Evaluating Quantitatve Research Flashcards
Internal consistency
Consistency of scores within a test, or degree of interrelatedness among items (extent to which items on measure are measuring same thing)
What is internal consistency assessed by
Cronbach’s coefficient alpha, ranging from 0-1
- 0 not related
- higher items covary the greater the internal consistency
- affected by number of items
What are good and poor example of internal consistency
.85 good
.45 poor
What is test-retest reliability
- method for estimating reliability
- coefficient between scores on repetitions of the same test
- if scores stable over time, more reliable
Can a measure be reliable, yet not valid or valid and not reliable
Reliable, yet not valid
Internal validity
Claim of researchers that any change in an outcome is a result of their intervention and not other factors
What are internal validity threats?
Factors that may alter the dependent variable separate from the effects of the independent variable
Mortality (experimental drop out)
People leaving study because of boredom, sickness, injury, inconvenience, discomfort, etc.
Testing
Participants may be one familiar with test, and this might influence future performances
Diffusion of treatments
If participants in experimental and control groups communicate with each other communication can impact scores
Blind study can fix
Halo effect
Researchers have some expectation about performance of participants and are in position of assessment
Maturation
Participants in study might mature naturally or change in many ways over time
History
Events other than the experimental treatment can affect results
Regression
Participants who have extreme score (high or low), scores naturally change in direction of mean
Selection bias
When groups are formed without random assignment, possibility that they will be biased