R has built in datasets which we can use to enhance our learning for example, here is the documentation for the cars dataset: https://rstudio-pubs-static.s3.amazonaws.com/481654_883a4b47c9b244d4859dd1db235f0165.html
Part 1. I’ve aleady written out the code for the built-in chickweight dataset you need to use to answer the next question. You may add code for including packages you may want to load in the below chunk.
# data("ChickWeight")
# View(ChickWeight)Q1. Using the variable ‘weight’, create a new variable reflecting different weight categories such as ‘low’, ‘medium’, ‘high’ (you can create your own version). Then refer to the documentation on t-tests and chi-squares and answer this question: Is there a statistical significant association between the diet and weight gain? Which test do you plan to use (and why)?
# You may need the below code if you're trying to create more than two categories
# check if _R_USE_PIPEBIND_ environment variable is set
if (!"TRUE" %in% Sys.getenv("_R_USE_PIPEBIND_", unset = NA)) {
# set _R_USE_PIPEBIND_ environment variable
Sys.setenv("_R_USE_PIPEBIND_" = "TRUE")
}Q2. Plot chickweight against diet (not the new variable you created), however you wish to. Which diet type seems to be correlated with higher weight gain?
Part 2.
Read in nc births from last week again.
Q 1. One of the questions in the previous assignment was as follows: “4. Do you think mothers who smoked are more or less likely to have babies with low birth weight?” Now use a statistical test to determine if there is a significant association between weight and smoking status. Write a very brief interpretation for your findings - does it align with what you expected? You’re a maternal and child health leader - what message would you spread to expecitng mothers or individuals interested in conceiving (are you able to make a recommendation?)?
Q 2. Is there an association between maternal age and term status? Hint: For a t-test outcome needs to be a quantitative variable. Select your varaibles accordinly (also test-dependent)
Q 3. It’s always neat to be able to observe the distribution of continous variables to visually get a sense of the mean and range. Try hist(var_of_your_choice) - do this for at least 2 variables.
Q 4. Bonus question, generate any plot of your choice but it cannot be the same as the one you generated earlier in this exercise.