Description Usage Format Author(s) References See Also Examples
Sample data from a visual-world eyetracking study included to demonstrate use of gmpm functions on multilevel data with difficult dependencies. Data are from a subset of Experiment 2 in Kronmuller and Barr (2007). A parametric reanalysis of these data can also be found in Barr (2008).
1 |
A data frame containing 2464 rows, each one an aggregate over multiple trials within the same condition for the same subject.
[,1] | SubjID | factor | experimental subject |
[,2] | Speaker | factor | identity of the speaker (same,diff) |
[,3] | Load | factor | was listener under cognitive load? |
[,4] | AggID | numeric | index of unique speaker*load*subject combinations |
[,5] | ms | factor | time in ms from description onset |
[,6] | t1 | numeric | time in 100 ms increments |
[,7] | T | numeric | count of looks to target |
[,8] | O | numeric | count of looks to non-target object |
[,9] | X | numeric | count of frames including looks to blank regions or blinks |
[,10] | N | numeric | total number for frames (T+O+X) |
[,11] | NT | numeric | total non-target frames (N-T) |
Dale J. Barr <dale.barr@ucr.edu>
Kronmuller, E. and Barr, D. J. (2007). Perspective-free pragmatics: Broken precedents and the recovery-from-preemption hypothesis. Journal of Memory and Language, 56, 436–455.
Barr, D. J. (2008). Analyzing 'visual world' eyetracking data using multilevel logistic regression. Journal of Memory and Language, 59, 457–474.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 | data(kb07)
# first let's fit the data using a binomial model
#
kb07.binom.gmpm <- gmpmCreate(cbind(T,NT) ~ t1*Speaker*Load | SubjID,
"binomial", kb07, ivars=c("Speaker","Load"))
# you will need to increase the number of runs in the command below
# (to, say, 999) to get sensible results
kb07.binom.gmpm <- gmpmEstimate(kb07.binom.gmpm, list(maxruns=19))
summary(kb07.binom.gmpm)
# now let's do a more powerful multinomial analysis
# where we break out looks to blank regions and blinks
# into a separate "junk" category (X)
kb07.mnom.gmpm <- gmpmCreate(cbind(O,T,X) ~ t1*Speaker*Load | SubjID,
"multinomial", kb07, ivars=c("Speaker","Load"))
# again, you'll need to increase the number of runs to get something
# sensible
kb07.mnom.gmpm <- gmpmEstimate(kb07.mnom.gmpm, list(maxruns=19))
summary(kb07.mnom.gmpm)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.