Skip to main content
JALT Testing & Evaluation SIG

Main navigation

  • Home
  • About TEVAL SIG
    • TEVAL's Mission
    • Officers
    • Constitution
  • Publications
    • Shiken Journal
    • Shiken Current Issue
    • Statistics Corner: Questions and Answers about Language Testing Statistics (book)
    • Shiken Back Issues
    • Submission Guidelines
    • TEVAL News
    • Shiken Issues Pre May 2012
  • News
    • TEVAL Talk Time
    • Grants Information
    • Introduction to jMetrik
    • Shiken Journal Call for Papers
    • Statistics Corner
    • Ethical Issues in Language Testing - Dr. Isbell
    • Dr. Isbell Pansig Plenary Slides
  • Events
  • Join
  • Contact Us
    • Facebook
    • email
User account menu
  • Log in

Breadcrumb

  1. Home

Statistics Corner: Chi-square and related statistics for 2 x 2 contingency tables

Article appearing in Shiken 17.1 (May 2013) pp. 33-40.

Author: James Dean Brown
University of Hawai'i at Manoa

Question:
I used to think that there was only one type of chi-square measure, but more recently, I have become confused by the variety of chi-square measures that exist. Can you explain the difference between a simple chi-square and a (1) likelihood ratio chi-square, (2) a continuity adjusted chi-square, and (3) a Mantel-Haenszel chi-square? Finally, when should each of these statistics be used and what is the differ- ence between a Yates and Pearson correction when used for chi-square data?

Download full article (PDF)

PanSIG Conference, Chiba, May 16-18, 2025

PanSIG
Kanda University of International Studies

SHIKEN

A Journal of Language Testing and Evaluation in Japan

Site editors

  • Reset your password
RSS feed

JALT is the Japan Association for Language Teaching, a nonprofit organization dedicated to the improvement of language teaching and learning. The TEVAL SIG is a Special Interest Group of researchers within JALT who are interested in testing and the evaluation of language learning outcomes.

Powered by Drupal