Upgrade to remove ads
Only $2.99/month

Psych Tests and Measurements Final

Terms in this set (41)

What tool of assessment was cited as particularly useful in learning about the reaction of simulated juries to certain types of evidence?
Behavioral Observation
A psychological interview requires
Direct, reciprocal communication
Of the following parties to the assessment enterprise, which group would be least likely to have read the Standards
Test takers
A psychological test almost always involves an analysis of
A sample of behavior
Which of the following contains the most detailed and authoritative information on responsible test use and test development
Standards for Psychological and Educational Tests
Which is the only type of scale that has an absolute zero point
Ratio
A 100-item achievement test is administered to 30 students. Students earn 1 point for each correct answer. In the test results there are three scores of 95. All of the other students score between 10 and 30. What measure of central tendency would be MOST representative of this set of scores
The median
What is the formula for the arithmetic mean as calculated from a frequency distribution
"Summation of fx" divided by N
Which statement is TRUE regarding this distribution of scores:
0, 11, 10, 0, 0, 5, 4, 3, 2, 1, 1, 1
There are two modes, 0 and 1
A student received a z score of 1 on a test of English as a Second Language and the distribution of test scores on that test was normal. This same student received a z score of 1 on another test of English as a Second Language and the distribution of test scores on this second test were highly skewed. Considering the percentile of this student
These two standard scores do not mean the same thing
If all scores in a set of test scores were the same, the variance would be equal to
Zero
In theory, the distribution of the normal curve ranges from
Negative infinity to positive infinity
A z score scale has sometimes been referred to as a "zero plus or minus one" scale because a z score scale
Has a mean set at 0 and a SD set at 1
A distribution of test scores on a Navy Seal Qualifying and Screening Examination administered to a class of high school seniors would be expected to yield
A positively skewed distribution of scores
If few scores fall on the negative side of the distribution, how is the distribution skewed
Negatively
It is incorrect to conceive of error "creeping into" the measurement process because
Error is a well acknowledged part of the measurement process
The term psychometric soundness is typically a reference to a test's reliability and
Validity
Norm-referenced testing and assessment is both a method of evaluation and a way of deriving meaning from tests scores. Individual test taker scores are evaluated against...
The scores of a comparison group of test takers
Psychological traits are EXPLICITLY present and exhibited to others...
Never
A basic assumption of all scaling is that
Traits and states can be quantified and measured
A reliability coefficient is
An index, a proportion of the total variance attributed to true variance, and unaffected by a systematic source of error
Which of the following would result in the LEAST appropriate estimate of reliability for a speed test?
Split-half from a single administration of the test
A confidence interval is a range or band of test scores that
Is likely to contain the true score
The standard error of measurement is
Used to infer how far an observed score is from the true score, also known as the standard error of a score, and is used in the context of classical test theory
Which type of reliability estimate would be appropriate only when evaluating the reliability of a test that measures a trait that is presumed to be relatively stable over time?
Test-retest
What is the difference between alternate forms and parallel forms of a test
Alternate forms do not necessarily yield test scores with equal means and variances
Which of the following is usually minimized when using split-half estimates of reliability as compared with test-retest or parallel/alternate form estimates of reliability
Time and expense
Test-retest estimates of reliability are referred to as measures of ---, and split-half reliability estimates are referred to as measures of ---
Stability; Internal consistency
"It's a measure of validity that is arrived at by a comprehensive analysis of how scores on the test relate to other test scores." This statement is a reference to
Construct validity
The form of criterion-related validity that reflects the degree to which a test score is correlated with a criterion measure obtained at the same time that the test score was obtained is known as
Concurrent validity
Which is NOT a method of evaluating the validity of a test
Evaluating the percentage of passing and failing grades on the test
Predictive and concurrent validity can be subsumed under
Criterion-related validity
Which is an example of convergent evidence for the construct validity of a test measuring fear of cats
A high correlation between the tests and an existing validated test measuring for fear of cats, and a high correlation with an existing validated test measuring more generalized fear
The form of criterion-related validity that reflects the degree to which a test score correlates with a criterion measure that was obtained some time subsequent to the test score is known as
Predictive validity
A test is considered to contain bias if
A factor inherent in the test systematically prevents accurate measurement
Which type of error has most likely occurred when a music critic's review of Lady Gaga's latest album is more positive than most people on the planet believe was warrented
Halo effect
A utility analysis of a test may best be thought of as
An evaluation of the costs and benefits of the test
An index of utility can be distinguished from an index of reliability and an index of validity in that an index of utility can tell us something about
...
Both Taylor-Russell and Naylor-Shine expectancy tables must include which type of validity coefficient
Concurrent criterion-related
A difference between the Naylor-Shine and Taylor-Russell expectancy tables is that
Naylor-Shine tables do not require that the criterion be dichotomized; the Taylor-Russell tables do
"An empirical standard used to divide a group of data into two or more distinct categories" is a description of a
Cut score
THIS SET IS OFTEN IN FOLDERS WITH...