Search for notes by fellow students, in your own course and all over the country.
Browse our notes for titles which look like what you need, you can preview any of the notes via a sample of the contents. After you're happy these are the notes you're after simply pop them into your shopping cart.
Title: Tests & Measurements - Research Methods - Psychology
Description: doctoral level notes. PsyD program, research methods, tests and measurement, raw scores, percentiles, standard scores, T-distribution, WAIS scores, standard deviation, norms, normative samples, reference groups, criterion-referenced tests, performance criteria, floor-effect, ceiling-effect, grade equivalent
Description: doctoral level notes. PsyD program, research methods, tests and measurement, raw scores, percentiles, standard scores, T-distribution, WAIS scores, standard deviation, norms, normative samples, reference groups, criterion-referenced tests, performance criteria, floor-effect, ceiling-effect, grade equivalent
Document Preview
Extracts from the notes are below, to see the PDF you'll receive please use the links above
Notes 6/10
•
•
•
•
•
•
•
Raw scores: Doesn’t mean anything w/o common frame of reference, a number (X) that
summarizes or captures some aspect of a person’s performance in the behavior samples
that make up psychological tests
Percentiles: indicate relative position of individual compared to reference group, ordinal level
of measurement, denote rank but are not set on equal intervals (not evenly distributed in
normal curve)
• quartiles break up percentiles into 4 different areas
• deciles break up percentiles into 10ths
• sensitive to minimum and maximum values
normalized/standard scores: raw scores transformed so they have meaning, in terms of their
position, as if they came from a normal distribution
standard scores: scores that have been linearly transformed to match the shape of the
original distribution but standardized in terms of their scores, allow scores to be compared
across scores, standard but not uniformed across different tests
• t distribution (M=50, SD = 10), z distribution (M = 0, SD = 1), WAIS subtest scores
(M=10, SD=3), WAIS IQ scores (M=100, SD=15)
• z= (raw test score- raw group mean)/raw group standard deviation
• formula for calculating standard score to put them on same metric
• new score = ((score old - mean old)/SD old)SD new + mean new
norm referenced: uses standards based on performance of specific groups of people to
provide info for interpreting scores
• norms: particular reference group
• normative samples: reference group upon which the norms were collected
• must be representative of group trying to study, sizes vary based on test,
recency of sample is important, relevant factors to consider depend on the
nature of the test and its intended purpose
types of norms: standardized normative sample (overall representative), normative sample
(any sample used and corrected to match normal distribution), reference group (any group to
whom we’re making comparisons; subgroup, local, convenience)
• performance criteria (criteria referenced): uses procedures to determine whether and
to what extent the desired levels of mastery criteria have been met, more subjective
than norm-referenced interpretation (require some judgement of what threshold is
needed to meet criteria)
• can require raters meet some threshold of reliability to lessen subjective
effects of judging criteria
• “testing the limits”: cases where sometimes a patient cannot tolerate
standardized assessment, give tasks to test the limit of how much patient will
cooperate to try to assess levels in other areas, unable to compare to norms
because it is not standardized
criterion-referenced tests: some test amount of knowledge, some test levels of competence
in a skill area
• sometimes used in reference to establish relationships between test scores and
future outcomes
• commonalities: meant to assess the extent to which test takers are proficient in
certain skills or knowledge domains, scored such that one person’s performance
doesn’t influence the relative standing of others
• content-related tests: often called achievement tests, planning for these kinds of
tests requires a table of specifications (helps consider universally what is related to
construct must be included relevant to audience and how heavily weighted they are)
Notes 6/10
•
•
•
•
•
typically involve: identifying/describing qualitative criteria for evaluating performance
or product of performance, developing method for applying criteria
Floor effect: test too easy, can’t see a lot of variability
ceiling effect: test too hard, no one can get to top, unknown true range
age equivalent: score on test same as average score on age level, does’t mean individual is
performing at level of specified age - only average score
grade equivalent: score on test same as average score on grade level
Title: Tests & Measurements - Research Methods - Psychology
Description: doctoral level notes. PsyD program, research methods, tests and measurement, raw scores, percentiles, standard scores, T-distribution, WAIS scores, standard deviation, norms, normative samples, reference groups, criterion-referenced tests, performance criteria, floor-effect, ceiling-effect, grade equivalent
Description: doctoral level notes. PsyD program, research methods, tests and measurement, raw scores, percentiles, standard scores, T-distribution, WAIS scores, standard deviation, norms, normative samples, reference groups, criterion-referenced tests, performance criteria, floor-effect, ceiling-effect, grade equivalent