Here are the answers to practice exam 1. The answer is in **bold**; my explanation, which you don't need to write on the exam, is in regular type.

**Where two correct answers are shown for these practice questions (such as chi-squared or G-test of independence), you must only write down one on the actual exam.**

**With a larger sample size, the standard deviation may be smaller or larger; standard error will be much smaller.**- Two nominal variables, age (nominal because just two categories, first-grade and about 80) and handedness; sample size greater than 1000:
**chi-squared or G-test of independence.** - Two nominal variables, sex of dog, sex of owner; sample size less than 1000:
**Fisher's exact test** - One nominal variable, birthplace; one ranked variable, dominance:
**Kruskal-Wallis test** - One measurement variable, heart rate; two nominal variables, before and after training, person (person is a nominal variable because you take two measurements from each person):
**Paired t-test or two-way anova without replication** **See if the data are normal and homoscedastic**- One nominal variable, mounting from right or left; theoretical expectation of 50:50 ratio; sample size less than 1000:
**exact test of goodness-of-fit** **The significant interaction term tells you that the effect of food on speed is different in males than in females**- One measurement variable, quadriceps strength; two nominal variables, sex and torn vs. untorn ACL:
**two-way anova with replication** - One measurement variable, HDL; two nominal variables, before vs. after, person (person is a nominal variable because you take two measurements from each person):
**paired t-test or two-way anova without replication** - Two nominal variables, Montana vs. Alberta, black vs. grizzly; sample size less than 1000:
**Fisher's exact test** - Two measurement variables, weight and alcohol intake; nothing in the question suggests a curved relationship:
**correlation/linear regression** - Two measurement variables, weight and alcohol intake; one nominal variable, extracurricular activity:
**ancova** - Two measurement variables, tryptophan (measurement because more than 5 values), time awake:
**correlation/linear regression** - Two measurement variables, prostate size, age; one nominal variable, steroid vs. no steroid:
**ancova** - One measurement variable, barkiness; one nominal variable, litter (individual is not a nominal variable, because only one measurment of barkiness per dog):
**Fisher's one-way anova or Welch's one-way anova** - One nominal variable, genotype; theoretical expectation of 1:3:3:9 ratio; total sample size greater than 1000:
**chi-squared or G-test of goodness-of-fit** - One nominal variable, presence or absence of both parasites; theoretical expectation of 1/6 present, 5/6 absent; total sample size greater than 1000:
**chi-squared or G-test of goodness-of-fit** - One measurement variable, melanopore index; three nominal variables: predator; tank (because more than one fish per tank); individual killifish (because more than one measurement per fish); each tank found with only one predator, each fish found with only one tank:
**nested anova** - One measurement variable, nitrogen fixation rate; one nominal variable, site (flask is not a nominal variable because only one mesurement per flask):
**Fisher's one-way anova or Welch's one-way anova** - One measurement variable, number of colonies; two nominal variables, bacterial species, detergent type; each species found with each detergent type; more than one measurement for each combination of species and detergent:
**two-way anova with replication** - One dependent measuremnt variable, lifespan; two independent measurement variables, exercise time and food consumption:
**multiple linear regression** - One measurement variable, light reflectance; two nominal variables, background, and individual crab (because two measurements per crab:
**paired t-test or two-way anova without replication** - One measurement variable, catabolic activity; two nominal variables, clonal strain, and line (line is a nominal variable because three measurements per line); each line found in only one strain:
**nested anova** - One measurement variable, tryptophan amount (measurement because more than 5 values); one nominal variable, asleep or awake; if there's a relationship, tryptophan affects the probability of falling asleep: :
**simple logistic regression** - Two measurment variables, age and running speed; question implies that relationship is curved (fastest at intermediate ages, slower at younger and older ages):
**curvilinear regression** - One nominal variable, vegetarian vs. carnivore; one ranked variable, order falling asleep:
**Kruskal-Wallis test** - Two measurement variables, fertilizer amount, number of tomatoes; question implies the relationship is curved (most tomatoes at intermediate fertilizer):
**curvilinear regression** - One dependent measurement variable, speed; two independent measurement variables, length and weight:
**multiple linear regression** - One nominal variable, back or feet; theoretical expectation of 1:1 ratio; sample size less than 1000:
**exact test of goodness-of-fit**