| xmpl {prefmod} | R Documentation |
Data to illustrate the usage of patt.design for rating scale (Likert type) items.
xmpl
A data frame with 100 observations on 5 numeric variables.
The first three variables (I1, I2, I3)
are the rating scale (Likert type) items with 5 response categories,
ranging from 1 (strong agreement) to 5 (strong disagreement).
I1response to item 1
I2response to item 2
I3response to item 3
SEX(1) male, (2) female
EDU(1) low education, (2) high education
Datasets in data files or Data frames used in patt.design require the follwing structure:
All values must be numeric.
The item responses must be in the leftmost columns (such as I1
to I3 above).
Categorical subject covariates follow the item responses (rightmost columns)
and their levels must be specified as consecutive integers. If in a used
datafile or dataframe these are defined as R factors they will be
converted to integers. This is not possible if characters are used
as factor levels and, consequently, patt.design will produce an error.
des <- patt.design(xmpl, nitems = 3, resptype = "rating", cov.sel = "SEX") head(des)