Home
Explore
Exams
Login
Get started
Home
Threats to Validity
Threats to Validity
0.0
(0)
Rate it
Studied by 0 people
Call Kai
Learn
Practice Test
Spaced Repetition
Match
Flashcards
Knowt Play
Card Sorting
1/65
There's no tags or description
Looks like no tags are added yet.
66 Terms
View all (66)
Star these 66
1
New cards
History — Definition
[Internal Validity] External event influences outcome during study
2
New cards
History — Most problematic in
One-group pretest-posttest; time series
3
New cards
History — Control strategy
Add control group; shorten study duration
4
New cards
History — Statistical adjustment options
Interrupted time series modeling
5
New cards
History — Impact on effect size
Inflates or deflates observed effects
6
New cards
History — Impact on power
Reduces interpretability
7
New cards
Maturation — Definition
[Internal Validity] Natural change over time affects outcome
8
New cards
Maturation — Most problematic in
Pretest-posttest designs
9
New cards
Maturation — Control strategy
Use control group; randomization
10
New cards
Maturation — Statistical adjustment options
Growth curve modeling
11
New cards
Maturation — Impact on effect size
May exaggerate treatment effects
12
New cards
Maturation — Impact on power
Minimal direct effect on power
13
New cards
Testing — Definition
[Internal Validity] Pretest influences posttest scores
14
New cards
Testing — Most problematic in
Pretest-posttest; Solomon design
15
New cards
Testing — Control strategy
Use Solomon four-group design
16
New cards
Testing — Statistical adjustment options
ANCOVA
17
New cards
Testing — Impact on effect size
Artificial score improvement
18
New cards
Testing — Impact on power
May inflate effect size
19
New cards
Instrumentation — Definition
[Internal Validity] Measurement tool changes over time
20
New cards
Instrumentation — Most problematic in
Longitudinal designs
21
New cards
Instrumentation — Control strategy
Standardize measurement tools
22
New cards
Instrumentation — Statistical adjustment options
Calibration checks
23
New cards
Instrumentation — Impact on effect size
Biases effect estimates
24
New cards
Instrumentation — Impact on power
Reduces precision
25
New cards
Selection — Definition
[Internal Validity] Groups differ at baseline
26
New cards
Selection — Most problematic in
Non-equivalent control group
27
New cards
Selection — Control strategy
Matching; randomization; ANCOVA
28
New cards
Selection — Statistical adjustment options
Propensity score matching
29
New cards
Selection — Impact on effect size
Over/underestimation of treatment effect
30
New cards
Selection — Impact on power
Lower statistical power
31
New cards
Regression to the Mean — Definition
[Internal Validity] Extreme scores naturally move toward average
32
New cards
Regression to the Mean — Most problematic in
Pretest-posttest with extreme groups
33
New cards
Regression to the Mean — Control strategy
Use control group
34
New cards
Regression to the Mean — Statistical adjustment options
Regression modeling
35
New cards
Regression to the Mean — Impact on effect size
Artificial improvement
36
New cards
Regression to the Mean — Impact on power
Reduces reliability of effect
37
New cards
Mortality (Attrition) — Definition
[Internal Validity] Participants drop out systematically
38
New cards
Mortality (Attrition) — Most problematic in
Longitudinal designs
39
New cards
Mortality (Attrition) — Control strategy
Intent-to-treat analysis
40
New cards
Mortality (Attrition) — Statistical adjustment options
Multiple imputation
41
New cards
Mortality (Attrition) — Impact on effect size
Biases effect estimates
42
New cards
Mortality (Attrition) — Impact on power
Reduces power
43
New cards
Interaction of Testing & Treatment — Definition
[External Validity] Pretest sensitizes participants
44
New cards
Interaction of Testing & Treatment — Most problematic in
Solomon design contexts
45
New cards
Interaction of Testing & Treatment — Control strategy
Use posttest-only designs
46
New cards
Interaction of Testing & Treatment — Statistical adjustment options
Compare pretested vs non-pretested groups
47
New cards
Interaction of Testing & Treatment — Impact on effect size
Limits generalizability
48
New cards
Interaction of Testing & Treatment — Impact on power
Minimal effect on power
49
New cards
Sampling Bias — Definition
[External Validity] Non-representative sample
50
New cards
Sampling Bias — Most problematic in
Quasi-experimental designs
51
New cards
Sampling Bias — Control strategy
Random sampling; replication
52
New cards
Sampling Bias — Statistical adjustment options
Weighting adjustments
53
New cards
Sampling Bias — Impact on effect size
Limits generalizability
54
New cards
Sampling Bias — Impact on power
Minimal statistical effect
55
New cards
Inadequate Operationalization — Definition
[Construct Validity] Poor definition of variables
56
New cards
Inadequate Operationalization — Most problematic in
All designs
57
New cards
Inadequate Operationalization — Control strategy
Use validated measures
58
New cards
Inadequate Operationalization — Statistical adjustment options
Factor analysis
59
New cards
Inadequate Operationalization — Impact on effect size
Weak or inconsistent effects
60
New cards
Inadequate Operationalization — Impact on power
Reduces statistical power
61
New cards
Low Statistical Power — Definition
[Statistical Conclusion Validity] Insufficient sample size
62
New cards
Low Statistical Power — Most problematic in
Small-sample studies
63
New cards
Low Statistical Power — Control strategy
A priori power analysis (G*Power)
64
New cards
Low Statistical Power — Statistical adjustment options
Increase N; reduce error variance
65
New cards
Low Statistical Power — Impact on effect size
Unstable effect size estimates
66
New cards
Low Statistical Power — Impact on power
High Type II error risk