Answer:
We conclude that there is no difference in potential mean sales per market in Region 1 and 2.
Step-by-step explanation:
We are given that a random sample of 12 supermarkets from Region 1 had mean sales of 84 with a standard deviation of 6.6. 
A random sample of 17 supermarkets from Region 2 had a mean sales of 78.3 with a standard deviation of 8.5. 
Let  = mean sales per market in Region 1.
 = mean sales per market in Region 1.
 = mean sales per market in Region 2.
  = mean sales per market in Region 2.
So, Null Hypothesis,  :
 :  = 0      {means that there is no difference in potential mean sales per market in Region 1 and 2}
 = 0      {means that there is no difference in potential mean sales per market in Region 1 and 2}
Alternate Hypothesis,  : >
 : >  0      {means that there is a difference in potential mean sales per market in Region 1 and 2}
 0      {means that there is a difference in potential mean sales per market in Region 1 and 2}
The test statistics that will be used here is <u>Two-sample t-test statistics</u> because we don't know about population standard deviations;
                             T.S.  =   ~
   ~  
where,  = sample mean sales in Region 1 = 84
 = sample mean sales in Region 1 = 84
 = sample mean sales in Region 2 = 78.3
 = sample mean sales in Region 2 = 78.3
 = sample standard deviation of sales in Region 1 = 6.6
  = sample standard deviation of sales in Region 1 = 6.6
 = sample standard deviation of sales in Region 2 = 8.5
  = sample standard deviation of sales in Region 2 = 8.5
 = sample of supermarkets from Region 1 = 12
 = sample of supermarkets from Region 1 = 12
 = sample of supermarkets from Region 2 = 17
 = sample of supermarkets from Region 2 = 17
Also,  =
  =  = 7.782
 = 7.782
So, <u><em>the test statistics</em></u> =   ~
  ~  

                                    =  1.943  
The value of t-test statistics is 1.943.
  
Now, at a 0.02 level of significance, the t table  gives a critical value of -2.472 and 2.473 at 27 degrees of freedom for the two-tailed test.
Since the value of our test statistics lies within the range of critical values of t, so we have<u><em> insufficient evidence to reject our null hypothesis</em></u> as it will not fall in the rejection region.
Therefore, we conclude that there is no difference in potential mean sales per market in Region 1 and 2.