Confidence Intervals

---

Overview

The data analysis conducted by the analysts at the Cocoabix biscuit factory group in islandX has revealed the 95% Confidence intervals for the mean number of chocolate chips per biscuit at two timeframes. The two timeframes are “2017 - before no eating policy” and “2018 - after no eating policy”.

The confidence intervals are important in providing an indication of the true mean number of chocolate chips per biscuit per timeframe at the Cocoabix biscuit factory group in islandX.

Confidence intervals

We can say that the 95% Confidence interval contains plausible values for the mean value of chocolate chips per biscuit at a given timeframe. We are 95% confident that the true mean falls within the interval. In the example in the plot below (for another brand of biscuits), note that five of the intervals do not contain the true mean (10 chocolate chips).

The management of the biscuit factory group in islandX are pleased that the two 95% confidence intervals provide a strong indication that there are more chocolate chips per biscuit post policy intervention ….however, they intend and need to perfom further analyses to formally compare the difference in means (before and after the policy). The null hypothesis in this case would be: “the difference in the mean number of chocolate chips per biscuit before and after the no-eating policy at Cocoabix factories in islandX is zero”. Additionally, the team is interested in examining whether the proportion of broken biscuits has changed post policy and to construct relevant confidence intervals.

A press release will take place on Christmas eve.