Homework 3

Course evals and exam grades

You work in the Office of Institutional Research at a large university, and your boss asks you to investigate the relationship between student evaluations and scores on the final exam in introductory statistics courses.

The assignment repository contains two CSV files to investigate this question. The exam_grades.csv file contains last semester’s final exam scores for 800 students taking introductory statistics across 30 different sections. The columns include:

• student_id: a unique 4-digit identifier for each student
• classid: which of the 30 sections the student was enrolled in
• professor_name: the name of the professor which whom the student took the class. Each professor teaches several sections
• exam_score: the student’s score on the final exam. All introductory students take the same final exam, regardless of course section

Your boss also provides you with information on each professor’s course evaluations. The professor_evals.csv file contains evaluations for each of the 10 professors teaching introductory statistics. The classid_1, classid_2, and classid_3 columns identify the 3 sections taught by each professor, while the three evalscore columns contain the professor’s corresponding course evaluation scores (out of 5) for each class.

Assigning course grades

For the second part of this assignment, you are a TA for a statistics course. At the end of the semester, the instructor asks you to calculate each student’s overall grade in the course.

You are provided with a CSV file (student_grades.csv), from Canvas, containing the grades for each student on each assignment in the course. Here are the instructions the professor gives you:

• There are 6 homeworks, 2 midterms, a final exam, and a project
• Each homework is scored out of 10. All other assignments are scored out of 100
• If a student did not submit an assignment, it is marked as NA in the CSV file. These missing assignments should receive a score of 0
• Homework is worth 15% of the course grade; midterm 1 is worth 15%; midterm 2 is worth 15%; the final exam is worth 25%; and the project is worth 30%
• The possible letter grades at the university are A, B, C, D, and F. There are no plus/minus options. Grades are assigned on a standard scale:
  • < 60 is F
  • 60 - 69 is D
  • 70 - 79 is C
  • 80 - 89 is B
  • 90+ is A

Your task is to calculate the overall course grade, reporting both the percentage and the letter grade.

Row-wise operations: Each row in the student_grades.csv file represents one student. To compute a grade for each student, you will want to combine scores within each row. To do this efficiently (without having to write out every homework column, e.g.), I recommend you look into row-wise dplyr operations. A good vignette is provided here.

Here is an example data frame adapted from the vignette:

df

## # A tibble: 2 × 4
##   name      x     y     z
##   <chr> <int> <int> <int>
## 1 Alice     1     3     5
## 2 Bob       2     4     6

Now suppose we want to compute the mean of the x, y, and z columns for each row. I can do this with a simple mutate call:

df |>
  mutate(row_mean = (x + y + z)/3)

## # A tibble: 2 × 5
##   name      x     y     z row_mean
##   <chr> <int> <int> <int>    <dbl>
## 1 Alice     1     3     5        3
## 2 Bob       2     4     6        4

Alternatively, I can compute the mean by first calling rowwise and then using the mean function:

df |>
  rowwise() |>
  mutate(row_mean = mean(c(x, y, z)))

## # A tibble: 2 × 5
## # Rowwise: 
##   name      x     y     z row_mean
##   <chr> <int> <int> <int>    <dbl>
## 1 Alice     1     3     5        3
## 2 Bob       2     4     6        4

However, it gets a bit tedious writing out all the column names explicitly! By leveraging rowwise and the c_across function, I can easily get the mean of, for example, all the numeric columns:

df |>
  rowwise() |>
  mutate(row_mean = mean(c_across(where(is.numeric))))

## # A tibble: 2 × 5
## # Rowwise: 
##   name      x     y     z row_mean
##   <chr> <int> <int> <int>    <dbl>
## 1 Alice     1     3     5        3
## 2 Bob       2     4     6        4

Here I haven’t had to list any of the columns explicitly, which is particularly nice when I have a lot of columns to work with. Note that if I want to avoid listing columns explicitly, I do need to use rowwise – the first solution, involving (x + y + z)/3, requires the explicit column names.

Student phone use

For the final part of this assignment, you are analyzing data from a study which investigates the amount of time students spend on their phone. Each student was asked to record their phone use (in hours) over the course of 5 days.

The phone_data.csv file contains the study data. year_in_school is the student’s year in college (1, 2, 3, or 4), and phone_hours contains the number of hours of phone use on each day. For example, 4, 6, 10, 7, 5 means 4 hours the first day, 6 the second day, etc.