Response: I would be interested in studying a population of pet dogs. The target population would be dogs from suburban households and who primarily stay inside. The two variables I would investigate are body score and minutes of exercise per week. Body score is an ordinal, categorical, subjective measure that a veterinarian assigns, and it is used in a similar manner as BMI in humans. A score of 4 or 5 is healthy weight, lower scores indicate underweight, and higher scores indicate obesity. Minutes of exercise per week is a continuous quantitative variable measuring the amount of time per week that the dog is engaged in intentional exercise (i.e. walk/run, fetch, training, etc.).

I expect the data for body score to primarily fall between the 4-8 range. The minutes of exercise data set would likely show a normal distribution with a majority of the data falling around the mean. The 5-number summary would be a good representation for both variables. A histogram or boxplot would be useful in visualizing both of these variables. I predict that minutes of exercise will be a predictor variable which influences weight. To assess this, I could use a side by side boxplot with body score on the x axis and weight on the y axis

From the response choose a third variable you might be interested in studying in their population and explain why. Discuss what statistics and visualizations you would expect would summarize its distribution. What features of its distribution would you want to examine?

Is this third variable likely related to one of the other variables already discussed? If so, what relationship would you hypothesize? Which is the predictor and which is the response variable? What visualization might illustrate this relationship well?