Travelogue:
It’s been a busy, fun, and productive 10 days since I last
wrote. The non-work highlight of this
period was having Talia here for much of last week during her fall break from
Vassar. Though she was hobbled by a
badly sprained ankle (who told that kid to play ultimate, anyhow?) and wearing
a walking boot, we still managed to explore a bunch of the city, with a big
assist from Diane who drove us around on Wednesday afternoon.
Among other adventures and just hanging out, Talia and I went to a lovely
Impressionist exhibition at Musee des
Beaux Arts, and ate some great food (Montreal deli, poutine, classic
French, and Taiwanese wonton soup). After
I dropped Talia off at the train station on a spectacular Saturday morning, I
took a big walk (about 10 km) through old Montreal, the historic port, back up
through Chinatown and to my apartment.
Then on Sunday, Diane took me and Andre, a new post-doc from Brazil (via
the University of Texas) to the Jean-Talon Market and back to her house in the
Town of Mount-Royal, an independent borough surrounded by the city of
Montreal. There we ate the goodies from
the Market (joined by Denis, Diane’s husband) and had a thoroughly enjoyable
late afternoon and early evening. My
French/Quebecois is improving—people aren’t immediately switching to English
when I engage them in conversation, and one merchant at Jean-Talon actually
complimented my French. Here is a selection of pictures from the week.
Psychologue: On the work side of things, I submitted one
manuscript, am close to submitting another, received encouraging new data from
my RA in Arkansas, gave what was apparently an enjoyable and coherent
colloquium at the Center for Research in Human Development, and started
planning a collaborative project with Diane and several other new colleagues
here at Concordia. Not bad, not bad at
all. So I think this blog has served very well one
of the purposes I hoped it would. I’m
writing more and more quickly than I have in a while, thinking pretty damn lucidly, and I’m fired up about what I’m getting accomplished and what’s
ahead. The (still preliminary) data on
the “privileged information” study is in particular very exciting, showing a
classic (for developmental psychology, at least) age by condition interaction.
In other words, children of different ages are showing different
patterns of responses to the two types (conventional vs. privileged) of
information. Even better, they are
doing so in a manner that is consistent with our hypothesis. As the kids would say, “W00t!!”
Of course this is
just the beginning of the scientific story.
Several of the questions following my colloquium last week were of the
form, “Well couldn’t result X be a function of factor Y?” This is par for the course. Science, and behavioral science in particular,
proceeds incrementally and competing/alternate explanations for
findings—especially those findings that are new or counterintuitive—must be
considered and tested for. These
competing explanations can fall into several categories, the most common of
which are 1: procedural/methodological (what if you did it THIS way), 2: experimental design issues (most notably a confound,
a variable that is unintentionally manipulated along with your independent
variable of interest), and for my field in particular, 3: some other (usually
unmeasured in the experiment) previously demonstrated ability or effect that
could be responsible for the new findings.
All of these can be—and usually are—legitimate explanations for or
arguments against a particular research finding that then must be discounted
via further experimentation, additional analysis of data, and/or logical
scientific arguments such as “It can’t be factor Y because Smith & Smith
have shown that factor Y is unrelated to phenomenon X.” In fact, I am designing a study to handle
one such alternate explanation for these new results.
However, these
sorts of arguments against new findings are one of the reasons many have
perished in the Publish or Perish game.
The field of psychology is particularly tough on its own members, as
most of the top journals have rejection rates of somewhere between 80% and 90%
and many other respected journals say no to over half of all papers
submitted. Now, the types of arguments
I’ve listed above are legitimate and necessary to help ensure that
well-designed, replicable studies that serve to advance a discipline are the
ones that are published and widely disseminated for the benefit of the science
and, hopefully, of humankind as well.
However, there are many other types of arguments that are levied against
research findings that have absolutely no place in the scientific
enterprise. Here are a few that many
of us have encountered over the years:
1.
This
can’t be right because it’s not what MY studies show. Not good enough. Please refer to categories 1, 2, and 3 above.
2. We tried that and it didn’t work.
I agree that failures to replicate are really important and are
underreported, but didn’t I just fail to replicate your failure to replicate? Mangez-le.
3.
You’re just reinventing the wheel. Right.
Like alloy rims aren’t any different from wooden spokes.
4.
Well I know a kid who can/couldn’t do that so
I don’t believe you. Sorry, we
report group data and probabilities. There are almost certainly kids like this
represented in my sample.
5.
What if scenarios. Counterfactuals, tall tales, just-so stories,
and thought experiments fit here (apologies to my philosopher friends—er, former friends—for that last
one). Look, I love speculation and
argument for the sake of argument as much as the next guy (depending on who the
next guy is) but as arguments against a novel finding or, worse yet, against
publication, these just burn me. There
are more than enough ways to argue a point legitimately without resorting to
make-believe. Buy me a beer and I’m MORE
than happy to consider an impossible premise like, “What would happen if a
hurricane formed at the end of October and is projected to take a path never
before recorded and you are scheduled to travel in precisely that direction at
precisely that time?”
Merde. Je suis fourre.
No comments:
Post a Comment