Determine if an experiment verifies the hypothesized results.
- actual_range - required, the range of data to test against expected values
- expected_range - required, the range of data to test against the ratio of row totals to column totals to the total
This example uses the CHISQ.TEST function to test whether the data in A2:B4 and A6:B8 are independent. The results of the test will show whether the data from the two ranges are related or not.
This example will return a #N/A error if the number of data points in A2:B4 and A6:B8 are different. This means that the two sets of data must have the same number of points in order to test if they are independent or not.
The CHISQ.TEST function is used to test for independence and calculate the chi-squared statistic. The result of the test is used to determine if a hypothesis is verified by an experiment. An error will be returned if the number of data points in the actual_range and expected_range are different.
- The CHISQ.TEST function calculates a chi-squared statistic based on the formula Aij = actual frequency in the i-th row, j-th column Eij = expected frequency in the i-th row, j-th column r = number of rows c = number of columns.
- The CHISQ.TEST function also calculates the probability that a value of the Ï‡2 statistic at least as high as the value calculated by the above formula could have happened by chance under the assumption of independence.
- The CHISQ.TEST function is appropriate when Eijâ€™s are not too small.
Frequently Asked Questions
What is CHISQ.TEST?
CHISQ.TEST is a statistical test that tests whether data fits a Ï‡2 (chi-squared) model.
What does CHISQ.TEST do?
CHISQ.TEST calculates a Ï‡2 statistic to determine if the data fits a Ï‡2 model.
When should CHISQ.TEST be used?
CHISQ.TEST is most appropriate when the expected range and the actual range of data points have similar numbers, and when the Eij's (expected frequencies) are not too small.
What does CHISQ.TEST return?
CHISQ.TEST returns #N/A if the actual and expected ranges have different numbers of data points.