Introduction

The article According To Super Bowl Ads, Americans Love America, Animals And Sex [https://projects.fivethirtyeight.com/super-bowl-ads/] references a data set that was manually collected by the website that published it. The data includes observations about 10 brands’ individual super bowl ads from 2000 to 2020. There are seven variables that the FiveThirtyEight team - descriptions for each of them can be found in the article.

Load libraries

library(tidyverse)

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.4
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.4.4     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.0
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

Import the CSV file

Saving the data set as ‘sbc’ for super bowl commercials.

View function to see the data

sbc <- read.csv("https://raw.githubusercontent.com/fivethirtyeight/superbowl-ads/main/superbowl-ads.csv", sep = ',')
glimpse(sbc)

## Rows: 244
## Columns: 11
## $ year                      <int> 2018, 2020, 2006, 2018, 2003, 2020, 2020, 20…
## $ brand                     <chr> "Toyota", "Bud Light", "Bud Light", "Hynudai…
## $ superbowl_ads_dot_com_url <chr> "https://superbowl-ads.com/good-odds-toyota/…
## $ youtube_url               <chr> "https://www.youtube.com/watch?v=zeBZvwYQ-hA…
## $ funny                     <chr> "False", "True", "True", "False", "True", "T…
## $ show_product_quickly      <chr> "False", "True", "False", "True", "True", "T…
## $ patriotic                 <chr> "False", "False", "False", "False", "False",…
## $ celebrity                 <chr> "False", "True", "False", "False", "False", …
## $ danger                    <chr> "False", "True", "True", "False", "True", "T…
## $ animals                   <chr> "False", "False", "True", "False", "True", "…
## $ use_sex                   <chr> "False", "False", "False", "False", "True", …

Cleaning the data

From the view, it can be seen that some values are missing. To tell R that these values are null, the na = “” is used.

The 7 characteristics tested are all True or False values, so those column types should be cleaned to logical.

The right most column titled na is full of null values, this column can be dropped.

sbc <- sbc |>
  mutate(
    #empty cells recognized as null
    na = "",
    #all 7 columns converted to logical type
    funny = as.logical(funny),
    show_product_quickly = as.logical(show_product_quickly),
    patriotic = as.logical(patriotic),
    celebrity = as.logical(celebrity),
    danger = as.logical(danger),
    animals = as.logical(animals),
    use_sex = as.logical(use_sex)
    )

#drop the null column 
sbc <- sbc[-12]

#quick view of the updated data set
glimpse(sbc)

## Rows: 244
## Columns: 11
## $ year                      <int> 2018, 2020, 2006, 2018, 2003, 2020, 2020, 20…
## $ brand                     <chr> "Toyota", "Bud Light", "Bud Light", "Hynudai…
## $ superbowl_ads_dot_com_url <chr> "https://superbowl-ads.com/good-odds-toyota/…
## $ youtube_url               <chr> "https://www.youtube.com/watch?v=zeBZvwYQ-hA…
## $ funny                     <lgl> FALSE, TRUE, TRUE, FALSE, TRUE, TRUE, TRUE, …
## $ show_product_quickly      <lgl> FALSE, TRUE, FALSE, TRUE, TRUE, TRUE, FALSE,…
## $ patriotic                 <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FA…
## $ celebrity                 <lgl> FALSE, TRUE, FALSE, FALSE, FALSE, TRUE, TRUE…
## $ danger                    <lgl> FALSE, TRUE, TRUE, FALSE, TRUE, TRUE, FALSE,…
## $ animals                   <lgl> FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, TRUE,…
## $ use_sex                   <lgl> FALSE, FALSE, FALSE, FALSE, TRUE, FALSE, FAL…

Let’s take a look at one randomly selected row

single_sbc <- sbc[2,]
glimpse(single_sbc)

## Rows: 1
## Columns: 11
## $ year                      <int> 2020
## $ brand                     <chr> "Bud Light"
## $ superbowl_ads_dot_com_url <chr> "https://superbowl-ads.com/2020-bud-light-se…
## $ youtube_url               <chr> "https://www.youtube.com/watch?v=nbbp0VW7z8w"
## $ funny                     <lgl> TRUE
## $ show_product_quickly      <lgl> TRUE
## $ patriotic                 <lgl> FALSE
## $ celebrity                 <lgl> TRUE
## $ danger                    <lgl> TRUE
## $ animals                   <lgl> FALSE
## $ use_sex                   <lgl> FALSE

We can see this set of observations are from a Bud Light commercial in 2020. The video contains elements that are funny, show the product in the first 10 seconds of the commercial, it features a recognizable celebrity, and it involves danger through violence, threats of violence, injuries, fighting or guns.

What can we see through graphs?

The below charts show the frequency of commercials per brand and the ratio that each brand has for the 7 specific measures.

# funny
ggplot(sbc, aes(x = brand, fill = funny)) +
geom_bar() + 
labs(title = "Commercial Funny Counts Per Brand")

# show_product_quickly
ggplot(sbc, aes(x = brand, fill = show_product_quickly)) +
geom_bar() + 
labs(title = "Commercial Product Being Shown Counts Per Brand")

# patriotic
ggplot(sbc, aes(x = brand, fill = patriotic)) +
geom_bar() + 
labs(title = "Commercial Patriotic Counts Per Brand")

# celebrity
ggplot(sbc, aes(x = brand, fill = celebrity)) +
geom_bar() + 
labs(title = "Commercial Celebrity Counts Per Brand")

# danger
ggplot(sbc, aes(x = brand, fill = danger)) +
geom_bar() + 
labs(title = "Commercial Danger Counts Per Brand")

# animals
ggplot(sbc, aes(x = brand, fill = animals)) +
geom_bar() + 
labs(title = "Commercial Animals Counts Per Brand")

# use_sex
ggplot(sbc, aes(x = brand, fill = use_sex)) +
geom_bar() + 
labs(title = "Commercial Sex Usage Counts Per Brand")

DATA607Assignment01RMD

Kaylie Evans