being able to understand biases and limitations with studies Flashcards
Internal validity
Extent of cause and effect claims. How well can the study rule out alternative explanations. How well does the study minimise systematic error or “bias.” Remember that internal validity relates to causality, to correlational research has low internal validity.
External validity
Extent that findings can be generalised.
Was the sample representative? Did pre-testing result in atypical behaviour?
Construct validity
extent to which instruments measure what they claim to measure?
Is the measure reliable? is the content of the measure relevant the variable it is trying to measure?
Statistical conclusion validity
Extent of valid conclusions from statistical inferences
Reliability
• Comparable results over time/observations
• Reliability does not guarantee validity, only a prerequisite for validity
• Reliable measure can be consistently measuring, or being influenced by, the wrong thing
Reliability of a scale is measured using chronbachs alpha, do each of the items in a scale measure the same thing.
Test-retest reliability
test repeated over time
Parallel forms reliability
: parallel forms of test applied over time
Split-half reliability
parallel forms of test applied intermingled
Observer/judgement reliability
Inter-observer reliability
Reliability maximised by
o Careful operationalization
o Adequate control of variables
Random error vs bias
- Random error: inconsistent and balances out, given large sample size and random assignment. should be reduced
- Bias: has a direction and must be eliminated
Convergent validity
Contructs that are expected to be related, are in fact related
Discriminant validity
Constructs that are not expected to have any relationship, in fact do not have any relationship
Predictive validity
A test accurately indicates an employee’s scores on a future criterion (e.g., conscientiousness).
