week 3 Flashcards
(22 cards)
Reliability
The extend to which rue measure correctly without measurement error
Validity
The extend to which we measure what we intended to measure without bias (systematic error)
Test - retest reliability
repeat the same measurement with the same participants
Issues with test-retest reliability
How stable is the variable over time? Is second measurement influence by first measurement
Parallel forms reliability
Use similar/equivalent measure with the same participants
Issues with parallel test reliability?
Are the measurement instruments parallel enough? Are the measurement instruments not too parallel?
Interitem reliability
Are all parts of the measurement instrument related ( instrument consists of item that all aim to) measure the same underlie ing construct/concept
Issues with interitem reliability?
Do all items contribute to the measurement of the concept?
Are the items too similar?
Replication
Repeat the whole study and compose the outcomes / same methods, different participants / researcher
Increasing the reliability/ eliminating measurement error
① standardize administration of the measure ② clarify instructions and questions ③ train observers ④minimize errors in coding data
Validity
The extend to which we measure what we intended to measure (without systematic error/bias )
Face validity
Does it appear to measure what it’s supposed to measure
Content validity
Does the measure cover all aspects of a construct? Requires independent observers
Construct validity
Associated in a logical way with instruments that measure comparable (or opposing) constructs (
= convergent) ② No association with instruments that measure (completely) different concepts (discriminant)
Criterian_related validity
Association with a particular behavior al criterion (in the present
= concurrent) in the future (predictive)
Statistical validity
Was data analysis done correctly
internal validity
Are alternative explanations ruled out?
External validity
Is the result generalizable?
Types of descriptive research
① survey = attitudes I lifestyles, behaviors, problems ② demographic = patterns of basic life events: birth, marriage ③ epidemiological = occurrence of disease and death
Cross-sectional
One-shot, cross-section of the population
Successive independent samples
Changes over time, different respondents each time I are samples comparable?
Longitudinal
Changes over time, same respondents more than once
