What are item-total correlations? What is their purpose in an item analysis?

Question

What will be an ideal response?

Answer 1

• Item-total correlations are a measure of the strength and direction of the relationship between the way test takers responded to one item and the way they responded to all of the items as a whole.

• Item-total correlations are calculated using the Pearson product-moment correlation formula.

• Items that have little or no correlation with the total item score may measure a different construct from that being measured by the other items.

• Testing practitioners use item-total correlations as an alternative to the discrimination index discussed earlier.

• When items on a test are all measuring a similar construct, it is reasonable to assume that, on average, individuals who answer a particular question correctly will obtain a higher total score on the test than will individuals who answer the question incorrectly.

• Test developers want to retain those items that strongly differentiate high-scoring individuals from lower scoring individuals. This is indicated by a positive item-total correlation.

• The item-total correlation coefficient can be negative. This would mean that people who answered a question correctly actually did worse on the test than people who got the question wrong.

• Negative item-total correlations are often indicative of a question that has been poorly worded and is confusing the more capable test takers. It could also indicate a question where the wrong answer was keyed as being correct.

• Questions with negative item-total correlations should be either reworked or removed from the test, because they will reduce the test’s reliability/precision.