Question

ou are a patrol sergeant attending a staff meeting. At this staff meeting the results of an important, and very public study, are being presented. This study was conducted to determine the effect of body cameras on the frequency of police/citizen physical confrontations.

The researcher conducted an experiment wherein he fitted half of the officers in your department with body cameras. Assume that for all practical purposes the experiment was done correctly.

The researcher concludes that police officers wearing body cameras are 70 percent less likely to be involved in a physical confrontation with a citizen. While reading the researcher’s report you notice a footnote next to the statistic (which is a t-score) that the researcher is using as the basis of his conclusion. The footnote indicates that the statistical significance of the statistic is .421, well over the .05 level of acceptable error in social science statistical analysis!

The chief (who has absolutely no knowledge of statistical analysis) asks, “Sergeant, you look like you have something on your mind. Let’s hear it!” Briefly, using what you have learned about statistical significance explain to the chief (in simple terms) how this statistical finding should be interpreted, with respect only to its statistical significance.

Answer 1

Conclusion made by the researcher is incorrect because the p value is greater than .05 level of significance.

We know that when the p value is more than level of significance, then we fail to reject the null hypothesis. So, in this case, we can never conclude that the police officers wearing body cameras are 70 percent less likely to be involved in a physical confrontation with a citizen because the result is insignificant as p value is 0.421.

So, conclusion is incorrect and we should not say that the police officers wearing body cameras are 70 percent less likely to be involved in a physical confrontation with a citizen.