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.**

- One measurement variable, bone density; two nominal variables, fractured vs. unfractured leg, person (because two measurements per person):
**paired t-test or two-way anova without replication** **If the average weights with or without aspirin are really the same, you'd get a difference of 103 grams or more 7 percent of the time by chance.**- One measurement variable, fructose amount; two nominal variables, apple variety and temperature (temperature is nominal because just two values); each temperature found in combination with each apple variety; multiple measurements for each combination of variety and temperature:
**two-way anova with replication** - One dependent measurement variable, weight of lichen litter; four independent measurement variables, age, diameter, height, distance:
**multiple linear regression** - One measurement variable, lesion area; two nominal variables, genotype and mouse (mouse is a nominal variable because multiple measurements per mouse); each mouse in combination with only one genotype:
**nested anova** - One measurement variable, distance from fluorescent corn field; one nominal variable, fluorescent or non-fluorescent pollen grain; if there's a cause-and-effect relationship, it's that distance affects percentage of pollen that are fluorescent:
**simple logistic regression** - One measurement variable, number of
*Klebsiella*; two nominal variables, before and after, person (person is nominal because two measurements per person):**paired t-test or two-way anova without replication** - Two measurment variables, droopiness of dogs, droopiness of owners:
**correlation/linear regression** - One nominal variable, light type; one measurement variable, pumpkin diameter:
**one-way anova** - One measurement variable, fish length; two nominal variables, temperature (nominal because only two temperatures), tank (because more than one measurement per tank); each tank found with only one temperature:
**nested anova** - Two nominal variables, corn vs. soybean, hole vs. no hole; sample size less than 1000:
**Fisher's exact test** - Two nominal variables, particle size and number of crickets; question says the relationship looks curved:
**polynomial regression** - One nominal variable, dead or alive; four measurement variables, age, height, weight, time; if there's a cause-and-effect relationship, the meaasurements affect survival:
**multiple logistic regression** **There may be many studies that were unpublished because they didn't show an effect.**- Two measurement variables, stride length, leg length; one nominal variable, exercise type:
**ancova** - One dependent measurement variable, number of offspring; three independent measurement variables, weight, length, age:
**multiple linear regression** - One measurement variable, number of eggs; two nominal variables, food type, chicken breed; each chicken breed found in combination with each food type; only one measurement per combination:
**two-way anova without replication** - One nominal variable, maggot species; one measurement variable, tetracycline dose (measurement because 11 different amounts); if there's a cause-and-effect relationship, tetracycline dose affects percentage of
*Phaenicia regina*:**simple logistic regression** - Two measurement variables, amount of salt, amount of fat; nothing in question suggests a non-linear relationship:
**correlation/linear regression** - One measurement variable, stain amount; two nominal variables, mutated vs. not mutated, individual fly (fly is a nominal variable because four measurements per fly); each fly found in only one mutation type:
**nested anova** **Look at histograms to see if the measurement variable is normal and homoscedastic.**- One measurement variable, days until distress; two nominal variables, sex of chicken, strain of virus; each sex in combination with each virus strain; multiple measurements for each combination of sex and strain:
**two-way anova with replication** - One measurement variable, level of T-cells; two nominal variables, donor vs. recipient and pair of people;
**paired t-test or two-way anova without replication** - One measurement variable, migration distance; two nominal variables, normal vs. prostate cancer cells, treated with EGF vs. untreated; each treatment in combination with each cell type; multiple measurements per combination:
**two-way anova with replication** **You should not very excited, because the chance of one P=0.021 out of 75 tests is high, even if all null hypotheses are true.**- One measurement variable, reflected light; two nominal variables, black vs. white background, individual crab:
**paired t-test or two-way anova without replication**. Note that the question says the crabs are "individually tagged." This means that you know the name of each crab and can keep the two measurements for each crab together, making crab identity a nominal variable. - One nominal variable, ant treatment; one measurement variable, number of ants:
**one-way anova** - One measurement variable, rotation speed; two nominal variables, before vs. after training, individual (individual is a nominal variable because two measurements per individual:
**paired t-test or two-way anova without replication** - Two measurement variables, weight of pig, amount of selenium (selenium amount is measurement because 8 different values:
**correlation/linear regression** - One measurement variable, number of goldenrod plants; one nominal variable, presence or absence of praying mantis; if there's a cause-and-effect relationship, number of goldenrods affects presence of praying mantis:
**simple logistic regression**