Martin, Lenora and others, on the divide between quantitative and
qualitative methods. I would like to go with Vera's push for both at once..
It seems to me that this is an area where theory has had a dangerous
effect. The popular tendency to wrap research procedures up in
theoretically opposing bundles creates illusions of difference and suggests
that we must sign allegiance to one camp or the other.
Not only is it possible to use multiple methods in any one study, but it is
unavoidable. Quantitative concepts are deeply embedded in qualitative
procedures. Whenever we decide that something occurred or didn't occur,
that there was more or less of it, that it was kindly rather than hurtful,
that it was red rather than green, we are engaged in quantitative analysis.
It is not possible to use language without using the categories and
quantities embedded in it.
The same argument can be made for qualitative judgements being buried
deeply in quantitative analyses. Martin alluded to them as 'assumptions'.
I like Aristotle's notion that the problem should determine the method
-It might be supposed that there was some single method of inquiry
applicable to all objects whose essential nature we are endeavouring to
ascertain ... in that case what we would seek would be this unique method.
But if there is no such single and general method ... our task becomes
still more difficult. In the case of each different subject we shall have
to determine the appropriate process of investigation. (de Anima, 1:1)
I thought one of the important messages in Mike's Cultural Psychology was
that there is a constant search for ways to include, remain aware of every
aspect of what we are investigating.
I have just had a letter from a researcher on the other side of the world
asking if I considered my research to be quantitative or qualitative. I
find it very hard to know how to answer such questions.
Graham
Graham Nuthall
I would like to re-visit this post a piece at a time to try to make a
series of points which i think is crucial:
>Martin, Lenora and others, on the divide between quantitative and
>qualitative methods. I would like to go with Vera's push for both at once..
>It seems to me that this is an area where theory has had a dangerous
>effect. The popular tendency to wrap research procedures up in
>theoretically opposing bundles creates illusions of difference and suggests
>that we must sign allegiance to one camp or the other.
The notion that qual and quant methods are best used in conjunction is fast
approaching the state of 'received wisdom' at least in educational research
circles. Graham seems to be warning us that any attempt to keep these
modes of inquiry separate is at heart an ideological move. In other words,
theory has become the tool of ideologues on each side. "Wrapping research
procedures up in theoretically opposing bundles" as Graham puts it, seems
to be missing an essential point -- research methods exist outside and
independently of theoretical perspectives.
With all due respect, I cannot imagine being more wrong on this matter.
Research methods are a form of reasoning, and all modes of reasoning depend
very crucially on both ontological and (less often cited) logical
assumptions and positions. Graham might retaliate by saying, but surely
observing is observing and counting is counting, and tallying is tallying.
Well, I say, yes and no. When Phil Jackson, coach of the Chicago Bulls, is
watching a basketball game and I am watching a basketball game, we are
literally not seeing the same thing. and it is not only a matter of
experience; it is a matter of both assumptions and logic. Counting is
counting; but what you choose to count, and the way you count and the way
you reconcile the outcome of the counting have both logical and ontological
outcomes. So, as much as we might want to brush away those things that fly
under the vague and general banner of theory, we cannot afford to do...
>Not only is it possible to use multiple methods in any one study, but it is
>unavoidable. Quantitative concepts are deeply embedded in qualitative
>procedures. Whenever we decide that something occurred or didn't occur,
>that there was more or less of it, that it was kindly rather than hurtful,
>that it was red rather than green, we are engaged in quantitative analysis.
>It is not possible to use language without using the categories and
>quantities embedded in it.
>The same argument can be made for qualitative judgements being buried
>deeply in quantitative analyses. Martin alluded to them as 'assumptions'.
The above statement is only true in a very qualified sense, and once we
acknowledge that qualified sense, then we open the door for determining a
genuine and authentic distinction between these two modes of research.
Qual and quant assumptions can only be inter-related, in the way that
Graham describes, if they share the same logical base. And furthermore,
that logical base is supported by a common ontological position. And the
common ontological position that Graham assumes is Nominalism. From a
nominalist frame, the distinctions between kindly and hurtful, and red and
green, are differences of degree of some more simple orienting frame. Take
the Phil Jackson example from above -- a nominalist would say that the
difference in perspective between Phil and me is traceable to the
differences along certain factorial dimensions. All we need to do as
inquirers is to isolate those factors and chart out their operation --
hopefully in a controlled fashion, but often in a quasi-controlled manner.
Therefore, any inquiry that focused on issues of meaning, on Phil and
myself as entities that are real in a holistic and cohesive way, is at best
a means to an end for finding those ultimate constituitive factors that
will finally settle what is actually going on, if only in a probabilistic
fashion.
Unllike nearly all researchers in education and other social venues, I
reject Nominalism for a more informed contemporary model of Realism that we
might call Semiotic Realism (in honor of Peirce, who explilcitly championed
these ideas). For Peirce, things can be real in three important ways:
they can be real as open possibilities, they can be real as existing
phenomena, and they can be real as mediated entities. In contrast, the
Nominalist can only take the second category seriously. The first category
is dismissed out of hand, and the third category is constantly being
re-defined as aggregates of existing phenomena. When people attack
quantitative assumptions and models, I think they are really trying to
attack Nominalist thinking and positions, without being able to articulate
what it is they want to do. Likewise, qual people are constantly at risk
of falling into old and familiar nominalist assumptions and thinking
patterns. It is no wonder the waters are so muddy.....
I could go on and on, but let me sum up here. The so-called qual vs quant
debate is a debate between nominalism and realism. These positions are
totally at odds with one another -- to see this consider the vast
differences between the Realist Peirce and the Nominalist Rorty, in spite
of the fact that each calls himself a 'pragmatist.' The people, such as
Graham, who say that the two methods can be reconciled, are simply calling
for a return to nominalism pure and simple. I think that the idea of a
genuinely realist and semiotic approach to empirical inquiry is extremely
exciting, and i have no desire to reach any reconciliation with nominalism
(which i feel is at heart fatally flawed).
end of rant :-)
gary shank
shank who-is-at duq.edu