Quantitative Flashcards
Simple Random Assignemnt
participants assigned at random, equal chance of any group
Block random assignment
similar to simple random, assures an even number of participants in each group
Stratified Random Assignment
participants divided into strata to ensure proportional representation of any important characteristics
Allocation Concealment
The person who determine if a subject was eligible for inclusion in the trail was unaware of which group the subject would be allocated to when this decision was made
Standard of Care group
group still receives the standard of care
Participant blinding
participants were unable to discriminate whether they had or had not received the treatment
Therapist Blinding
therapists were unable to discriminate which participants had or had not received the treatment
Assessor blinding
assessors were unable to discriminate which participants had or had not received the treatment
Pragmatic Clinical Trial
Has the following characteristics:
1. Minimal exclusion criteria
2. Looser restrictions on recruitment
3. Treatment is less rigidly controlled, better mimics actual clinical practice
4. Control group receives standard of care
Pre-Test/Post Test Design
Given a pre test, then treatment (or control) and then given post test
Post-Test Only
not given a pre-test beforehand, start with treatment or control, and then given post test.
Factorial Design
More than 1 independent variable
testing to compare the factors together
Crossover/Repeated Measures Design
each experimental unit is given each of the different treatment levels over the course of the experiment.
Washout Period: time between interventions that eliminates effect of intervention on outcome measure
Quasi Experimental, Time Series Designs
One group Pre-Test/Post-Test
One group repeated measures
interrupted time series
Quasi, Non Equivalent Group Designs
Non-Equivalent Pre-Test/Post-Test with Control Group
Non-Equivalent Post-Test only with Control group
Historical COntrol group
One Group Pre-Test/Post-Test Design
Can be used when
1. Previous research has documented behavior of control group
2. time period is short enough to minimize effect of other variables
Repeated Measures Design
Similar to one group pre-test/post-test, but multiple post tests administered
no control group
Interrupted Time Series Design
multiple measurements both before and after treatment administered
multiple pre-tests help minimize time as a potential confounding variable
Non-equivalent pre-test/post test with control group
Treatment and control groups, but no randomization
groups may already be intact or participants may choose their group
Non Equivalent Post Test Only with Control Group
Same as non-equivalent pre-test/post test with control group, no pre-tests administered
Historical Control Group
same as non-equivalent pre test/post test, but control group is from historical data collected at a previous time
Single Subject Designs
group research can hide individual difference and generate results that do not represent the behavior of any individual
can also serve as pre-cursor to RCTs in order to better understand the effect of the treatment on the outcome
each participant receives a treatment and serves as their own control
Single Subject Design Examples
ABA or ABAB Designs
Multiple Baseline
Multiple Treatment
ABA or ABAB Designs
A is the baseline, B is the intervention
data collected at baseline, during intervention, after intervention is removed, and possibly again during intervention
Multiple Treament Design
A = Baseline, B = Intervention 1, C=Intervention 2
Data collected at baseline, during intervention 1, after the intervention is removed, and again during intervention 2
Observational Study Designs
Case-Control
Cohort
Descriptive Study Designs
Cross Sectional
Case Series
Case Study
Types of Experimental Designs
Pre-Test/Post Test
Post-Test Only
Factorial Design
Crossover
