Tailed Tests


Your audience is your client, the investment group. But the group has hired one consultant and she is reasonably well-versed in statistics, so you are speaking to her for the most part. As such, you need to be clear about what tests you are running (what hypotheses), what level of confidence, etc. but you can assume that she can follow the statistics. However, key takeaways should still be noted so that the report is accessible to others who are not statisticians.
Your client is interested in knowing if there is a significant difference in AVERAGE WEEKLY SALES ($) for:
⦁ Restaurants that are SWAM vs. are NOT SWAM (women or minority owned businesses).
⦁ Restaurants that are franchises vs. are not franchises.
⦁ Restaurants that have a website vs. do not have a website.
⦁ Restaurants that have road frontage vs. do not have road frontage.
⦁ Restaurants that are in localities with a special meals tax vs. those without a meal tax.
⦁ Restaurants serving American food vs. any non-American food types (Asian, Italian, Mexican, Other combined as one type).
You must:
⦁ Perform the desired hypothesis test (assuming equal variance and not “pairing” the observations) in Excel.
⦁ Create tables of Excel results in the paper. (You can decide how to organize the tables.)
⦁ Write out your hypothesis for each test…they must appear in the paper.
⦁ Keep in mind you need to think about one-tailed vs. two-tailed and be clear about your assumption.
⦁ Your hypotheses must reflect these assumptions.
⦁ You need to discuss what level of confidence you are working with.
⦁ Discuss results and what can be learned from them.
⦁ Be precise in your language.
⦁ Use TIMES ROMAN 12 point font.
⦁ Use 1″ margins
⦁ No more than 2.5 pages
A one-tailed and two-tailed statistical hypothesis testing were done on the data using the average weekly sales as defined above. The results are below for each category. Confidence level is 95% for ALL tests (alpha=0.05) and assuming equal variances.
DF= Degree of Freedom
P = p-value
If the absolute value of the test statistic (T Stat) is greater than the critical value (T Critical), then we reject the null hypothesis (that two possibilities are the same)
Women or minority-owned business designation (SWAM)

Is franchise or part of national brand

Has Website or not

If restaurant has road frontage (visibility)

Area has meals tax added on top of state sales tax

Serving American or other food