Comparing Variances. Is a t test required?

#1
O.k. guys. I'm new on here. I was given these 2 questions to do for a grad. school program entrance assignment, and I have no idea how to compute them. So, please help me figure this stuff out. Thank you!


Question 1:
A corporate manager granted two groups with either overtime or not and records the production: Overtime: 14, 19, 22, 28, 18, 22, 24, 26, 26, 27; Without overtime: 12, 16, 14, 18, 20, 24, 23, 22, 16, 20. Use Alpha 0.05. Compare the variances. Should we do a t test?

Question 2:
Two advertising firms examined the results of two ads by checking on whether the respondents actually purchased anything when they telephoned. Ad 1 had 240 people call and 169 bought something. Ad 2 had 275 callers and 147 bought something. Using an alpha of 0.05 and both of the approaches mentioned, show whether there is a difference between the ads.

Please help. I need to get this solved in the next couple days. Thank you!

Jennifer
 

JohnM

TS Contributor
#2
On #1, you'll need to first do an F-test to compare the variances between the overtime and no-overtime groups - if they are not statistically different, then it's OK to use a t-test to compare the average production levels (a t-test assumes that the variances of the two groups are not significantly different)

On #2, use a z-test to compare the two percentages
this example is somewhat similar:
http://www.talkstats.com/showthread.php?t=724