CH. 3 Flashcards
the degree to which a test measures what it claims, or purports, to be measuring.
Construct validity
the extent to which test items have sufficient breadth to capture the full range of the construct intended to be measured
content validity
whether a test looks like it measures what it is supposed to measure
face validity
interitem vs interobserver reliability
interitem- if the items of a test are all measuring the same, the correlation would be high.
interobserver- degree to whcih rater consistently code the same observed behavior.
scale rating from 0-10
likert scale
parametric vs. nonparametric statistics
parametric- used with variable that are measured on either an interval or ratio scale.
nonparametric- test hypotheses for variable that use either a nominal or ordinal scale for measurement.
randomly divide all items that are supposed to measure the same trait or construct into two sets
split-half reliablity
formed by randomly selecting from relatively uniform subpopulations called strata
stratified sample
useful when a sampling frame of elements is not available, often the case for large populations spread out across a wide area.
cluster sampling
simplest type of nonprobability sampling method. Ex. partcipate in an experiment in your psych course because youwere availible
availability sampling
when one or more questions influence the responses to subsequent questions.
context effects
proportionate vs. disproportionate samping
pro- ensures the sample is selected so that the distribution of characteristics in the sample matches those of the population.
dis- selecting a higher fraction from one group that will help your data
try to infer what the sample population will think
inferential statistics
sequence varies in some regular, periodic pattern
periodicity
requires some procedure that generates numbers or otherwise identifies cases strictly on the basic of chance
simple random sampling
variant of simple random sampling, first element is selected from a list and then every nth element is selected
systematic random sampling
quotas are set to ensure that the sample represents certain characteristics in proportion to their prevalence int he population
quota sampling
some difference in the characteristics of the sample compared with those of the population from which it is selected
sampling error
the list from which the elements of the population are slected
sampling frame
that an association is not likely to be due to chance
statistical significance
