For this dataset, I will use the Social Security Administration’s historical table of the five most popular male and female baby names by year. The original data is presented in a wide format, with separate columns for each ranked female and male name. I will first recreate and save the original structure as a CSV file so that the raw data is preserved before any transformations are performed. Using tidyr and dplyr, I will transform the dataset from wide to long format so that each row represents a single name for a particular year, sex, and rank. Instead of maintaining separate columns for each ranking and sex, the tidy dataset will contain variables such as year, sex, rank, and name. I will also standardize the variable names and check for missing or inconsistent values. After tidying the data, I will analyze how the most popular baby names have changed over time. I plan to identify names that appeared in the top five most frequently, determine which names held the number-one position most often, and examine changes in naming patterns across years or decades. Visualizations will be used to show changes in name popularity and compare patterns between male and female names.
Data Source
The data for this analysis comes from the Social Security Administration (SSA) and contains the five most popular baby names for girls and boys for each year from 1926 through 2025. The original data was saved as baby_names_raw.csv and committed to GitHub before any transformations were performed.
The raw dataset is stored in a wide format, with separate columns for each combination of sex and rank.
Year Girls_Rank_1 Girls_Rank_2 Girls_Rank_3 Girls_Rank_4 Girls_Rank_5
1 2025 Olivia Charlotte Emma Amelia Sophia
2 2024 Olivia Emma Amelia Charlotte Mia
3 2023 Olivia Emma Charlotte Amelia Sophia
4 2022 Olivia Emma Charlotte Amelia Sophia
5 2021 Olivia Emma Charlotte Amelia Ava
6 2020 Olivia Emma Ava Sophia Charlotte
Boys_Rank_1 Boys_Rank_2 Boys_Rank_3 Boys_Rank_4 Boys_Rank_5
1 Liam Noah Oliver Theodore Henry
2 Liam Noah Oliver Theodore James
3 Liam Noah Oliver James Elijah
4 Liam Noah Oliver James Elijah
5 Liam Noah Oliver Elijah James
6 Liam Noah Oliver Elijah William
In the original dataset, each row represents one year. However, the baby names are spread across ten separate columns: five ranked names for girls and five ranked names for boys. Therefore, information about sex and rank is stored within the column names rather than as separate variables.
This wide structure is useful for displaying rankings, but it is not ideal for analysis. A tidy structure should have one observation per row and separate variables for year, sex, rank, and name.
Year Girls_Rank_1 Girls_Rank_2 Girls_Rank_3 Girls_Rank_4 Girls_Rank_5
1 2025 Olivia Charlotte Emma Amelia Sophia
2 2024 Olivia Emma Amelia Charlotte Mia
3 2023 Olivia Emma Charlotte Amelia Sophia
4 2022 Olivia Emma Charlotte Amelia Sophia
5 2021 Olivia Emma Charlotte Amelia Ava
6 2020 Olivia Emma Ava Sophia Charlotte
Boys_Rank_1 Boys_Rank_2 Boys_Rank_3 Boys_Rank_4 Boys_Rank_5
1 Liam Noah Oliver Theodore Henry
2 Liam Noah Oliver Theodore James
3 Liam Noah Oliver James Elijah
4 Liam Noah Oliver James Elijah
5 Liam Noah Oliver Elijah James
6 Liam Noah Oliver Elijah William
Transformation Steps
To convert the dataset from wide to tidy format, I use pivot_longer() to combine the ten ranking columns into two temporary variables: one containing the original column name and another containing the corresponding baby name.
I then use separate() to divide the original column names into sex and rank. Finally, I standardize the sex labels and convert rank to a numeric variable.
# A tibble: 10 × 4
year sex rank name
<int> <chr> <int> <chr>
1 1926 Female 1 Mary
2 1926 Female 2 Dorothy
3 1926 Female 3 Betty
4 1926 Female 4 Helen
5 1926 Female 5 Margaret
6 1926 Male 1 Robert
7 1926 Male 2 John
8 1926 Male 3 James
9 1926 Male 4 William
10 1926 Male 5 Charles
After tidying, each row represents one ranked baby name for a specific year and sex. The final variables are year, sex, rank, and name. Missing values are checked after transformation so that they can be identified before beginning the analysis.
Analytical Methods
The analysis uses only the tidy version of the dataset. I examine which names appeared most frequently in the top five rankings, which names held the number-one position most often, and how the most popular names changed over time. I also compare patterns between female and male names. Counts and grouped summaries are calculated using dplyr, and visualizations are created using ggplot2.
