A researcher wants to see if people really vote the way they claim. Using exit polls, the researcher found that in a sample of 100 voters, 65 claimed they voted for a candidate. The voting data showed that only 53% of people actually voted for the candidate. In her hypothesis test conclusion the researcher found a p-value that led her to reject the null hypothesis. But, her sample was flawed, so the hypothesis test was incorrect, and the null hypothesis was actually true. This is an example of: