Repeated Measures Designs: Advantages, Disadvantages, and Types, Study notes of Design

Download Repeated Measures Designs: Advantages, Disadvantages, and Types and more Study notes Design in PDF only on Docsity!

Repeated Measures Design

Mark Conaway October 11, 1999

Common design:

Repeated measures designs

• Take measurements on same subject over time or under different conditions.
• Same basic idea as a randomized block design: - treatment effects measured on ``units'' that are similar as possible.

Repeated measures designs

Disadvantages

• May not be feasible
• May not give realistic assessments of treatment effects
• Analyses more difficult
  • usually need to take into account associations between observations taken from same individual

Repeated measures designs

Cross-over Designs

• Subjects receive every treatment
• Most common is ``two-period, two-treatment''
  • Subjects are randomly assigned to receive either
    • A in period 1, B in period 2 or
    • B in period 1, A in period 2

Cross-over designs: Example

• Treatments: Impermeable (IP) / Semi- Permeable (SP)
• Outcomes: Skin temperature, heat storage, oxygen consumption
• Protocol:
  • 6 men studied under both types of clothing.
  • 3 men randomized to order (IP, SP), 3 men to (SP, IP) Rissanen and Rintamaki (1997) Ergonomics p. 141-150.

Cross-over designs: Example

• Why a crossover design and not a completely randomized design?
• Would expect large amounts of variability in heat storage, oxygen consumption, etc. from different men.
• Would expect small variability in these measures from the same man at two different times

Cross-over designs: Example 2

Possible designs

• Completely randomized?
• Randomized block?
• Cross-over:
  • Each subject observed under each condition
  • Randomize order.
  • One week period between observations.

Cross-over designs: Example 2

• Precision determined by variation in ``time to exhaustion'' by a subject over multiple occasions. - Avoids basing precision on variation in time to exhaustion between different subjects

Completely randomized design or randomized block design or a cross-over design?

• Is the natural variability within a subject likely to be small relative to the natural variability across subjects?
• Are there likely to be carry-over effects?
• Are there likely to be ``drop-outs''?
• Is a cross-over design feasible?

Completely randomized design or randomized block design or a cross-over design?

• No definitive statistical answer to the question.
• Answer depends on knowledge of
  • experimental material and
  • the treatments to be studied

Measurements over time

• Important to consider individual subject profiles over time.
• Ignoring individual subjects can give misleading impression of - variation - direction of effects

Ignoring individual patients can misrepresent variation Modified data from Crowder and Hand. Analysis of Repeated Measures

Ignoring individual patients can misrepresent direction of effects

Ignoring individual patients can misrepresent direction of effects