Presenting different length texts

[Mark A. Casteel]

I thought I'd respond back to the list and tell folks who are interested
one potential way to solve the problem of presenting stories with different
numbers of sentences. First, let me thank all those folks who responded to
my plea for help and offered assistance: Edward Carney, Natasha Tokowicz,
Richard Anderson, and Anthony Capobianco.

Here was the problem: Some stories might be only 12 sentences long while
other stories might be 14 sentences long. One idea that seemed fairly easy
to implement was simply to include a variable in the stories List object
that tells E-Prime the durations of sentence 13 and sentence 14. Since some
stories will have actual sentences for these variables, the values of the
duration variable (e.g., Sen13dur; Sen14dur) for those stories would be set
to -1 (i.e., wait for subject input). In the duration dialogue box for both
sentence 13 and sentence 14, the value would then be set to [Sen13dur] or
[Sen14dur], respectively. For those stories that are shorter, and don't
contain a sentence 13 or 14, the values of the Sen13dur and Sen14dur
variables would be set to 0. Now keep in mind that E-Prime will actually
log nonsense values under the reaction times for sentence 13 and sentence
14, and these bogus values are not easy to pick out. Therefore, make sure
the data for the Sen13dur and Sen14dur variables get logged by E-Prime. You
can use the values of 0 in these variables as flags that the RTs for those
sentences should be discounted in any data analysis.  I hope this makes
sense and is helpful to those of you who had similar questions. Thanks
again to those of you who helped me out.  -- Mark


*********************************
Mark A. Casteel, Ph.D.
Associate Professor of Psychology
Penn State York
1031 Edgecomb Ave.
York, PA  17403
(717) 771-4028
*********************************