variables for a study
research tradition, theory, availability of equipment & techniques
reliability
ability to produce similar results when repeated; reliability of
- physical measure: physical variables
- Population estimates: opinion, attitude; problem of estimating average
- judgements: degree of agreement among observers (interrater reliability)
- psychological test: correlation between scores
test-retest reliability
same test twice, long time in between
parallel-forms reliability
parallel rest as second test –> less reliable
split-half reliability
2 parallel forms in one single test
accuracy
measure that produces results that agree with known standards
validity
extent to which it measures what you intend to measure
face validity
how well a measure instrument appears to what it was designed to measure
content validity
how adequately the content of a test is
ecological validity
measures what people done in real life
criterion related validity
how adequately a test score can be used to infer an individuals value on some criterion measure
- -> concurrent validity = scores of two measure test at the same time correlate?
- -> predictive validity = score predicts related behavior in the future?
construct validity
test measures underlying theoretical construct?
high: participants who score high on the test also behave as predicted by the theory
range effect
occurs when values of a variable have upper or lower limit
- -> floor effect = decrease of mean, task too difficult
- -> ceiling effect = increase of mean, task too easy
behavioral measures
record behavior: frequency, latency, number of errors
physiological measures
blood pressure, heart rate
self-report measure
rating scales, reliability & validity problems (demand characteristics, cooperative attitude, negative attitude, social desirability)
implicit measures
measures attitude, prejudice –> not under conscious control
measurement error
- systematic measurement error (bias): questions are formulated wrongly, low validity
- random measurement error: person is tired, distracts, bad day (environment), low validity & reliability
experimenter bias
behavior of experimenter influences results
expectancy effect
threats to internal and external validity
- -> single-blind technique = experimenter does not know which experimental condition a subject had been assigned to
- -> double-blind technique = experimenter & participant don’t know what they are assigned to
- -> automating experiment
frequency method
record number of times that behavior occurs in period
duration method
record how long behavior lasts
interval method
record if behavior occurs in different time periods
time sampling
scan group for a specific time period
