19.542 critical thinking

From: Humanist Discussion Group (by way of Willard McCarty willard.mccarty_at_kcl.ac.uk>
Date: Fri, 6 Jan 2006 10:35:18 +0000

               Humanist Discussion Group, Vol. 19, No. 542.
       Centre for Computing in the Humanities, King's College London
                     Submit to: humanist_at_princeton.edu

   [1] From: Willard McCarty <willard.mccarty_at_kcl.ac.uk> (46)
         Subject: critical thinking about critical thinking

   [2] From: Ryan Deschamps <Ryan.Deschamps_at_Dal.Ca> (45)
         Subject: Re: 19.534 (critical) thinking and button pushing

         Date: Fri, 06 Jan 2006 06:23:14 +0000
         From: Willard McCarty <willard.mccarty_at_kcl.ac.uk>
         Subject: critical thinking about critical thinking

On this subject of some recent discussion, I quote from Alec Fisher,
Critical Thinking: An Introduction (Cambridge, 2001) -- who on page 1
ironically demonstrates the need for what he professes to be
teaching. After noting the putative failure of traditional efforts to
teach critical skills indirectly, in courses dedicated to the
conventional disciplinary subjects, Fisher notes that, "The result is
that many teachers have become interested in teaching these skills
*directly*. This is what this text aims to do. It teaches a range of
*transferable* thinking skills, but it does so *explicitly* and
*directly*.... they will be taught in a way that expressly aims to
facilitate their transfer to other subjects and other contexts"
(original emphasis). I am reminded of my hero Dr Johnson's railing
against cant, i.e. "the particular language or jargon of a class....
a pet phrase, a trick of words; esp. a stock phrase that is much
affected at the time, or is repeared as a matter of habit or form"
(OED). "Transferable skills" is certainly one of our current pets.
But what does it mean?

Consider a skill, say carpentry or brick-laying. What, do you suppose,
would a master carpenter think of justifying what he or she does by
the claim that it is transferable? The implication seems to be that
carpentry is not or cannot be valued for what it accomplishes
directly, rather for how it can be used in some "other contexts" --
unspecified. One would hope that a master carpenter would have deep
insights by virtue of his or her skill into the building of things
and could, perhaps, adapt if wood were not to hand, though only to a
limited degree. (Stone-cutting, for example, is a very different
skill.) But if you're going to hire a carpenter, you'll do so for
the skills in carpentry, and you would certainly look for a carpenter
who actually cared about his or her particular skill and wasn't
minded to be transferring it somewhere else, no?

I am thinking, then, that selling what we do as "transferable"
devalues it and so makes us quite vulnerable. It's always been clear,
has it not, that success in the world does not require an advanced
degree. (I don't suppose I'm the only one who was reminded of this
fact by family members not as privileged as I was, and later by
successful but less well-educated persons than I.) Perhaps it is true, as a
colleague of mine at Toronto used to say, that we need to "package"
ourselves so as to be attractive to those who pay for us. But it
seems to me that *critical* thinking is required to keep the goods
from becoming the packaging.



Dr Willard McCarty | Reader in Humanities Computing | Centre for
Computing in the Humanities | King's College London | Kay House, 7
Arundel Street | London WC2R 3DX | U.K. | +44 (0)20 7848-2784 fax:
-2980 || willard.mccarty_at_kcl.ac.uk www.kcl.ac.uk/humanities/cch/wlm/

         Date: Fri, 06 Jan 2006 06:26:05 +0000
         From: Ryan Deschamps <Ryan.Deschamps_at_Dal.Ca>
         Subject: Re: 19.534 (critical) thinking and button pushing


Although I would like to take a look at how Hacking comes to his
conclusion that
"one enters data and presses a button" and that this has an impact on critical
thinking, but I would say that his comments are somewhat overstated.

Statistical packages (and I am mostly thinking about minitab and SPSS) really
only do the math for a statistical process. The decision to analyse via
t-test, time-series model, parametric or non-parametric etc. still depends on
the user of the package. While I think there are still "critical thinking"
risks, they are the same kinds of risks that one might encounter analysing
Joyce's _Ulysses_ without a computer. For example, since it is very easy to
conduct the tests, one might run a variety of tests and pick the ones that
produce the more interesting results (this might also not be a problem!). In
the same sense, one might be more likely to analyse _Ulysses_ using reader's
response theorem because the book is particularly suited to a reader's response

My larger critical thinking concern comes with more qualitative products like
Nudist (which I have to admit I have never used myself). But again, the
problems are more universal than computing packages. For example, choosing a
taxomony for the organization of themes or concepts has a considerable
influence on the results of the study.

I also have a concern about online surveys, data collection and the like, since
it is very difficult to establish a good understanding of how a survey will
work in such a broad arena as the World Wide Web.

In the absense of statistical software, statistical tests would have to be done
by a statistician, who could, in fact, influence process more drastically than
a computer. For example, a busy statistician could dictate how he/she want the
information organized (to facilitate ease of calculation), and in fact could
have a number of co-workers (typists, clerks, assistants, etc.) that could
potentially have influence as well.

Software also enables others as prone to mathematical errors (like myself) to
scrutinize data and criticize accordingly. And there are plenty of
mathematicians out there willing to scrutinize the tiniest numerical error in a
system (they aren't very nice about it either!).

This is not to say critical thinking errors do not occur. For
example, through
the familiar econometric Granger "causality" test, it can be shown that an
increase in sales of Christmas card causes Christmas to come (not my own idea,
but I do not have the citation).

Also, if Hacking means to say that computing makes it so easy to do statistical
analysis that researchers can be prone to choosing stats for projects that
could be better analysed through qualitative means -- sure, that's a fair
criticism. But, if it is so easy, why not do both?

Ryan. . .

Ryan Deschamps
Received on Fri Jan 06 2006 - 05:55:30 EST

This archive was generated by hypermail 2.2.0 : Fri Jan 06 2006 - 05:55:30 EST