Basic Statistics
Scores
CTT and IRT
Reliability
Validity
100
Ethnicity is an example of this level of measurement
What is nominal?
100
Jose just got his test score back and saw that he is in the 91st percentile, which indicates that the test had this type of referenced scores.
What is norm-referenced?
100

This is the difficulty level of this item.

What is .5?
100
This type of error is measured using test-retest reliability
What is time-sampling error?
100
A counselor administered a test to middle school students to determine which elective they would enjoy most in high school. This purpose suggests the need for this type of validity evidence.
What is predictive criterion-related validity?
200
This is the level of measurement that percentiles are.
What is ordinal?
200
This is the z score for someone who scored 600 on the GRE

What is 1?

(score-mean)/SD

(600-500)/100

200

This is the chance level of this item.

What is .1
200
You can split a test in half many, many different ways. Taking the average of the correlations between all those different split-halves gives you a measure of this type of reliability.
What is inter-item or internal consistency?
200
A table of specifications can be used as this type of validity evidence.

What is content related validity?

(shows relevance and representativeness)

300

This value is the IQR for the data displayed above

What is 13
300
Bob's z score was -0.30, so this was his T score.

What is 47?

Mean + (z-score x SD)

50 + (-.30 x 10)

300
45 of 50 high-ability children got the question right, and 20 of 50 low-ability children got the question right. This is the difficulty for the item.

What is .65

(45+20) / 100

300
Based on the Spearman-Brown formula, this is what happens when you lengthen a test.
What is increase the reliability?
300
The Naglieri Nonverbal Ability Test was used to identify traditionally underrepresented gifted children, and it accurately identified 15 children who were not identified with a traditional intelligence test. The evidence provided here is an example of this type of validity evidence.
What is concurrent criterion-related validity?
400
This statistic describes the average variability in the scores in the same units as the scores.
What is standard deviation?
400
DAILY DOUBLE! To earn an A in this course, the cut score is 90% of points, which is an example of this type of cut score.
What is a fixed cut score?
400
5 of 20 low-ability children and 13 of 20 high-ability children got the question right. This is the D value.

What is .40

13/20 - 5/20

400
This is another term for test-retest reliabililty.
What is stability reliability?
400
The math subscale of the Iowa Tests of Basic Skills was used as a pre- and posttest for a math intervention study, which did show improvement in the subscale scores. These experimental results are an example of this type of validity evidence.
What is convergent validity?
500
If you collect data from only psychiatric patients (a homogeneous group), this will happen to the correlation coefficient.
What is attenuate (or lessen)?
500
Jorge's IQ is 120, which makes this his percentile rank

What is 91st?

z = (120-100)/15 = 1.33

Large area on Urbina table

500
This is the classical test theory formula

What is X = T + E?

(observed score = true score + error)

500
Given that IQ test A has reliability of .96 and IQ test B has reliability of .90, this IQ test has the larger standard error of measurement.
What is IQ test B?
500
This value, which is the percent of correct decisions or classifications made with the use of test scores, is used for criterion-related validity evidence when the criterion is dichotomous.
What is a hit rate?