1. Which of the following is NOT true about criterion-referenced testing?
a. They are used when one whishes to compare a test score to a predetermined value or a set criterion.
b. Many states have gone to using criterion-referenced testing as a result of the recent passing of No Child Left
Behind legislation.
c. They are useful when one is wishing to compare a score to an individual’s peer group.
d. As compared to norm-referenced testing, they are an alternative way of understanding test scores.
2. If you have the raw scores of two individuals on a norm-referenced test, which of the following is important in terms
of making meaning of those scores?
a. Measures of central tendency (mean, median, mode)
b. The relative positions of the scores to the rest of the group
c. Whether higher scores are better than lower scores
d. How an individual feels about his or her score
e. All of these are important.
3. On normally distributed curves, which of the following is NOT true?
a. The percentile rank mean is always p = 50.
b. The mean of z scores is always 0.
c. The mean of the standard deviation is 0.
d. The mean of standard scores vary as a function of the type of standard score (e.g., T score mean is 50, DIQ
mean is 100).
e. The mean of the publisher type score varies as a function of the publisher.
4. Which of the following is NOT always true?
a. If you know the z score, you can determine a percentile by looking at a conversion chart.
b. If you know the z score, you can determine most common standard scores (e.g., T score, DIQ, SAT, NCE,
etc.).
c. If you know the z score, you can determine the standard deviation of the test.
d. All of these are true.
Chapter7—StatisticalConcepts:CreatingNewScorestoInterpretTestData