Experimental Designs Part 2 Flashcards
What would threaten internal validity?
History Reliability of Measures Mortality Maturation Regression to the Mean
Unreliable measures can do what to a study?
Invalidate
What is another name for external validity?
Generalizability
An RCT only considers the average of what?
Group of Subjects
What is the most commonly used design in research?
Pretest-Posttest randomized experimental design
In the classic experiment subjects are evaluated before and after the intervention so that what between groups can be considered?
Pre-treatment Differences
What are test factors in pretreatment differences between groups as a covariate?
Analysis of Covariance (ANCOVA)
The ANCOVA test statistically removes the effect of what from the analysis?
Covariates
What compares with an alternate form of treatment?
2-group pretest-posttest design
What type of trial can groups not be compared after randomization because no pretest is used?
Post-test only randomization controlled trial
What is often used when several explanatory variables are involved in a study?
Factorial Design
Factorial design is useful to determine if any interaction exists between?
Variables
Explanatory variables are categorized as what?
Factors
Levels
When treatment is provided to one group, while the other group receives a placebo or alternate treatment is known as a?
Crossover Design
With crossover design, group assignments are switched at some point in time without who’s knowledge?
Doctor or Subject
With crossover design each group receives what?
Active and alternate treatment
What are the crossover design limitations?
Carry-over effects
High Dropout rates
Treatment sequencing
The therapeutic effects of the first intervention continue during the second intervention in a cross over design is known as what?
Carry-Over Effects
There are high dropout rates in a crossover design because there are what?
2 or more periods of treatments
What is treatment sequencing?
When pt’s respond differently when treatment 1 is given before treatment 2 that if the order is reversed
What design is similar to the randomized designs minus random assignment to groups?
Quasi-Experimental Designs
The lack of randomization is a major factor that makes claims about causality based on quasi-experimental evenidence what?
Doubtful
What designs don’t utilize randomization or a comparison groups?
Non-Experimental
Non-experimental designs are not capable of determining what?
Effect of an intervention
Non-experimental designs include?
Survey and observational research
Case studies and case series
What design has large sample sizes, large time ranges, safety estimations, cost effectiveness and risk/benefit?
Cohort Studies