A nationally known supermarket decided to promote its own brand of soft drinks on TV for two weeks. Before the ad campaign, the company randomly selected 21 of its stores across the United States to be part of a study to measure the campaign’s effectiveness. During a specified half-hour period on a certain Monday morning, all the stores in the sample counted the number of cans of its own brand of soft drink sold. After the campaign, a similar count was made. The average difference was an increase of 75 cans, with a standard deviation of difference of 30 cans. Using this information, construct a 90% confidence interval to estimate the population average difference in soft drink sales for this company’s brand before and after the ad campaign. Assume the differences in soft drink sales for the company’s brand are normally distributed in the population.