Question Type 10: Determining the type of reliability Exercises

Ten students take a geometry test and then retake the same test two weeks later. Their scores correlate at 0.68. Identify the reliability type and assess whether the test is acceptably reliable.

A teacher administers the same mathematics test to a class on Monday and then again on Friday. The Pearson correlation between the two sets of scores is 0.85. Identify the type of reliability being assessed and comment on the adequacy of the reliability coefficient.

Parallel‐forms reliability is assessed by administering Form X and Form Y of a chemistry test to the same students; the correlation is 0.60. Describe this reliability type and evaluate the test’s consistency.

A new depression inventory is administered twice to the same sample one week apart, yielding a reliability coefficient of 0.95. Name the reliability type and explain what this high coefficient implies.

Two versions (Form A and Form B) of an English vocabulary test are administered to the same group, yielding a correlation of 0.72 between scores. Determine the type of reliability and interpret the coefficient.

A performance art exhibition is rated independently by two judges. The intraclass correlation coefficient (ICC) for their ratings is found to be 0.90. Identify the reliability type and discuss its adequacy.

In a 20‐item psychology test, odd–even split‐half reliability yields a correlation of 0.78 between the two halves. Calculate the estimated reliability of the full test using the Spearman–Brown formula, and name the type of reliability measured.

A researcher calculates an average inter‐item correlation of 0.3 on an 8‐item measure. Compute Cronbach’s alpha and specify the form of reliability.

A 5‐item scale has item variances of 2, 1.5, 2.5, 1, and 1, and total test variance of 15. Calculate Cronbach’s alpha and identify the type of reliability.

A 10‐item true/false test has $\sum p_i q_i = 2.1$ and total test score variance $\sigma_t^2 = 4$. Calculate the KR-20 coefficient and state the type of reliability.

Using the Spearman–Brown prophecy formula, how many items are needed to achieve a reliability of 0.80 if the current 12‐item test has split‐half reliability $r_{hh} = 0.50$?

Given the following correlation matrix for a 3‐item scale: $r_{12}=0.4$, $r_{13}=0.5$, $r_{23}=0.6$, compute Cronbach’s alpha and identify the type of reliability.