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 a road vs. far, presence or absence of bird nest; total sample size 200 trees: ** Fisher's exact test of independence **

**2. ** Two nominal variables, beetle species, on feet vs. on back; total sample size 25×4=100 clicks; ** Fisher's exact test of independence **

**3. **Two nominal variables, blood type, snore vs. not snore; total sample size is 355+353=1308; **Fisher's exact test of independence**. Note that I put the questions in random order after I write them, so it is possible to have three questions in a row with the same answer.

**4. ** ** total number of amphipods: measurement
salinity of the water: measurement
type of plant: nominal
amount of human disturbance: ranked **

Whether to consider the name of each beach 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. ** ** salt concentration: measurement
oxygen content: measurement
weight: measurement
sex: nominal **

**8. ** One nominal variable, food type; theoretical expectation (1/3 in each trough) if the null is true; total sample size is 20 mice; ** exact test of goodness of fit **

**9. ** ** number of fireflies: measurement
sand particle size: measurement
percentage of the area that is bare sand: measurement
presence 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**

**15. ** Two nominal variables, salt substitute vs. salt, heart attack vs. no heart attack; total sample size 800+700=1500 people; ** chi-square test of independence ** or **G-test of independence**

**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 greater than 1000; ** chi-square test of independence ** or **G-test of independence**

**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 **

**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 **