Contingency Table

Chi-Square, Cramer's V, and Lambda for a Rows by Columns Contingency Table

For a contingency table containing up to 5 rows and 5 columns, this page will:

	º	perform a chi-square analysis [the logic and computational details of chi-square tests are described in Chapter 8 of Concepts and Applications];
	º	calculate Cramer's V, which is a measure of the strength of association among the levels of the row and column variables [for a 2x2 table, Cramer's V is equal to the absolute value of the phi coefficient];
	º	and calculate the two asymmetrical versions of lambda, the Goodman- Kruskal index of predictive association, along with some other measures relevant to categorical prediction. [Click here for a brief explanation of lambda.]

To begin, select the number of rows and the number of columns by clicking the appropriate buttons below; then enter your data into the appropriate cells of the data-entry matrix. After all data have been entered, click the «Calculate» button.

Select the number of rows:
Select the number of columns:

Data Entry_Q

	B₁	B₂	B₃	B₄	B₅	Totals
A₁
A₂
A₃
A₄
A₅

Totals

Chi-Square	df	P	Message Box

Cramer's V =

Percentage Deviations_Q

B₁

B₂

B₃

B₄

B₅

A₁

A₂

A₃

A₄

A₅

Standardized Residuals_Q

B₁

B₂

B₃

B₄

B₅

A₁

A₂

A₃

A₄

A₅

Lambda for predicting	Standard Error	.95 CI Limits
Lambda for predicting	Standard Error	Lower	Upper
A from B:
B from A:

[Click here for a brief explanation of lambda.]

Estimated Probability of Correct Prediction

when Predicting:
A without knowledge of B
A from B
B without knowledge of A
B from A

Home

Click this link only if you did not arrive here via the VassarStats main page.