When trapping sticklebacks, should a lump of orange cheddar cheese be added in hopes of improving the catch? This is not recommended if you are hoping to examine gut contents because some sticklebacks actually eat the stuff. In other circumstances, however, is cheese recommended?
The debate has raged for years in the Schluter lab. McPhail swore by the cheesy approach, Schluter also recommends it, but his lab members are exceedingly skeptical. The question was put to the test for the first time on June 5, 2003, by Nathan Millar and others (identify yourselves, please). The test was carried out in Klein Lake, which has a rather low density of sticklebacks. Minnow traps were placed along the shore where they sat overnight. The results are given below. Each number refers to the count in a single trap (20 traps total).
|
Cheese
added |
No cheese
added |
|
|
14 |
0 |
|
|
12 |
3 |
|
|
0 |
4 |
|
|
3 |
3 |
|
|
12 |
1 |
|
|
23 |
1 |
|
|
0 |
3 |
|
|
22 |
4 |
|
|
0 |
1 |
|
|
|
0 |
|
|
|
13 |
|
|
9.6 |
3.0 |
MEAN |
**in fact, one of the traps under the no cheese columns had cheese in it, at least initially.
A two-sample t-test after a square-root (+0.5) transformation gave t = 1.7386, df = 18, P-value = 0.0992.
An approximate Wilcoxon test (Mann-Whitney U-test) gave Z = 0.9628, P-value = 0.3357
Thus, the results are inconclusive. Each camp can continue to hold on to private beliefs, with the assurance that no data (yet!) will prove them wrong.