PSYC 3250 MIDTERM QUESTIONS WITH COMPLETE
SOLUTIONS
`how do you calculate percentiles? Correct Answer ·
Percentile: When the distribution of Z raw scores is normal, Z
scores can be directly concerned to percentiles.
· Percentile ranks can be calculated with respect to the test
sample or a normative sample.
oAssume that every personality trait is normally distributed
(most things are)
`measuring psychological traits Correct Answer The first step
in developing a test to measure a trait is to define the trait or
construct to be measured
The 'domain' is all possible behaviours that could conceivably be
indicative of a particular construct
When developing test items, we engage in "domain sampling"
(i.e., taking a sample of all possible behaviours that are
indicative of the trait)
a norm-based score Correct Answer indicates where an
individual stand in comparison with the particular normative
group that defines the set of standards.
a psychological test Correct Answer a systematic procedure
for comparing the behaviour of two or more people
additivity Correct Answer · To measure something, we must
count units of measurement (e.g. number of items answered
correctly on a history test)
o Has to make sense to add those numbers - differences between
them has to be the same (Likert items, time)
, § Does not make sense with rank order.
· To add units meaningfully requires that the units are
psychologically identical
oEach item is equally psychologically meaningful / difficult
big variance Correct Answer very different
classical test theory views reliability in terms of links among
Correct Answer Observed scores
True scores
Measurement error
common issues that arise in item writing Correct Answer Item
length
Vocabulary
Double barreled
Double negatives
Ambiguity of meaning
construct method Correct Answer Start with psychological
theory underlying the construct/ trait
Come up with a definition of what the trait is/ isn't
What behaviours are/ are not a characteristic of the trait?
Content of the item is clearly related to the definition
Use item analysis to determine which items to keep
contrasted groups Correct Answer approach focuses on
looking for differences in the way two groups respond to the
same item
SOLUTIONS
`how do you calculate percentiles? Correct Answer ·
Percentile: When the distribution of Z raw scores is normal, Z
scores can be directly concerned to percentiles.
· Percentile ranks can be calculated with respect to the test
sample or a normative sample.
oAssume that every personality trait is normally distributed
(most things are)
`measuring psychological traits Correct Answer The first step
in developing a test to measure a trait is to define the trait or
construct to be measured
The 'domain' is all possible behaviours that could conceivably be
indicative of a particular construct
When developing test items, we engage in "domain sampling"
(i.e., taking a sample of all possible behaviours that are
indicative of the trait)
a norm-based score Correct Answer indicates where an
individual stand in comparison with the particular normative
group that defines the set of standards.
a psychological test Correct Answer a systematic procedure
for comparing the behaviour of two or more people
additivity Correct Answer · To measure something, we must
count units of measurement (e.g. number of items answered
correctly on a history test)
o Has to make sense to add those numbers - differences between
them has to be the same (Likert items, time)
, § Does not make sense with rank order.
· To add units meaningfully requires that the units are
psychologically identical
oEach item is equally psychologically meaningful / difficult
big variance Correct Answer very different
classical test theory views reliability in terms of links among
Correct Answer Observed scores
True scores
Measurement error
common issues that arise in item writing Correct Answer Item
length
Vocabulary
Double barreled
Double negatives
Ambiguity of meaning
construct method Correct Answer Start with psychological
theory underlying the construct/ trait
Come up with a definition of what the trait is/ isn't
What behaviours are/ are not a characteristic of the trait?
Content of the item is clearly related to the definition
Use item analysis to determine which items to keep
contrasted groups Correct Answer approach focuses on
looking for differences in the way two groups respond to the
same item
