Rubric

  1. Using your dataset’s description and/or codebook, briefly describe how the data were collected (the sampling strategy).
    • (1 p) Indicate the source of the information, document name or url and page number and
    • (1 p) summarize the data collection strategy.
    • (2 p) What potential issues (estimation bias, etc) may be a result of this collection strategy, or how does the strategy used protect against potential sampling issues?

(For example, if there is an interest to learn about different types of people and a certain type has a very low population frequency, then a good study design would use stratified sampling to oversample people from that group to assure that estimates have adequate sample sizes for that group.)

  1. Using either a numerical variable or a two-level categorical variable, calculate and interpret a confidence interval for the population mean or proportion.
    • (1 p) Identify and describe the variable,
    • (1 p) use t.test() or binom.test() to calculate the mean and confidence interval, and
    • (2 p) interpret the confidence interval.
    • (2 p) Using plotting code from the last two classes, plot the data, estimate, and confidence interval in a single well-labelled plot.

Turn in your master HW file with these sections appended to the bottom.

(After this HW is graded, move the section on the data collection strategy up to the top where your dataset is described.)