How to Use the Chi-Square in Qualitative Studies

When working with qualitative data, researchers may use a chi-square test to compare the results of observed date with data they expect to obtain according to a specific hypothesis. For example, when working with Mendel's laws of genetics, you would expect 50 percent of offspring to be male. If the actual observed number of males is other than that, however, you may want to determine the "goodness to fit" between the expected and observed result. The chi-square test examines the null hypothesis or the lack of correlation between expected and observed values.

Things You'll Need

  • Computer spreadsheet suite
  • Observed data values
  • Expected data values
Show More

Instructions

  1. Implementing a Chi-Square Test

    • 1

      Open up Microsoft Excel or another spreadsheet software.

    • 2

      Enter your observed results on one row, then type in your expected results in a second row.

    • 3

      Place the cursor where you wish to have the chi-square test results displayed and left-click the mouse.

    • 4

      Open the "Function Wizard" and select the "Statistical" function category.

    • 5

      Click on the function labeled "CHITEST" in the right section, then select "Next."

    • 6

      Highlight your observed data in the first row, which will appear in the "actual_range" box.

    • 7

      Click on the "expected_range" box and highlight your expected data in the second row.

    • 8

      Observe the results of the chi-square test, which will appear in the cell you originally selected.

    Analyzing a Chi-Square Test

    • 9

      Accept your null hypothesis if the chi-square test value, or p-value, is more than 0.05 percent since the deviation is small enough to be considered insignificant.

    • 10

      Conclude that some factor other than probability is operating for deviation and reject your null hypothesis if the p-value is equal to or greater than 0.05 percent.

    • 11

      Conduct your test again if the p-value is greater than 0.10 percent and determine whether experimental error has occurred.

Learnify Hub © www.0685.com All Rights Reserved