Lesson 2b: Inline Code for Reproducible Writing

Learning goals

By the end of this lesson, you will be able to:

  • Explain what inline R code is
  • Use inline code to report values in sentences
  • Format numbers for professional writing
  • Use inline code with summaries and model results
  • Avoid common reproducibility mistakes

What is inline code?

Inline code is R code that runs inside a sentence. Instead of typing numbers by hand, you let R calculate and insert them automatically when the document is rendered.

This keeps your writing and your analysis synchronized.

Why inline code matters in public health

Public health decisions depend on accurate numbers. Inline code:

  • prevents copy and paste errors
  • ensures reported values match the analysis
  • allows results to update automatically
  • supports transparency and trust

In this course, all reported statistics should come from inline code whenever possible.

A simple example

First create a vector of values.

ages <- c(23, 35, 41, 29, 50)

Now write a sentence using inline code.

Tip

This includes a backtick, the letter r, a space, your R code, and a closing backtick.

The sample included 5 participants, and the mean age was 35.6 years.

When you render the document, those numbers are inserted automatically.

What happens when data change?

Change one value.

ages[1] <- 28

Now re-render the document.

The mean age is now 36.6 years.

Because the number comes from inline code, it updates automatically.

Formatting inline values

You will often need to format numbers.

Rounding

Use the round() function.

The average age was 36.6 years.

Percentages

prop_smoker <- 0.237

Approximately 23.7% of participants were smokers.

Sample sizes

The analytic sample included 5 participants.

Inline code with small datasets

df <- data.frame(
  bmi = c(21.7, 27.4, 31.2, 29.8, 26.3, 24.9),
  smoker = c("No", "Yes", "No", "No", "No", "Yes")
)
mean_bmi <- mean(df$bmi)
sd_bmi <- sd(df$bmi)
prop_smoker <- mean(df$smoker == "Yes")

The mean BMI was 26.9 kg/m² (SD = 3.4), and 33.3% of participants reported smoking.

Inline code with regression models

m1 <- lm(bmi ~ smoker, data = df)
coef(m1)
(Intercept)   smokerYes 
      27.25       -1.10 
b_smoker <- coef(m1)[2]

Compared with non-smokers, smokers had an estimated BMI difference of -1.1 units in this sample.

Common mistakes to avoid

Avoid these errors:

  • Typing numbers into your paper by hand
  • Copying results from the Console
  • Forgetting to re-render after changing data
  • Using inline code before the object exists

If a number appears in your report, it should almost always come from inline code.

Course expectations

In Projects 1–4:

  • summary statistics should be written using inline code
  • regression coefficients should be reported using inline code
  • tables and figures should be generated in R

This is how real scientific manuscripts are written.

Practice section

Create a section in your own homework file and complete the following:

  1. Create a numeric vector with at least 6 values
  2. Use inline code to report:
    • sample size
    • mean
    • standard deviation
  3. Change one value
  4. Re-render and confirm the sentence updates

Starter:

x <- c(4, 6, 9, 12, 15, 18)

Write a sentence like:

The sample size was 6, the mean was 10.67, and the standard deviation was 5.35.

Key takeaways

Inline code connects your writing to your analysis. It keeps numbers accurate, makes your work reproducible, and is a core skill for data-driven public health.

Looking ahead

You will use inline code in every project, especially when writing your final manuscript in Quarto.