SCREEN-L Archives

January 1994


Options: Use Monospaced Font
Show Text Part by Default
Condense Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Film and TV Studies Discussion List <[log in to unmask]>
Mon, 31 Jan 1994 00:25:33 EST
Message of Sun, 30 Jan 1994 15:04:09 EST from <[log in to unmask]>
Film and TV Studies Discussion List <[log in to unmask]>
text/plain (34 lines)
Norman and others:  Sorry the Kuleshov experiment report was published
in 1982 and is available only as a journal article in JOURNAL OF
BROADCASTING 26(sUMMER 1982):685-696.
The actual experiment was not a simple experimental/control group
experiment; it was seriously more complex (multiple Latin square,
or something like that is the technical description, I think).
The statistics were as complex as the experimental design, using
a four-way analysis of variance.  Four-way analysis yields tricky
and sometimes confounding results.  In this case the results were
in the direction generally predicted by Kuleshov and other theoretical
considerations.  What made it interesting was that variables other
than simple picture content and order of presentation affected the results.
I'll take my evidence from wherever I can, as long as I can have
some confidence that it is approximately reliable and valid.
I don't quite understand the experiment that Norman describes and
shall reserve comment for when I do get to understand it and the
It was not my intention to get into a debate about experimental vs.
other kinds of evidence.  I mentioned the paper since the discussions
of Kuleshov "experiment" that I'm familiar with are flabby and
not basically richer than Pudovkin's original 1924 report.
Finally:  This is a topic that I haven't had occasion to be interested
in since completing the paper in 1981 (i.e., 12 years ago).
Cal Pryluck, Radio-Television-Film, Temple University, Philadelphia
<[log in to unmask]>  <PRYLUCK@TEMPLEVM>