Practice 5

PUBH 523/623

Overview

This practice is mostly hands-on. You will start transforming the HRS dataset the way real analysts do: keeping and reordering the rows and columns you need, creating new variables, and combining conditions to answer more specific questions about respondents. Along the way you’ll get a little more practice with the pipe (|>) and with data frames in general — but the focus this week is on the new dplyr verbs.

Topics covered (Lessons 20-23):

• Intro to data transformations and summarizations
• Data transformation: rows
• Data transformation: columns
• Data transformations: advanced

Directions

Download this practice .qmd from the course GitHub repository, rename it, and save it in your course folder.

1. Download practice_05.qmd from GitHub.
2. Rename the file to Lastname_Firstinitial_Practice_05.qmd (or lastname_firstinitial_practice_05.qmd), replacing with your actual last name and first initial.
3. Save it inside your practice folder. You may create a practice_05 subfolder if you prefer.

TipHow to use this file

• The boxes with the yellow stripe explain what to do. Keep these in your submitted file.
• Do all your work below and outside of the yellow-striped boxes.
  • You will need to create code chunks within this document to run your R code.
• Please delete the following sections before submitting:
  • Overview
  • Directions
  • Tip boxes (green stripe)
  • Caution boxes / Needed Steps (orange/red stripe)
  • Submission checklist
  • Note on AI usage

ImportantNeeded setup

• You should already have hrs_data.rds from Practice 4 (link on the practice page). Place it in your data folder if you haven’t already.
  • You can check out the codebook here
• Load the tidyverse, here, and rio packages. tidyverse includes both dplyr (for this week’s functions) and forcats (for the factor functions in Question 4).

Questions

Question 1: dplyr Basics

In Lesson 20, we learned that dplyr functions are “verbs” that take a data frame in and return a data frame out, and that they can be grouped into four categories based on what they operate on: rows, columns, groups, and joins.

Part A: Input and output

WarningTask

In 1 sentence, describe the general input and output of most dplyr functions. What type of object typically goes in, and what type of object comes out?

Write your answer here.

Part B: Sort the verbs

WarningTask

For each function below, fill in the table with:

1. Whether it operates on rows, columns, or is part of the groups category (which we have not covered yet, but you saw it mentioned in Lesson 20)
2. What the function does, in your own words
3. What output you’d expect if you ran it on hrs_data (for example: the same number of rows but fewer columns, fewer rows but all the same columns, the rows reordered, etc.)

Function	Rows, Columns, or Groups (not yet covered)?	What does it do?	Expected output on hrs_data
filter()
select()
mutate()
arrange()
distinct()
relocate()
rename()
group_by()
summarize()

Question 2: Data transformation: rows

In Lesson 21, we learned three functions for transforming rows: filter(), arrange(), and distinct().

Part A: filter()

Using hrs_data:

WarningTask 1

Use filter() to keep only respondents whose cond_count (total number of chronic conditions) is 5 or more. Assign this to hrs_01_filter_conditions, and use nrow() to report how many respondents that is.

WarningTask 2

Use filter() to keep only respondents whose srh (self-rated health) is exactly "Poor". Assign this to hrs_02_filter_srh.

TipTip

Remember: if you misspell or mis-capitalize a value for a categorical/factor variable, filter() will not throw an error — it will just silently return 0 rows. Always check your output!

Part B: arrange()

WarningTask 1

Sort hrs_data by cesd in descending order first, and within ties, by age_yr in ascending order. Assign the result to hrs_03_arrange, and print it using tibble().

WarningTask 2

In 1 sentence, describe what would change about the output if you arranged by age_yr first and cesd second instead.

Write your answer here.

Part C: distinct()

WarningTask 1

Use distinct() on degree and race_original together to see how many unique combinations of these two variables exist in hrs_data. Assign this to hrs_04_distinct, and print it using tibble().

WarningTask 2

Use distinct() on degree again, but this time with .keep_all = TRUE, so you keep the full row (all columns) for the first respondent found at each unique degree value. Assign this to hrs_05_distinct_all.

Question 3: Data transformation: columns

In Lesson 22, we learned four functions for transforming columns: select(), relocate(), mutate(), and rename().

Part A: select() and relocate()

WarningTask 1

Use select() to keep only id, age_yr, ed, degree, income, cesd, and srh from hrs_data. Assign this to hrs_06_vars_subset.

WarningTask 2

Starting from hrs_06_vars_subset, use relocate() to move srh so that it comes right after id. Assign this to hrs_07_reloc, and print it using tibble().

Part B: mutate()

WarningTask

Using hrs_data, create a new variable called height_ft that converts height (recorded in meters) into feet, using the conversion 1 meter = 3.28084 feet. Round height_ft to 1 decimal place using round() inside your mutate() call. Assign the result to hrs_08_mutate, and use relocate() so that height_ft sits right next to height for easy comparison. Print it using tibble().

