Introduction to Power Analysis: May 14, 9 a.m. to 12 noon, 5628 Math Sciences (Visualization Portal), sign up
here .
This workshop will introduce the concepts necessary for conducting power analyses on a conceptual as well as a
mechanical level. Suggestions for ways to increase statistical power, other than simply increasing the sample size,
will be given.
Note: Non-UCLA researchers may attend this workshop only if space is available for
all interested UCLA researchers. There is no online component to this workshop,
and this workshop will not be recorded. The notes for this workshop are
here.
Multiple Imputation Techniques in SAS: May 21, 9 a.m. to 12 noon, 5628 Math Sciences (Visualization Portal), sign up
here .
The purpose of this seminar is to discuss commonly used methods for handling missing data and common issues
that could arise when these techniques are used. In particular, we will focus on one of the most popular
methods called multiple imputation. We will demonstrate the two main methods for implementing multiple
imputation in SAS which are multivariate normal and the fully conditional specification also known as
chained equations.
Note: Non-UCLA researchers may attend this workshop only if space is available for
all interested UCLA researchers. There is no online component to this workshop,
and this workshop will not be recorded. The notes for this workshop are
here.
Item Response Models in Stata 14: May 28, 10 a.m. to 12 noon, 5628 Math
Sciences (Visualization Portal), sign up
here .
This talk will demonstrate the new models for Item Response Theory in Stata 14. The new -irt- command
will fit 1PL, 2PL and 3PL models for binary items as well as many popular models for ordered and
unordered categorical items. It will even fit hybrid models for combinations of binary and categorical items.
-irt- is also compatible with the -svy- prefix so parameter estimates and standard errors can be adjusted
for complex sampling designs. We will also demonstrate how to detect differential item response (DIF).
Note: Non-UCLA researchers may attend this workshop only if space is available for
all interested UCLA researchers. There is no online component to this workshop,
and this workshop will not be recorded.
The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California.