Question 1 and 2 of Part II is simply asking if there were any significant differences among estimate of total elk in 1992, estimate of total elk in 1994, and the management objective for that area (i.e., 2500 total elk). The logical approach is to:
Examine the overall test to determine if there was at least one difference, i.e., test the Ho: u92= u94= uMgmt
Test Ho: u92= u94 , which is the "test" you should have requested with the contrast vector (i.e., contrast = 1 -1 0), and
Note: if step #2 results in "no significant difference", then you simply want to test Ho: (u92+ u94) / 2 = uMgmt, i.e., compare the Mgmt Objective to the average of 1992 and 1994. This is the second default test that program Aerial Survey does for you (assuming you entered the Mgmt Objective on line 3 of the Test Survey window).
If you are conducting a 2-tailed test (i.e., Ho: u92 = u94) and using an alpha=0.10 (i.e., CI = 90%), then the critical region is p<0.05. In other words, the p value reported in your printout must be less than 0.05 for you to reject the null hypothesis. To avoid confusion, set alpha=0.10 (i.e., 90%CI) and treat each test as a 2-tailed test. It also might help you keep things straight if you write out the null and alternate hypothesis being tested at each step in the printout.
Question 3 of Part II is asking you to compare the estimated calf-cow ratio in 1994 to the Mgmt Objective for that area. You can use the same analysis approach as you did for total elk (see above) or you can make a simple 2-group comparision (i.e., only enter 1994's estimate and the Mgmt Objective in the Test Surveys window).
 |
 |
 |
Revised: 09 September 2005
