Posted by: wolf | December 11, 2007


Continuing (somewhat) the guiding theme of the previous post…

I believe there is a general problem with a lot of research in psychology (and probably other research domains too, but I don’t really know) in that people very often stick to the methods they have learned in undergrad and graduate classes, but don’t really look into totally different methods once they have settled for some general research questions. Of course, people do use newer developments, like when somebody has learned to do linear regression and afterwards learns structural equation modeling; but that does not really change the frame of reference.

When I started to think about the problem of an optimally compressed representation of a number of data points (and remember, I am a total rookie in those issues) versus estimating the parameters of some pre-specified function like a linear regression, I realized that psychologists like me do have a very much fixed mindset about regression problems – very often people just estimate the linear fit and sometimes polynomials, but they already shy away from transformations. Of course, a straight regression line is one of the most compressed representations one can think of, but the miproblem


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: