Assessing quality of exam MCQs Item discrimination II in Excel Point-Biserial correlation

Assessing quality of exam questions (MCQs)
Item discrimination analysis in
Excel
Point-Biserial correlation coefficient (rpb)
Item discrimination
Item refers to a question.
Assessing exam quality.
Assessing each question in exam sheet as either good or bad question.
Differentiates between students who have mastered the subject contents from those
who have not.
Determines how well an item is discriminating between good and poor students.
Point-Biserial (P.B.) assists educators identifying good or bad test question.
Point Biserial correlation coefficient
One of the item discrimination index.
Measures the discriminating power of an item (question).
A method of measuring the relationship between student’s total exam scores
(continuous variable, from 0-100 %) and a student’s item (Q) mark (dichotomous
Variable; two categories, correct or wrong answer).
Parametric test based on Pearson’s product moment correlation.
Measures the strength of relationship between two set of data (2 variables).
One variable is continuous (total mark in percentage).
Another variables is categorical, 2 categories (correct or wrong answer).
Measures the direction of the relationship.
Calculates the significance of the relationship by producing P-Value.
Values for point-biserial range from -1.0 to +1.0.
Positive value indicates that learners who answered a given question correctly
also have a high score on the overall exam and those who answered the same
question wrongly, have low total mark in the test.
A high positive value indicates the question is discriminating high-performing
students from low-performing students.
Negative value means, high-performing students are getting the answer wrong,
and the low performing students are getting it right. (Remove item from exam)
Zero value means, item is not discriminating between students.