Measurment Error Confounds Flashcards
Random error
Obscures results
Constant error
Biases results
Extranious variables
Variables out of the scientist control
Confounding variables
Extranious varaibles that disproportionately affect one level of the IV more than the other
Introduce treat to internal validity
Selection
Bias resulting from selection or assignments of participnats to different levels of the IV
History
Uncontrolled events that take placce between testing occassions
Maturation
Intrinsic chnages in the characteristic of participants between different test occassions
Instrumentation
Changes in the sensitivity or reliability of measurment instrument during the course of the study
Reactivity
Awarness that they are being observed altering behaviour
Precision
Exactness (consistency)
Accuracy
Corectness (truthfulness)
Reliability :precision (consistency)
The extent to which our measure would provide the same result under the same conditions
Validity(accuracy or truthfulness)
The extent to which it is measuring the construct we are interested in
Test-retest reliability
Measures flactuations from one time to another
Beware of order effect
Inter-rater reliability
Measures flatuations between observers
Parallel forms reliability
If we administer different versions of our measure to the same participants would we achieve the same results?
Beware of order effect
Internal consistency
determines whether all items (ex: in a quesionnaire) measure the same construct
bewre of order effect
Content Validity
Does our test measure the construct fully
Face validity
does it look like a good test
Criterion Validity
Does our measure give results which are in agreement with other measures of the same thing? (aka previous studies)
Concurent - comparison of new test with established test
Predicitve - does the test predict outcome on another variable
Construct validity
Is te construct we’re trying to measure valid?
Convergent validity
Connected to tests of the same and related constructs
Discriminant validity
doesn’t have a connection with test of different or unrelated constructs
True Causation
It can only be established when neccesity and sufficiency criteria have been satisfied
Random sample
each member of the population has an equal chance of being selected
usually quasi-random
Systematic
draw from the population at fixed intervals, problematic in populations with a periodic function
Stratfied sample
Proportional: specified groups appear in numbers proportional to their size in the population
Disproportional: specified groups which ar not equally represented in the population are selected in equal proportions
Cluster sample
reseracher samples an entire group or cluster from the population ofi nterest
opportunity/Conveniece sample
people who are available atm
Snowball sampling
recruit a small number of participants and then use those initial contacts to recruit further participants
Reactivity
Awareness they’re being observed may alter behaviour
