Test Standardisation in Psychology – UGC NET – Solved PYQs

Topic Cover: Test standardization: Reliability, validity and Norms Codes

1. Read each of the following two statements, Assertion (A) and Reason (R) and indicate your answer using the codes given below: [DEC 2014]

Assertion (A): The squared correlation between the true score and the obtained score is called reliability index.

Reason (R): According to the Classical Rellability theory, the mean of the true scores and the mean of the obtained scores are equal.

Codes:

(1) Both (A) and (R) are true and (R) is correct explanation of (A)

(2) Both (A) and (R) are true, but (R) is not the correct explanation of (A)

(3) (A) is true, but (R) is false

(4) (A) is false, but (R) is true

2. A numerical ability test with 50 items has a Cronbach alpha of 80. If the test is randomly divided into two halves of the some length, the expected correlation between the two halves would be [DEC 2014]

(1) 0.40

(2) 0.56

(3) 0.67

(4) 0.80

3. Read each of the following two statements, Assertion (A) and Reason (R) and indicate your answer using the codes given below: [JUNE 2015]

Assertion (A): According to the True Score Model in the Classical Test Theory (CTT), for a population on a large sample, Mean of the obtained scores = Mean of the true scores.

Reason (R): In the CTT, the errors of measurement are assumed to be independently distributed with a mean and variance equal to zero.

Codes:

(1) Both (A) and (R) are true and (R) is the correct explanation of (A)

(2) Both (A) and (R) are true, but (R) is not the correct explanation of (A)

(3) (A) is true, but (R) is false

(4) (A) is false, but (R) is true

4. Read each of the following two statements, Assertion (A) and Reason (R) and indicate your answer using codes given below: [JULY 2016]

Assertion (A): According to Classical Test Theory (CTT), reliability can be conceptualised as the ratio of true score variance to the observed score variance.

Reason (R): According to the CTT, the random error of measurement is assumed to be uncorrelated with the true score.

Codes:

(1) Both (A) and (R) are true and (R) is the correct explanation of (A)

(2) Both (A) and (R) are true, but (R) is not the correct explanation of (A)

(3) (A) is true, but (R) is false

(4) (A) is false, but (R) is true

Directions (Q. Nos 5-9)

Read the following paragraph and answer the questions which are as follows:

A researcher working in the area of human intelligence wanted to develop a new verbal test of intelligence. For this purpose, he wrote 200 objective multiple-choice items. Each item had four alternatives, one of which was right and the three remaining ones were wrong. He administered this initial version to a sample of 400 subjects. Using P27 and P73 as the cut-off points, he obtained two groups Low Scoring and High Scoring respectively. The item difficulty index and the item discrimination index were computed for each of the 200 items. To assess whether each item significantly discriminates between Low scoring and High scoring groups, appropriate statistical significance tests were also employed. This item analysis process retained 150 items. The final version of 150 items was administered to a fresh sample of 750 subjects. The appropriate measures of internal consistency reliability were obtained. The test norms were developed using normalized T scores (mean = 100, SD = 20) and the Wechsler type deviation IQ scores. To assess the dimensionality of verbal intelligence, the researcher also carried out the factor analysis of item scores.

5. Which one of the following tests of statistical significance can be employed in the above study to the item significantly evaluate whether discriminates between the Low scoring and High scoring groups? [JAN 2017]

(1) Chi-square

(2) Point-biserial correlation

(3) T-test

(4) Biserial correlation

6. Which of the following reliability coefficients can be computed in the above study to assess the internal consistency? [JAN 2017]

I. Odd-even reliability coefficient
II. Kuder-Richardsen reliability coefficient
III. Cronbach alpha coefficient
IV. Interscorer reliability coefficient

Codes:

(1) Only I

(2) I and III

(3) I, II and III

(4) II, III and IV

7. For the above test, what is the expected percentile rank for the T-score of 120? [JAN 2017]

(1) 60

(2) 68

(3) 75

(4) 84