Textbook Notes (280,000)
CA (170,000)
Western (10,000)
PSYCH (5,000)
Chapter 4

Chapter 4 Notes This is for Psych 2060, taken online, for the Muchinsky 9th Ed. textbook. These are very detailed and easy to understand notes on Chapter 4 that I made myself. Headings and bullets make it easy to read, and I assure you there is enough

Course Code
PSYCH 2060
Hayden Woodley

This preview shows pages 1-3. to view the full 14 pages of the document.
Chapter 4 – Predictors: Psychological Assessments
-predictor: any variable used to forecast a criterion
oe.g. barometric pressure is used to forecast rainfall
Assessing the Quality of Predictors – Reliability
-psychometric: the measurement of properties of the mind, the
standard used to measure the quality of psychological
oconsists of reliability and validity
-reliability: a standard for evaluating tests that refers to the
consistency, stability, or equivalence of test scores, often
contrasted with validity
-test-retest reliability: a type of reliability that reveals the
stability of test scores upon repeated applications of the test
ocoefficient of stability used to measure the stability of a
test overtime by correlating the two sets of scores from
two different administrations of a test
oreliability coefficients of around 0.7-0.8 are professionally
oshorter the time interval between the administration of
tests the higher the coefficient
-equivalent-form reliability: a type of reliability that reveals
the equivalence of test scores between two versions or forms of
the test
oalso known as parallel reliability
ocoefficient of equivalence used to measure the extent to
which the two forms are equivalent measures of the same
orarely used because it is hard to come up with one good
test, let alone two
oit is hard to construct two tests whose scores have similar
meanings such that they are truly parallel measures
-internal-consistency reliability: a type of reliability that
reveals the homogeneity of the items comprising a test
ofrequently used in IO psychology to assess homogeneity of
a test
osplit-half reliability
divide items into odd- and even-numbered items
two sets of scores are correlated
if the test is internally consistent, should have high
degree of similarity between responses (number of
right/wrong answers)
oCronbach’s alpha or Kuder-Richardson 20 (KR20)
each test item is treated like a minitest

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

response of each item is correlated with the response
of all the other items
matrix of inter-item correlations is formed whose
average is related to the homogeneity of the test
-inter-rater reliability: a type of reliability that reveals the
degree of agreement among the assessments of two or more
ofrequently used in IO psychology
oalso called conspect reliability
oe.g. multiple job interviewers evaluating job candidates
omuch more subjective
oestimation of inter-rater reliability is usually expressed as a
correlation and reflects the degree of agreement among
Assessing the Quality of Predictors – Validity
-validity: a standard for evaluating tests that refers to the
accuracy or appropriateness of drawing inferences from test
-construct validity: the degree to which a test is an accurate
and faithful measure of the construct it purports to measure
oconstructs are abstractions/ideas that require a real,
tangible way to assess them
oe.g. intelligence, leadership, motivation, mechanical
oconvergent validity coefficients are the correlation
coefficients between the scores from our new test of a
construct and the existing measures of that construct
oscores on the test should not be related to concepts that
we know are not related to our construct
low divergent validity coefficients
otests that manifest a high degree of construct validity are
the most widely respected and frequently used assessment
instruments in IO psychology
oconstruct validation is the process of demonstrating
evidence for five linkages or inferences
X is a measure of construct 1, Y is a measure of
construct 2
link between X and Y is the only one that can be
tested because it is the only inference involving two
variables that are directly measured
assess construct validity by assess links between X
and construct 1, and Y and construct 2
relationship between the two constructs can be used
for constructing theories of job performance

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

last linkage is between x and construct 2 (e.g.
inference between employment test score and job
-criterion-related validity: the degree to which a test forecasts
or is statistically related to a criterion
ofrequently used
oconcurrent criterion-related validity
concerned with how well a predictor can predict a
criterion at the same time
no time interval between collecting the predictor and
criterion data
e.g. predict a worker’s level of productivity (criterion)
on the basis of a test (predictor)
collect productivity data from workers
give them a test
correlate their scores with productivity records
opredictive criterion-related validity
collect predictor information and use it to forecast
future criterion performance
e.g. use a student’s high school class rank to predict
overall college GPA
odetermine whether there is a relationship between
predictor scores and criterion scores
oif yes, use predictor scores to select people on whom there
are no criterion scores
ovalidity coefficient: a statistical index often expressed as
a correlation coefficient that reveals the degree of
association between two variables
desired range of 0.3-0.4
square correlation coefficient to calculate how much
variance in the criterion we can account for by using
the predictor
tests with low validity coefficients are not
inadequate, they simply attest to the complexity of
human behaviour
-content validity: the degree to which subject matter experts
agree that the items in a test are a representative sample of the
domain of knowledge the test purports to measure
oface validity: the appearance that items in a test are
appropriate for the intended use of the test by the
individuals who take the test
oestimate of content validity made my test makers, face
validity made by test takers
You're Reading a Preview

Unlock to view full version