Skip to content
Omnikalc

Yule’s Q and Y association calculator

Enter the four counts of a 2 by 2 table to get Yule’s Q, Yule’s Y, the odds ratio, the phi coefficient, the standard error of Q and a 95 percent confidence interval.

Updated on
Your inputs

Value out of range

Value out of range

Value out of range

Value out of range

Results

Yule’s Q

0.800

Yule’s Y (coefficient of colligation)

0.500

Odds ratio

9.00

Phi coefficient
0.492
Standard error of Q
0.0849
Lower limit of the 95 % interval of Q
0.634
Upper limit of the 95 % interval of Q
0.966
Total count
100

The Yule’s Q and Y association calculator gives Yule’s Q, Yule’s Y, the odds ratio, the phi coefficient and a 95 percent confidence interval for Q from the four counts of a 2 by 2 table.

How the calculation works

  • Enter the counts a and b of the first row and c and d of the second row.
  • The calculator forms the odds ratio from the cross products ad and bc.
  • Read Q, Y, the odds ratio, phi and the interval.

Example

A table with 30, 10, 15 and 45 observations, 100 in all, has an odds ratio of 9.00. Yule’s Q is 0.800, with a standard error of 0.0849 and an interval from 0.634 to 0.966. Yule’s Y is 0.500 and the phi coefficient 0.492.

Why it matters

The calculator helps students, sociologists, epidemiologists and analysts who study the link between two yes-no variables, such as exposure and outcome or two survey answers, and want a bounded measure of association that does not depend on the margins of the table.

Good to know

Put the exposure or first variable in the rows and the outcome in the columns, and keep the same order for every comparison. Q is more extreme than a correlation: a Q of 0.8 corresponds to an odds ratio of 9 but a phi of about 0.5. A zero cell gives ±1 and an infinite odds ratio, so add 0.5 to all cells. The interval is an approximation that works well when every cell is at least about 5. Q and Y are symmetric, so swapping the rows or the columns changes only the sign, and they are most useful as descriptive measures next to a test such as the chi-square or the Fisher exact test.

The Cramér V calculator handles larger tables, the Fisher exact test calculator tests small 2 by 2 tables, the chi-square test calculator tests independence and the Matthews correlation coefficient calculator scores a binary classifier.

Frequently asked questions

What is Yule’s Q?

Yule’s Q is a measure of association for a 2 by 2 table, from −1 to +1. It is Q = (ad − bc) / (ad + bc) for the counts a, b, c and d, and it is a transformation of the odds ratio: Q = (OR − 1) / (OR + 1). A value of 0 means no association, and +1 or −1 a complete one in the sense that one cell is empty.

What is Yule’s Y?

Yule’s Y, the coefficient of colligation, is Y = (√ad − √bc) / (√ad + √bc). It is another transformation of the odds ratio, Y = (√OR − 1) / (√OR + 1), and takes more moderate values than Q. For the same table Y is always smaller in absolute value, which makes it closer to a correlation coefficient.

How does it differ from phi?

The phi coefficient is the Pearson correlation of two binary variables, and its maximum value depends on the marginal totals: it cannot reach 1 when the margins differ. Yule’s Q is a function of the odds ratio only and does not depend on the margins, which makes it comparable across tables with different marginal distributions.

How is the interval computed?

The standard error of Q is 0.5 × (1 − Q²) × √(1/a + 1/b + 1/c + 1/d). The 95 percent interval is Q ± 1.96 times that error, limited to the range from −1 to 1. Cells with small counts give a large error, and a zero cell makes Q reach ±1 and the formula invalid, so add 0.5 to every cell in that case.

Was this calculator helpful?

Thanks for your feedback!

Results are estimates provided for information purposes. They do not replace professional advice.