Analysis 1: Most Frequent “Top 5” Names
Code
top_names <- baby_names_tidy %>%count(sex, name, sort =TRUE) %>%group_by(sex) %>%slice_max(order_by = n, n =10, with_ties =FALSE) %>%ungroup()top_names
# A tibble: 20 × 3
sex name n
<chr> <chr> <int>
1 Female Mary 42
2 Female Emma 24
3 Female Linda 23
4 Female Patricia 22
5 Female Barbara 21
6 Female Jennifer 21
7 Female Jessica 21
8 Female Olivia 21
9 Female Sarah 19
10 Female Ashley 18
11 Male James 62
12 Male Michael 62
13 Male John 47
14 Male Robert 46
15 Male David 39
16 Male William 36
17 Male Christopher 29
18 Male Joshua 26
19 Male Matthew 26
20 Male Jacob 21
Analysis 2: Visualize the Most Frequent “Top 5” Names
Code
ggplot(top_names, aes(x =reorder(name, n), y = n, fill = sex)) +geom_col(show.legend =FALSE) +coord_flip() +facet_wrap(~sex, scales ="free_y") +scale_fill_manual(values =c("Female"="lightpink", "Male"="lightblue") ) +labs(title ="Most Frequent Top-Five Baby Names, 1926–2025",x ="Name",y ="Number of Years in the Top Five" ) +theme_minimal()
Analysis 3: Most Frequent #1 Names
Code
number_one_names <- baby_names_tidy %>%filter(rank ==1) %>%count(sex, name, sort =TRUE) %>%group_by(sex) %>%slice_max(order_by = n, n =10, with_ties =FALSE) %>%ungroup()number_one_names
# A tibble: 17 × 3
sex name n
<chr> <chr> <int>
1 Female Mary 30
2 Female Jennifer 15
3 Female Emily 12
4 Female Jessica 9
5 Female Lisa 8
6 Female Olivia 7
7 Female Emma 6
8 Female Linda 6
9 Female Sophia 3
10 Female Ashley 2
11 Male Michael 44
12 Male Robert 15
13 Male Jacob 14
14 Male James 13
15 Male Liam 9
16 Male Noah 4
17 Male David 1
Analysis 4: Time Based Visualization of Names
Code
number_one_by_year <- baby_names_tidy %>%filter(rank ==1)ggplot(number_one_by_year, aes(x = year, y = name, fill = sex)) +geom_tile() +facet_wrap(~sex, scales ="free_y") +scale_fill_manual(values =c("Female"="lightpink", "Male"="lightblue") ) +labs(title ="Number-One Baby Names Over Time, 1926–2025",subtitle ="Each tile represents a year in which a name ranked #1",x ="Year",y ="Name" ) +guides(fill ="none") +theme_minimal()
Results
The analysis shows that several names remained popular for long periods of time. Among female names, Mary appeared in the top five for 42 of the 100 years, substantially more often than any other female name. Emma appeared for 24 years and Linda for 23 years. Among male names, James and Michael each appeared in the top five for 62 years, followed by John with 47 years and Robert with 46 years.
Looking specifically at the number-one ranked name reveals even stronger patterns. Mary was the most frequently ranked number-one female name, holding the top position for 30 years. Jennifer followed with 15 years, while Emily held the top position for 12 years. For male names, Michael was especially dominant, ranking number one for 44 years. Robert ranked first for 15 years, Jacob for 14 years, and James for 13 years.
The visualization of number-one names over time also shows that popularity often occurred in distinct periods. Female number-one names changed more frequently, while male rankings were dominated by a smaller number of names for longer periods. Michael’s extended period as the number-one male name is particularly visible in the data.
Transforming the SSA baby-name data from wide to long format made it much easier to analyze changes in name popularity across years, sex, and rank. Instead of storing sex and rank information within separate column names, the tidy dataset represents these characteristics as individual variables that can be grouped, filtered, summarized, and visualized directly.
The analysis shows that some names maintained popularity for decades. Mary was the most persistent female name in both top-five appearances and number-one rankings, while James and Michael had the most top-five appearances among male names. Michael was particularly dominant at rank one, holding the position for 44 of the 100 years analyzed. Overall, the results also suggest greater turnover among number-one female names than among male names during this period.
