How to Calculate Effect Size for Change

Many statistical studies are interested in the changes caused by the administration of a treatment. The effect size of a study is a statistic that indicates the magnitude of the change that occurs from the administration of treatment. For example, researchers may be interested in how caffeine affects a person's running speed. The statistical analysis on the data of a study will give the information needed to perform a calculation of the effect size.

Instructions

    • 1

      Compute the means of both groups. The two groups in your study are called the control group (the group without treatment administration or before treatment administration) and the treatment group (the group with treatment administration or after treatment administration). Add the values for one group to get a sum. Divide this sum by the number of data points in that group. Do the same for the other groups. Call these values "mc" (for control) and "mt" (for treatment).

    • 2

      Compute the variances for each group. Subtract the mean for each group from every individual data point in its group. Square the resulting numbers and take their means by summing and dividing by the number of data points, just as you did for the original values. These means are the variances for the two groups. Call them "vc" (for control) and "vt" (for treatment).

    • 3

      Find the pooled variance of your sample. The pooled variance represents the variation in your data and is needed for the effect size calculation. Multiply "vc" by one less than the number of data points in the control group. Call this value "v1." Multiply "vt" by one less than the number of data points in the treatment group. Call this value "v2." Add v1 and v2. Divide the resulting number by two less than the total number of data points (the number of data points in the control group added to that of the treatment group). The result is a single number, the pooled variance of your data.

    • 4

      Convert the pooled variance into pooled standard deviation. Take the square root of the pooled variance. Call this new number "sp."

    • 5

      Compute the effect size. Subtract the control group's mean from the treatment group's mean and divide the result by "sp." This is the effect size for change in your study.

Learnify Hub © www.0685.com All Rights Reserved