'm struggling to think of one for Tennant and so far only have "Gaudis", which is crap and ludicrously obscure. Perhaps I should wait till he's done some episodes.

[espresso-addict.livejournal.com]

A darn good technique which deserves a wider audience. I must have (just?) missed the B7 Q study; the conclusions make an interesting re-read.

[kalypso-v.livejournal.com]

Thanks again! It took me time to come up with the word, and I'm still not quite sure, but I thought it worked on about three levels (Troughton must have said "goodness" at some point, mustn't he?) and I loved the pic even if it meant smaller type.

I don't think I'll bother doing all ten, though - just my Top Three! But it may be like babies, you need to see them before you know what their name is.

Having said that...

[executrix.livejournal.com]

If it had been a few years later I'd have sworn it was Executrix.

It was--I popped in en route to the time travelers' convo at MIT last month.

[altariel.livejournal.com]

Oh, bravissisma!

[altariel.livejournal.com]

It really does deserve a wider audience. I posted the results in May 1999, and I think I started the research about a year before that. Took me ages to work through it - well, it was being done in my spare time.

[altariel.livejournal.com]

TARDIS powered E'trix!

[executrix.livejournal.com]

Considering how large I am on the outside, that's a frightening thought.

[espresso-addict.livejournal.com]

I don't have my e-mails from then anymore, but I think I joined FC sometime late in 1999. Now you've got me trying to work out which account fits me best...

[kendokamel.livejournal.com]

Woohoo!

Now, let's see if that motivates me to do any work on my research proposal tonight.

[altariel.livejournal.com]

I hope it does! Just remember that Q is where all the cool kids hang out! :-D

[altariel.livejournal.com]

You may have been found on an account as yet undiscovered...

[espresso-addict.livejournal.com]

3, with a hefty dash of 4, seems to fit the bill.

[kalypso-v.livejournal.com]

Actually, I'd be quite tempted to do it again to see if my ratings had shifted at all.

[espresso-addict.livejournal.com]

Be very interesting to analyse a different series & see whether the same accounts emerged.

[altariel.livejournal.com]

The main difficulty with analyzing another show would be the number of episodes. 52 is a very good number for a Q-set. Although what matters most is familiarity with the items (here, the episode titles), so there's no theoretical problem per se... having said that, I have difficulty connecting, for example, DS9 episode titles with content.

[altariel.livejournal.com]

I would have predicted at least a bit of Beautiful Suffering in the mix!

[espresso-addict.livejournal.com]

Me too, but Sarcophagus is among my least faves, & Blake is the only one of the other positives that I really rate.

[espresso-addict.livejournal.com]

Whistles innocently -- isn't that what episode guides are for?

[altariel.livejournal.com]

That's true. There are some decent bits of software for doing Q online now, I wonder if that would be easier than the cut-out-and-keep version I did (people printed off a list of episodes, cut it up to do the sort manually, and then emailed their responses to me).

[espresso-addict.livejournal.com]

Easier, but much less fun! I'm guessing that you'd get far more responses from an online form than one that required people to e-mail.

[altariel.livejournal.com]

Someone's developed a version where you can drag the items around the screen, just like doing the real thing.

[espresso-addict.livejournal.com]

Nifty!

[kerravonsen]

Yes, with only 29 respondents to the original study, I wouldn't be surprised if there were more accounts to be found.

[altariel.livejournal.com]

29 is a little bit on the low side for a Q study, but it's sampling the concourse that really matters (i.e. the information that goes to make up the items of the Q-set). Here I had a very well-defined Q-set (the list of episodes), which I think makes up for the slightly low number of participants: four of the five factors were very strongly defined. I guess some more participants might have added clarity to the fifth one, and IIRC there was a sixth, very significant factor I couldn't interpret - more participants might have helped with that. Or thrown up a completely new account!

[altariel.livejournal.com]

Oh, you have changed the font size! It works very well.

[kalypso-v.livejournal.com]

IIRC there was a sixth, very significant factor I couldn't interpret

Now I'm really intrigued!

[kalypso-v.livejournal.com]

Yes, I decided I was being feeble when, unlike with Gnorthernness, there was no reason why I couldn't fit it in with a bit more fiddling. Thanks again!

[altariel.livejournal.com]

Statistically significant, I should say. Unfortunately, I couldn't interpret it in any way. I can't remember why off the top of my head - I suspect it didn't have a single person exemplifying it, and so I had no extra material in comments to help me make an interpretation.

I've used a different piece of software in Q-studies since doing this one (a lot of which I number-crunched by hand!), and that gives some extra stats which sometimes help; perhaps I might have squeezed an interpretation out that way. Loving that icon, btw.

It's tempting to do one on how fans perceive 'quality' in fanfiction, but I think I'd be mad to try.

[katlinel.livejournal.com]

Here's to happy memories of fiddling with scissors and slips of paper!

[altariel.livejournal.com]

It's the only way to do Q really. None of this fancy online gubbins.

[katlinel.livejournal.com]

Oh yes, because, all this fancy-schmancy-pants, online malarkey would probably snoot at my browser.

Besides, scissors, paper - it's Blue Peter Q. And that's very important.

(Still hankers for the days of Changing Rooms when they showed you how to decorate a plain lampshade like an Advent crown.