Question

The effect of a monetary incentive on performance on a cognitive task was investigated. The researcher predicted that greater monetary incentives would result in higher performance. Participants were told that they would receive either 5 cents, 25 cents, or 50 cents for each word puzzle that they correctly solved. A statistical test showed that there was a significant effect of incentive on performance at the .01 level. The greater the incentive, the more puzzles were solved. 1.) What statistical test was performed? 2.) Explain why rejecting the null hypothesis does not mean that the results of a study prove that the null hypothesis is false.

Answer 1

Here there are two variables one is the incentive in cents and performance in terms of the number of puzzles solved. A correlation test could have be performed to check whether there exist any kind of relation between the two variables.. Since the statistical test showed a significance effect at 1% level of significance we conclude that there exist a positive correlation between the variables. The hypothesis to be tested is vs

The study will be based on a sample collected at a particular time. The sample belongs to a large population of performers and the conclusion would be purely based on the sample collected and not based on the population. Eventhough we believe that sample precisely represent the populaation units, mistakes are inevitable. So rejection or acceptance of a hypothesis will be purely based on the sample available to us. We always keep a 5% chance that the hypothesis may be accepted. Here in this case the chance is 1%. so the decision to reject or accept do not imply that null hypothesis is false or true respectlvely.