Exploratory Data Analysis With R

Exploratory Data Analysis

How can you do this work?

Now, we don’t want to totally discredit the use of paper

• Famous Historical Examples

• Can be Excellent for Planning

But now that it’s 2025, how can we do this work?

But now that it’s 2025, how can we do this work?

• The coding language = R

• The Integrated Development Environment (IDE) = Posit Cloud

• How to think of this: You write an english paper in Google Docs. In that case english is the language and you use the Google Docs environment to actually type and edit that paper. 

Posit Cloud Layout

RStudio Project Layout

Packages

Start by Always Getting Your Toolboxes

• You do this by running the library function for each package you are going to use.

• For this tutorial, we will use.

  • Tidyverse is made up of a collection of packages that include tools making it easier to visualize, summarize, and wrangle data.

  • Openintro is a package referenced in our textbook, Introduction to Modern Statistics. Among other things, it includes a number of datasets. During this tutorial, you will generate similar visualizations to those seen when reading.

• The code you type and run

library(tidyverse)
library(openintro)
?births14

Questions to Ask to Start Getting to know your data

• What are the observational units/cases?

• What are the variables?

• How is my data organized?

  • You want Tidy Data

    • Each row is an observation

    • Each column is a variable

Data for this Introduction to EDA

• Throughout the rest of this tutorial we will explore the births14

  • This dataset is included in the openintro package

  • Since it is part of an R package can type ?births14 to access its documentation and learn more about it.

• Description

  • “Every year, the US releases to the public a large data set containing information on births recorded in the country. This data set has been of interest to medical researchers who are studying the relation between habits and practices of expectant mothers and the birth of their children. This is a random sample of 1,000 cases from the data set released in 2014.”[@openintro]

Answering Questions to Start Getting to Know the Births14 dataset

• Running the head function allows you to see the first 6 rows of the data frame

  • Type and run this code

    head(births14)

    # A tibble: 6 × 13
       fage  mage mature      weeks premie visits gained weight lowbirthweight sex  
      <int> <dbl> <chr>       <dbl> <chr>   <dbl>  <dbl>  <dbl> <chr>          <chr>
    1    34    34 younger mom    37 full …     14     28   6.96 not low        male 
    2    36    31 younger mom    41 full …     12     41   8.86 not low        fema…
    3    37    36 mature mom     37 full …     10     28   7.51 not low        fema…
    4    NA    16 younger mom    38 full …     NA     29   6.19 not low        male 
    5    32    31 younger mom    36 premie     12     48   6.75 not low        fema…
    6    32    26 younger mom    39 full …     14     45   6.69 not low        fema…
    # ℹ 3 more variables: habit <chr>, marital <chr>, whitemom <chr>

Answering Questions to Start Getting to Know the Births14 dataset

• The glimpse function allows you to see how many variables, what type each is, and how many observations are in the data frame.

  • Type and Run this code

    glimpse(births14)

    Rows: 1,000
    Columns: 13
    $ fage           <int> 34, 36, 37, NA, 32, 32, 37, 29, 30, 29, 30, 34, 28, 28,…
    $ mage           <dbl> 34, 31, 36, 16, 31, 26, 36, 24, 32, 26, 34, 27, 22, 31,…
    $ mature         <chr> "younger mom", "younger mom", "mature mom", "younger mo…
    $ weeks          <dbl> 37, 41, 37, 38, 36, 39, 36, 40, 39, 39, 42, 40, 40, 39,…
    $ premie         <chr> "full term", "full term", "full term", "full term", "pr…
    $ visits         <dbl> 14, 12, 10, NA, 12, 14, 10, 13, 15, 11, 14, 16, 20, 15,…
    $ gained         <dbl> 28, 41, 28, 29, 48, 45, 20, 65, 25, 22, 40, 30, 31, NA,…
    $ weight         <dbl> 6.96, 8.86, 7.51, 6.19, 6.75, 6.69, 6.13, 6.74, 8.94, 9…
    $ lowbirthweight <chr> "not low", "not low", "not low", "not low", "not low", …
    $ sex            <chr> "male", "female", "female", "male", "female", "female",…
    $ habit          <chr> "nonsmoker", "nonsmoker", "nonsmoker", "nonsmoker", "no…
    $ marital        <chr> "married", "married", "married", "not married", "marrie…
    $ whitemom       <chr> "white", "white", "not white", "white", "white", "white…

Now that you know a little about the data, start asking questions.

