Psychometrics Flashcards
Predictive validity
similar to criterion in that it deals with the predictive nature of a question/task on a test; this type is often used when comparing tests scores on a work related task, to how an employee is actually scored on that task by their employer.
Criterion validity
a measure of how well what is being tested can predict future uses of the same kind; i.e. a personality test may predict a certain behaviour, but it is only valid if this behaviour is later shown in real life.
Content validity
mainly used in clinical psychology, measures an assessment on how well it encompasses the entire part of personality or symptoms it is testing for, and requires authorities to validate that these ‘parts’ make up the whole picture of what an illness, etc. is.
Construct validity
arguably one of the most important in psychometric testing, aims to review whether the test is actually testing what it’s meant to.
The variance of the obtained test scores is equal to the:
The sum of the true score variance and the error variance
The correlation bw 2 sets if test scores indicates the:
Proportion of variance in one test associated with variance in the other test
Standardization
A series of test items administered to people of varying ages, races, and social backgrounds to determine which questions will be retained on the final version of the test
The statistic in a test manual that indicates how close an obtained score is to a true score is the:
Standard error of measurement
Criterion referenced measurement
.
Norm referenced measurement
.
Constant error
.
Nonparametric statistics should be used instead of parametric statistics when the:
Form of the distribution is not known
Sampling error refers to
Tendency of sample statistics to differ from population parameters
Standard error of the mean
Take the variance of the population and dividing it by the square root of the sample size
What is another name for a non-directional alternative hypothesis
A two tailed test
