Week 3 - 3006PSY || with 100% Errorless Solutions.
A measurement error that is constant across uses of a measure is said to be what? correct answers
A systematic Error
T/F: The standard error of measurement (SEM) is closely related to the concept of central
tendency? correct answers False
T/F: The standard error of measurement (SEM) is useful for converting individual raw scores to
percentile marks? correct answers False
If we knew the exact error of measurement in an individual's observed score, and we subtracted
this measure from their observed score, we would get their... correct answers True Score
Errors of measurement are produced by: correct answers - Changes in test items
- Changes in the examinees
- Changes in instrument administration
A statistic used to estimate the extent to which an observed score deviates from a true score
correct answers Standard error of measurement (SEM)
A measure of sampling error correct answers Standard Error of the mean
An index of the extent to which one individual's score vary over tests presumed to be parallel
correct answers Standard error of a score
A statistic used to estimate how large a difference between two scores should be before the
difference is considered statistically significant correct answers Standard error of the difference.
A group of employees took Skills Test "A" on Monday and again on Wednesday. Assuming no
practice or other instruction had occurred and given that the test had high reliability (r = .90),
then differences between the two test scores for each employee should be, "on average." correct
answers Quite Small
Reliability of measurement is concerned with: correct answers consistency
The longer the time interval between a prediction and the event to be predicted, the: correct
answers Smaller are the chances of making a good prediction
A grammar test is given twice within a few days to a group of word-processors. The highest
possible score is 40. One examinee received a score of 35 items correct the first time, and 18
correct the second time. If these results are similar for other students in the class, then the test is
most likely:
Lacking in relevance correct answers Unreliable
, This reliability coefficient is usually greater over the short-term (time) than the long term: correct
answers Test-retest
Which of the following would not a method of building reliability into a test?
a) Including many items in one test
b) Measuring the wrong skill
c) Two employers rating the same employees performance differently
d) Using formal assessment when an informal one might have been used. correct answers B
At minimum, all tests must have _____ validity and _____ _____ reliability. correct answers
Content
internal consistency
A human resource professional wants to compute the reliability of a test she has constructed after
a single administration. What type of reliability is applicable? correct answers Internal
consistency
If a thermometer measured the temperature in an oven as 180ºC five days in a row when the
temperature was actually 177ºC, this measuring instrument would be considered: correct answers
Reliable but not valid - because the thermometer is measuring the SAME temperate it is reliable,
but the actual temperate being measured is not correct, thus the measure is not valid.
A test of job performance is positively related to past supervisor ratings, job attendance, and
positive attitudes for similar employees. This information pertains most directly to: correct
answers Predictive Validity
A group of employees takes the Smith self-concept inventory and the Brown self-concept
inventory which are known to measure the same "thing." If the self-concept scores from the
Wilson and the Harry inventories are unrelated, which kind of validity appears to be lacking?
correct answers Construct validity
Face validity is most likely to contribute to better: correct answers Reliability
If we want to substitute a long measure with a shorter version of the same measure, and we
include both measures in a test to see how they compare, we are primarily concerned with:
correct answers Concurrent Validity
Which type of validity evidence is most likely to include most if not all of the other types of
validity? correct answers Construct validity
A test of extraversion negatively correlated with a measure of introversion would use what type
of validity? correct answers Construct
Four psychologists agree that a test measures the skills needed by a clinical master's student,
which type of validity would be appropriate? correct answers Content
A measurement error that is constant across uses of a measure is said to be what? correct answers
A systematic Error
T/F: The standard error of measurement (SEM) is closely related to the concept of central
tendency? correct answers False
T/F: The standard error of measurement (SEM) is useful for converting individual raw scores to
percentile marks? correct answers False
If we knew the exact error of measurement in an individual's observed score, and we subtracted
this measure from their observed score, we would get their... correct answers True Score
Errors of measurement are produced by: correct answers - Changes in test items
- Changes in the examinees
- Changes in instrument administration
A statistic used to estimate the extent to which an observed score deviates from a true score
correct answers Standard error of measurement (SEM)
A measure of sampling error correct answers Standard Error of the mean
An index of the extent to which one individual's score vary over tests presumed to be parallel
correct answers Standard error of a score
A statistic used to estimate how large a difference between two scores should be before the
difference is considered statistically significant correct answers Standard error of the difference.
A group of employees took Skills Test "A" on Monday and again on Wednesday. Assuming no
practice or other instruction had occurred and given that the test had high reliability (r = .90),
then differences between the two test scores for each employee should be, "on average." correct
answers Quite Small
Reliability of measurement is concerned with: correct answers consistency
The longer the time interval between a prediction and the event to be predicted, the: correct
answers Smaller are the chances of making a good prediction
A grammar test is given twice within a few days to a group of word-processors. The highest
possible score is 40. One examinee received a score of 35 items correct the first time, and 18
correct the second time. If these results are similar for other students in the class, then the test is
most likely:
Lacking in relevance correct answers Unreliable
, This reliability coefficient is usually greater over the short-term (time) than the long term: correct
answers Test-retest
Which of the following would not a method of building reliability into a test?
a) Including many items in one test
b) Measuring the wrong skill
c) Two employers rating the same employees performance differently
d) Using formal assessment when an informal one might have been used. correct answers B
At minimum, all tests must have _____ validity and _____ _____ reliability. correct answers
Content
internal consistency
A human resource professional wants to compute the reliability of a test she has constructed after
a single administration. What type of reliability is applicable? correct answers Internal
consistency
If a thermometer measured the temperature in an oven as 180ºC five days in a row when the
temperature was actually 177ºC, this measuring instrument would be considered: correct answers
Reliable but not valid - because the thermometer is measuring the SAME temperate it is reliable,
but the actual temperate being measured is not correct, thus the measure is not valid.
A test of job performance is positively related to past supervisor ratings, job attendance, and
positive attitudes for similar employees. This information pertains most directly to: correct
answers Predictive Validity
A group of employees takes the Smith self-concept inventory and the Brown self-concept
inventory which are known to measure the same "thing." If the self-concept scores from the
Wilson and the Harry inventories are unrelated, which kind of validity appears to be lacking?
correct answers Construct validity
Face validity is most likely to contribute to better: correct answers Reliability
If we want to substitute a long measure with a shorter version of the same measure, and we
include both measures in a test to see how they compare, we are primarily concerned with:
correct answers Concurrent Validity
Which type of validity evidence is most likely to include most if not all of the other types of
validity? correct answers Construct validity
A test of extraversion negatively correlated with a measure of introversion would use what type
of validity? correct answers Construct
Four psychologists agree that a test measures the skills needed by a clinical master's student,
which type of validity would be appropriate? correct answers Content
