DOC PREVIEW
CMU STA 36402-36608 - Handout

This preview shows page 1 out of 2 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 2 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 2 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

2/23/2010 36-402/608 ADA-II H. SeltmanHandout #12: Repeated Measures (Part 1)1. Definition of “experimental unit”: the unit to which treatment is applied2. Definition of “multivariate” response models: models that deal with multiple out-come measurements per experimental unit and that appropriately take into accountcorrelation among these measurements.3. Definition of “repeated measures” models: multivariate response models in whichthe same kind of outcome is measured at different times or locations.4. Approaches to multivariate analysis(a) Do separate analyses for each outcome, because they are not strongly related(in terms of meaning and goals).(b) Combine appropriate outcomes on each subject to make a single outcome thatrelates to the goals of the experiment, then analyze by non-RM methods.(c) Use RM methods that model the correlation directly.(d) Use hierarchical or mixed models to model the correlation through randomintercepts and/or slopes. (For repeated measures models only)5. Recognizing study types(a) Longitudinal: the same subjects are measured repeatedly over time(b) Crossover: longitudinal studies in which treatment changes over time (usuallyin “counterbalanced” order)(c) Split-plot experiments: experiments with two or more treatment variables thatare applied at different levels, usually “between subjects” and “within sub-jects”. Also just called “mixed within/between repeated measures” experi-ments6. Next class: Analysis of these models using classic RM methods and Hotelling’s T27. Example: Monkey memory study(a) Eighteen monkeys were trained to discriminate different object pairs at 2, 4,8, 12 and 16 weeks before treatment.(b) Eleven monkeys were (presumably randomly) assigned to the “treated” (as op-posed to “control”) group and had brain surgery to remove their hippocampalformations.(c) After surgery (or no surgery) the monkeys were tested to measure the percentcorrect discrimination of the pairs.(d) The goal was to test the effects of the surgery (and thus the role of the hip-pocampus) on short term vs. long term memory. Short term was defined inadvance as less than 5 weeks.●●●●●●●55 60 65 70 75 80 8550 55 60 65 70 75% Words Remembered After SurgeryShort Term (2+4 weeks)Long Term (6+8+12 weeks)●CONTROLTREATED8. Breakout and


View Full Document

CMU STA 36402-36608 - Handout

Documents in this Course
Load more
Download Handout
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Handout and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Handout 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?