• Keep it simple to start.

• Know that, you’ll go around the wheel a lot and new questions will arise following your conclusions, but you have to start somewhere.

• Examples:

  • How much do babies in this sample tend to weigh?

  • How many weeks did pregnancies tend to last?

  • What percent of babies were born to mothers who were married?

How much do babies in this sample tend to weigh?

• Start by visualizing the distribution of the weight variable.

  births14 |>
    ggplot(aes(x = weight)) +
    geom_histogram(binwidth = 0.5) +
    labs(
      title = "How much do babies in this sample weigh at birth?",
      x = "Weight (Pounds)",
      y = "Number of Babies"
    )

• Notes:

  • |> is a pipe symbol and should be read as “AND THEN”

  • + is used to add layers to a ggplot.

How much do babies in this sample weigh? Statistics

• We now need to generate statistics to better describe the weights.

• Central Tendency

  • Mean

  • Median

• Spread of the data

  • Standard Deviation: Average distance from the mean

  • Interquartile Range: The middle 50% of the data

Code to generate these statistics

1births14 |>
2  filter(!is.na(weight)) |>
3  summarize(
4    Mean = mean(weight),
5    Median = median(weight),
6    Standard_Deviation = sd(weight),
7    IQR = IQR(weight)
  )

# A tibble: 1 × 4
   Mean Median Standard_Deviation   IQR
  <dbl>  <dbl>              <dbl> <dbl>
1  7.20   7.31               1.31  1.46

How much do babies in this sample tend to weigh?

# A tibble: 1 × 4
   Mean Median Standard_Deviation   IQR
  <dbl>  <dbl>              <dbl> <dbl>
1  7.20   7.31               1.31  1.46

Now that you’ve learned something, what new question comes to mind?

What is the relationship between an expecting mother’s smoking habits and the weight of their baby?

More about ggplot layers

What is the relationship between an expecting mother’s smoking habits and the weight of their baby? Visualize

births14 |>
  filter(!is.na(habit)) |>
  ggplot(aes(x = weight)) +
  geom_histogram(binwidth = 0.5) +
  facet_wrap(~habit, nrow = 2)

What is the relationship between an expecting mother’s smoking habits and the weight of their baby? Statistics

The group_by function does as it says. It puts babies, in this case, into two groups, one for those with moms who smoked and one for those with moms who did not smoke.

births14 |> 
  filter(!is.na(weight)) |> 
  group_by(habit) |>
  summarize( 
    Mean = mean(weight), 
    Median = median(weight), 
    Standard_Deviation = sd(weight), 
    IQR = IQR(weight) 
  )

# A tibble: 3 × 5
  habit      Mean Median Standard_Deviation   IQR
  <chr>     <dbl>  <dbl>              <dbl> <dbl>
1 nonsmoker  7.27   7.35               1.23  1.49
2 smoker     6.68   7.03               1.60  1.85
3 <NA>       7.05   7.1                1.91  1.99

Is there a relationship between the weight of a baby and length of the pregnancy?

births14 |>
1  ggplot(aes(x=weeks, y = weight)) +
2  geom_point() +
3  labs(
    title = "Weight vs Length of Pregnancy",
    x = "Length of Pregnancy (weeks)",
    y = "Weight (pounds)"
  )

Adding a trend line to a scatter plot

births14 |>
  ggplot(aes(x=weeks, y = weight)) + 
  geom_point() + 
1  geom_smooth(method = 'lm') +
  labs(
    title = "Weight vs Length of Pregnancy",
    x = "Length of Pregnancy (weeks)",
    y = "Weight (pounds)"
  )

Bibliography

Rochette, R., & Lê, S. (2020). Grammar of graphics: gg basics. QCBS R Workshop Series. 1