TipTip

Remember, inside mutate() you use a single =, not <- or ==, to define your new column.

Part C: rename()

WarningTask

Using hrs_data, rename cesd to depression_score and srh to self_rated_health. Assign the result to hrs_09_rename, and print names(hrs_09_rename) to confirm the change.

Question 4: Data transformation: advanced

In Lesson 23, we covered more advanced techniques: combining conditions with &, |, and %in%; checking our work; creating variables with case_when(); and working with factors.

Part A: Combining conditions

WarningTask 1

Before writing any code: do you expect more respondents to have diabetes AND high blood pressure (bp), or diabetes OR high blood pressure? Explain your reasoning in 1 sentence.

WarningTask 2

Use filter() with & to find respondents who have diabetes and have high blood pressure. Assign this to hrs_10_filter_diab_bp.

WarningTask 3

Use filter() with | to find respondents who have diabetes or have high blood pressure. Assign this to hrs_11_filter_diab_or_bp.

WarningTask 4

Use nrow() (or count()) on both objects to check whether your prediction from Task 1 was correct.

WarningTask 5

Now combine multiple logic statements in a single filter(): find respondents who have diabetes and have high blood pressure, or who have more than 5 chronic conditions. Assign this to hrs_12_filter_combo. Think carefully about where your parentheses go.

Part B: case_when()

WarningTask

Using hrs_data, create a new variable called income_cat with 4 categories based on income:

• "<25k" if income is less than $25,000
• "25k-50k" if income is at least $25,000 but less than $50,000
• "50k+" if income is $50,000 or more
• "unknown" if income is missing (use .default for this case)

Then use fct() to convert income_cat into a factor. Assign your final result to hrs_13_case_when, and check your work using count().

TipTip

For the middle category, you’ll need an & statement inside case_when(), similar to how we combined conditions in Part A.

Part C: Factors

WarningTask 1

Check the current class of hrs_data$srh using class().

WarningTask 2

Use fct_relevel() to reorder the levels of srh into descending order from “Excellent” to “Poor.” Call the new column srh_releveled, and assign the result to hrs_14_relevel. Confirm the new order with levels().

TipTip

If you misspell one of the levels inside fct_relevel(), R will not throw an error — it will silently drop that level to the bottom of the order (and print a warning in the console). Double-check your spelling against the original values in srh!

WarningTask 3

Now suppose you had instead written fct_relevel(srh, "Excellent", "Very good", "Good", "Fair", "Poor") — with a lowercase “g” in "Very good". Run this version and look at the warning message R gives you. In 1-2 sentences, explain what the warning is telling you and why it happens.

Part D: Selecting columns by pattern or type

WarningTask 1

Use select() with contains() to grab all columns in hrs_data whose name contains "smoke". Assign this to hrs_15_select_contains, and print it using tibble().

WarningTask 2

Use select() with where(is.factor) to grab all factor columns in hrs_data, and pipe the result into names() so you just see the column names.

Submission Checklist

If having a checklist helps you stay organized, check off each item below as you complete it. You do not need to submit this checklist.

• Input/output of dplyr functions described (Q1 Part A)
• dplyr functions sorted into rows/columns/groups, with what they do and expected output (Q1 Part B)
• hrs_01_filter_conditions and hrs_02_filter_srh created with filter() (Q2 Part A)
• hrs_03_arrange created with arrange() on two variables, with explanation (Q2 Part B)
• hrs_04_distinct and hrs_05_distinct_all created with distinct() (Q2 Part C)
• hrs_06_vars_subset created with select(), then srh relocated after id in hrs_07_reloc (Q3 Part A)
• hrs_08_mutate created with height_ft and relocated next to height (Q3 Part B)
• hrs_09_rename created with rename(), names confirmed (Q3 Part C)
• Prediction made, hrs_10_filter_diab_bp and hrs_11_filter_diab_or_bp created and compared, hrs_12_filter_combo written with combined &/| statement (Q4 Part A)
• hrs_13_case_when created with case_when() (4 categories, including .default), converted to a factor with fct(), checked with count() (Q4 Part B)
• hrs_14_relevel created with fct_relevel() on srh, confirmed with levels() (Q4 Part C)
• hrs_15_select_contains created with contains(), and columns selected with where(is.factor) (Q4 Part D)
• Document renders to HTML with no errors
• Files named correctly: Lastname_Firstinitial_Practice_05.qmd and Lastname_Firstinitial_Practice_05.html (or lowercase versions)
• Both .qmd and .html submitted to Brightspace

Note on AI usage

I used GenAI (Claude) to help me draft this practice assignment. It helped me brainstorm ideas for the assignment, and I directed it to help you complete the needed tasks that I showed in the lessons.