Flashcards in SCC Chapter 2 Statistical Conclusion Validity & Internal Validity Deck (14):

1

## Validity is a property of inferences, not...

### Designs or inferences

2

## Three questions to ask in exploring threats to validity

###
1. How would this threat apply in this case?

2. Is there evidence that the threat is plausible rather than just possible?

3. Does the threat operate in the same direction as the observed effect, so that it could partially or totally explain the observed findings?

3

## Ruling out threats is a ________ enterprise

### falsificationist

4

##
Statistical conclusion validity concerns what two related statistical inferences?

###
1. whether the presumed cause and effect covary

2. how strongly they covary

5

## What are 8 threats to statistical conclusion validity?

###
1 Low statistical power

2 Violated assumptions

3 Fishing & the error rate problem

4 Unreliability of measures

5 Restriction of range

6 Unreliability of treatment implementation

7 Extraneous variance in experiment

8 Heterogeneity of units

9 Inaccurate effect size estimation

6

## Fishing & the error rate problem refers to

### inflated type I & II error rates

7

## Restriction of range refers to

### floor or ceiling effects

8

## Unreliability of treatment implementation refers to

### poorly standardized treatment

9

## Extraneous variance in experiment refers to

### distracting noises, administrative changes, etc.

10

## Heterogeneity of units refers to

### obscures covariation between treatment and outcome

11

## Inaccurate effect size estimation refers to

### outliers, wrong effect size measure

12

## What are 8 Methods to increase power?

###
1. Use matching, stratifying, blocking

2. Measure and correct for covariates

3. Use larger samples

4. Use equal cell sample sizes

5. Improve measurement

6. Increase the strength of treatment

7. Increase variability of treatment

8. Use a within-participants design

13

## What is Internal Validity

### inferences about whether observed covariation reflects a causal relationship in the form in which the variables were manipulated or measured.

14