Assessment week 4 Flashcards
What measurement is the amount of error in any given score?
Standard error of measurement (SEM)
True or false: the lower the reliability of the measure, the higher the standard error of measurement (SEM)
True: the reliability of a measure has an inverse relationship with the standard error of measurement (SEM)
True or false: larger SEMs have smaller confidence intervals
False: larger SEMs have bigger confidence intervals
True or false: the more confident you are, the larger the interval (range)
True: The larger the standard deviation, the larger the confidence interval, and the more confident you are
* So, 95% (2 SD) CI will be larger than 68% CI (1 SD)
What measurement allows us to determine if there is a significant difference between two scores on two different domains?
The standard error of difference (SEdiff)
What term conveys the consistency of a measure?
Reliability
What term conveys how well a measure assesses what it claims?
Validity
What type of validity refers to the degree to which evidence and theory support
interpretation of test scores entailed by proposed uses of a test?
Construct validity
What type of validity refers to whether the measure contains appropriate items for the construct?
Content validity
What type of validity analyzes the interrelationships of items?
Internal structure
What type of validity is used to create composite scores?
Internal structure
What type of construct validity tests that constructs that are expected to be related are, in fact, related?
Convergent validity.
example: the scores of a self-esteem test and an extraversion test are likely to be correlated—individuals scoring high in self-esteem are more likely to score high in extroversion. These two tests would then have high convergent validity.
What type of construct validity tests that constructs that should have no relationship do, in fact, not have any relationship?
Discriminant/divergent validity
Example: if discriminant validity is high, scores on a test designed to assess aggressiveness should not be positively correlated with scores from tests designed to assess intelligence.
What is a typical range for a validity coefficient?
0.30-0.40
What type of validity measures how well one measure predicts an outcome for another measure?
Criterion-related validity
What type of criterion-related validity has a lag between when the measures are given and thus measures if a test accurately predicts what it is supposed to predict?
Predictive validity
What type of criterion-related validity is when both measures are given at the same time?
Concurrent validity
What measure indicates how confident we are in predicting future performance?
Standard error of the estimate (SEest)
example: predicting college freshman grades based on ACT/SAT scores
What quantity indicates how far an individual falls from the average?
Standard deviation (SD)
example: how far above or below the mean a score falls?
What is a measure of how confident we are in a person’s score on a given measure at that moment?
Standard error of measurement (SEM)
example: if the person took the same test again, within what range would we estimate the score would be the second time?
What measurement determines how well you identify individuals with the issue you are assessing?
Sensitivity
What measurement determines how well you do NOT identify individuals that don’t have the issue you are assessing?
Specificity
True or false: Specificity is the ability to identify true positives
False: specificity is the ability to identify true negatives
True or false: sensitivity is the ability to identify true negatives
False: sensitivity is the ability to identify true positives
What type of validity can determine whether a measure gives you information you would not get from a different measure?
Incremental validity
What validity-threatening response set is the tendency of survey respondents to answer questions in a manner that will be viewed favorably by others?
Social desirability response set/bias
What validity-threatening response set is the tendency of survey respondents to select a positive response option disproportionately more frequently?
Acquiescence response set/bias
What validity-threatening response set is the tendency of survey respondents to answer questions in a manner that will be viewed unfavorably by others, or that departs from norms?
Deviant response set/bias