We actually have a small confession to make. We fudged the data from this study a bit! Recall from the video that the actual data showed that 10 of the 34 subjects in the yawn seed group (29.4%) and 4 of the 16 subjects in the control group (25.0%) yawned.

(r) Calculate the observed statistic for this study. How has it changed from the statistic you used earlier? Predict how this will change the p-value of the study and therefore the strength of evidence against the null hypothesis.


(s) Determine the p-value for the actual data:
  • In the Comparing Proportions applet below, change the two-way table counts to match the actual study results. (Remember to enter the counts not the totals and to press Use Table.)
  • Confirm your calculation of the observed statistic (treating "yawn" as a success).
  • Check the Show Shuffle Options box. (Scroll to the right in the window.)
  • Enter 1000 for the Number of Shuffles and press Shuffle.
  • Use the applet to compute the new p-value by entering the observed statistic in the Count Samples box and press Count.
  • Take a screen shot of your results (make sure your estimate of the p-value appears) and add to your report.

Was your conjecture correct?

Back Next