Sample Size Calculation

For comparing means of two independent groups with repeated measure:

  • • Data types: Outcome variables are numeric (continuous)
  • • Experiment: Comparison of 3D contents with 2D for memory retention and recall process
  • • Design: Participants will be randomly assigned into 3D group and 2D group
  • • Analysis: Two-sample independent f-test at alpha 0.05 (one sided)

Parameters Specified:

  • • Power (1-Beta): 80%
  • • Difference between two means: 0.5 (used Cohen’s d for medium sample)
  • • Standard deviation (assumed to be equal for both groups): 1
  • • Number of repetitions: 4
  • • Sample calculated: Group sample sizes of 26 and 26 achieve 80% power to detect a difference of 0.06 in a design with 4 repeated measurements having a compound symmetry covariance structure when the standard deviation was 0.11, and the alpha level is 0.050. Further, keeping 30% margin due to repetition, the total sample size was N = 68 (34 for each group)
PASS user interface window

Figure 7.1 PASS user interface window.

• Tool used: We used PASS software (see the main window of PASS in Fig. 7.1) for this calculation. The report is generated by the software with the previously specified parameters.

< Prev   CONTENTS   Source   Next >