Biological Data Analysis: Exam 1 Answers

Here are the answers to the exam questions. The answer is in bold and the explanation for the answer is in regular type. If you have quick questions about your exam, you can talk with me before or after class. If you'd like to talk with me about the exam outside of class, e-mail me to set up a time to meet.

1. Two nominal variables, near cedar trees vs. far, presence or absence of rust spots; total sample size 100×2=200 leaves: Fisher's exact test of independence

2. Two nominal variables, slug species, 15% vs. 30% water; total sample size 50×4=200 slugs; Fisher's exact test of independence

3. Two nominal variables, blood type, snore vs. not snore; total sample size is 655+653=1308; chi-square test of independence or a G-test of independence. Note that if you said "chi-square or G-test of independence," you got points off; you have to pick one or the other, for this and all other questions this semester. You also got points off if you just said "chi-square test" without specifying whether it was a test of goodness-of-fit or independence.

4. total number of salamanders: measurement
pH of the soil: measurement
number of dead logs: measurement
white oak vs. not white oak leaves: nominal
shadiness: ranked

Whether to consider the quadrat a nominal variable is a gray area, we'll talk about it later in the semester. You didn't get points off for including it or not including it.

5. Two nominal variables, sweet corn vs. yellow dent, E-strain vs. Z-strain; total sample size is 800+940=1740 larvae; chi-square test of goodness-of-fit or a G-test of goodness-of-fit

6. "For this experiment, there is a 2% probability of getting 9 or fewer dogs rolling in cow feces, or 9 or fewer dogs rolling in dead raccoon, if the null hypothesis is true." Note that I didn't take points off if you described a one-tailed test, although I could have; I guess I was in a good mood when I graded the exams.

7. mannose concentration: measurement
oxygen content: measurement
weight: measurement
sex: nominal

8. One nominal variable, grass vs. hay vs. peelings; theoretical expectation (1/3 in each trough) if the null is true; total sample size is 20 goats; exact test of goodness of fit

9. number of fireflies: measurement
sand particle size: measurement
percentage of the area that is bare sand: measurement
resence or absence of a streetlight: nominal

Whether to consider the quadrat a nominal variable is a gray area, we'll talk about it later in the semester. You didn't get points off for including it or not including it.

10. One nominal variable, lights vs. no lights; theoretical expectation (5/11 of turtles in lighted sections); total sample size 55+22=77 nests: exact test of goodness of fit

11. order of unrolling: ranked
weight: measurement
sex: nominal

12. Exact tests give more accurate estimates of the P-value, so we use them when the sample size is small. THe computations for exact tests are difficult for computers when the samples sizes are large, so we use chi-square or G-tests then.

13. The statistical null is that 42.3% of bird poops will be on cars. One nominal variable, on a car vs. on asphalt; theoretical expectation; total sample size 61+57=118 poops; exact test of goodness-of-fit

14. behavior order: ranked
amount of sugar eaten: measurement
age in months: measurement sex: nominal

Some of you put that the kinds of food eaten was a nominal variable; I didn't intend that, but the question is a little vague, so you got credit with or without that variable

15. Two nominal variables, salt substitute vs. salt, heart attack vs. no heart attack; total sample size 300+300=600 people; Fisher's exact test

16. Because the two diets are the same, the null hypothesis is true, so the probability of a P-value of 5% or less is 5%

17. Two nominal variables, kind of cat, adopted in first week vs. not adopted; total sample size 54+41+38+21+19+4+2=179; Fisher's exact test

18. Three nominal variables, which bay it it; mouth vs. inside of bay; Gpi90 vs. Gpi100 allele; Cochran-Mantel-Haenszel test

19. alpha, 0.05
power, 0.90 OR beta, 0.10 but not both power and beta, because they just tell you the same thing.
effect size, 15% fewer snorers

20. increasing alpha makes needed sample size go down
decreasing power OR increasing beta makes the needed sample size go down
increasing effect size makes the needed sample size go down


Return to the Biological Data Analysis syllabus