This project investigates the relationship between car horsepower and highway fuel efficiency. The dataset contains 5,076 observations and 18 variables describing cars, including make, model year, fuel type, transmission, horsepower, torque, and city and highway miles per gallon (MPG).
Horsepower measures engine power, while highway MPG indicates how many miles a vehicle can travel per gallon of fuel on the highway. These variables are quantitative because they are measured numerically. Fuel type and car make are categorical variables because they identify groups.
My main research question is, “Do cars with higher horsepower tend to have lower highway fuel efficiency?” I will use summary statistics, a scatterplot, and linear regression to investigate this relationship. I will also use fuel type to distinguish groups in the visualization.
# Load the libraries needed for data cleaning and visualization.library(tidyverse)
── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr 1.2.1 ✔ readr 2.2.0
✔ forcats 1.0.1 ✔ stringr 1.6.0
✔ ggplot2 4.0.3 ✔ tibble 3.3.1
✔ lubridate 1.9.5 ✔ tidyr 1.3.2
✔ purrr 1.2.2
── 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
library(ggplot2)# Import the car dataset from the CSV file.cars <-read_csv("C:/Users/akeeb/Downloads/cars.csv")
Rows: 5076 Columns: 18
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): Engine Information.Driveline, Engine Information.Engine Type, Engin...
dbl (9): Dimensions.Height, Dimensions.Length, Dimensions.Width, Engine Info...
lgl (1): Engine Information.Hybrid
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
# Standardize column names so they are easier to reference in R.names(cars) <-make.names(names(cars))# Display the first six rows of the dataset.head(cars)
# Rename the columns used in the project.cars <- cars %>%rename(make = Identification.Make,year = Identification.Year,fuel_type = Fuel.Information.Fuel.Type,city_mpg = Fuel.Information.City.mpg,highway_mpg = Fuel.Information.Highway.mpg,horsepower = Engine.Information.Engine.Statistics.Horsepower,torque = Engine.Information.Engine.Statistics.Torque )# Remove rows with missing values in the variables used for analysis.cars <- cars %>%drop_na(make, year, fuel_type, city_mpg, highway_mpg, horsepower, torque)# Display the number of rows and columns after cleaning.dim(cars)
[1] 5076 18
# Display summary statistics for the cleaned dataset.summary(cars)
Dimensions.Height Dimensions.Length Dimensions.Width
Min. : 1.0 Min. : 2.0 Min. : 1
1st Qu.:104.0 1st Qu.: 60.0 1st Qu.: 62
Median :152.0 Median :128.0 Median :158
Mean :145.6 Mean :127.8 Mean :144
3rd Qu.:193.0 3rd Qu.:198.0 3rd Qu.:219
Max. :255.0 Max. :255.0 Max. :254
Engine.Information.Driveline Engine.Information.Engine.Type
Length :5076 Length :5076
N.unique : 4 N.unique : 535
N.blank : 0 N.blank : 0
Min.nchar: 15 Min.nchar: 28
Max.nchar: 17 Max.nchar: 60
Engine.Information.Hybrid Engine.Information.Number.of.Forward.Gears
Mode:logical Min. :4.000
TRUE:5076 1st Qu.:5.000
Median :6.000
Mean :5.519
3rd Qu.:6.000
Max. :8.000
Engine.Information.Transmission city_mpg fuel_type
Length :5076 Min. : 8.00 Length :5076
N.unique : 11 1st Qu.:14.00 N.unique : 4
N.blank : 0 Median :17.00 N.blank : 0
Min.nchar: 14 Mean :17.28 Min.nchar: 3
Max.nchar: 30 3rd Qu.:20.00 Max.nchar: 22
Max. :38.00
highway_mpg Identification.Classification Identification.ID
Min. : 11.00 Length :5076 Length :5076
1st Qu.: 20.00 N.unique : 2 N.unique :5030
Median : 24.00 N.blank : 0 N.blank : 0
Mean : 24.13 Min.nchar: 19 Min.nchar: 12
3rd Qu.: 28.00 Max.nchar: 22 Max.nchar: 67
Max. :223.00
make Identification.Model.Year year horsepower
Length :5076 Length :5076 Min. :2009 Min. :100.0
N.unique : 47 N.unique : 918 1st Qu.:2010 1st Qu.:190.0
N.blank : 0 N.blank : 0 Median :2011 Median :266.0
Min.nchar: 3 Min.nchar: 11 Mean :2011 Mean :270.5
Max.nchar: 18 Max.nchar: 48 3rd Qu.:2011 3rd Qu.:317.0
Max. :2012 Max. :638.0
torque
Min. : 98.0
1st Qu.:187.0
Median :260.0
Mean :272.7
3rd Qu.:335.0
Max. :774.0
# Examine horsepower, highway MPG, and city MPG.summary(cars$horsepower)
Min. 1st Qu. Median Mean 3rd Qu. Max.
100.0 190.0 266.0 270.5 317.0 638.0
summary(cars$highway_mpg)
Min. 1st Qu. Median Mean 3rd Qu. Max.
11.00 20.00 24.00 24.13 28.00 223.00
summary(cars$city_mpg)
Min. 1st Qu. Median Mean 3rd Qu. Max.
8.00 14.00 17.00 17.28 20.00 38.00
# Count the number of cars in each fuel category.table(cars$fuel_type)
# Plot horsepower against highway MPG to investigate their relationship.# Color the points by fuel type to compare categorical groups.# Add a meaningful title, axis labels, legend, and source caption.# Use a custom color palette and change the default ggplot theme.ggplot(cars, aes(x = horsepower, y = highway_mpg, color = fuel_type)) +geom_point(alpha =0.6) +scale_color_manual(values =c("Gasoline"="#0072B2","E85"="#D55E00","Diesel fuel"="#009E73","Compressed natural gas"="#CC79A7" )) +labs(title ="Horsepower and Highway Fuel Efficiency",x ="Horsepower",y ="Highway Fuel Efficiency (MPG)",color ="Fuel Type",caption ="Source: Insert the original dataset source and link" ) +theme_minimal()
# Build a linear regression model predicting highway MPG using horsepower.model <-lm(highway_mpg ~ horsepower, data = cars)# Display the regression coefficients, p-value, and adjusted R-squared.summary(model)
Call:
lm(formula = highway_mpg ~ horsepower, data = cars)
Residuals:
Min 1Q Median 3Q Max
-13.824 -3.026 0.198 3.243 201.170
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 34.2220194 0.2292682 149.27 <2e-16 ***
horsepower -0.0373255 0.0007994 -46.69 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 5.427 on 5074 degrees of freedom
Multiple R-squared: 0.3005, Adjusted R-squared: 0.3004
F-statistic: 2180 on 1 and 5074 DF, p-value: < 2.2e-16
# Arrange the diagnostic plots in a two-by-two layout.par(mfrow =c(2, 2))# Check the regression model assumptions using diagnostic plots.plot(model)
# Return the plotting layout to normal.par(mfrow =c(1, 1))
Regression Results
The linear regression model examines whether horsepower is associated with highway fuel efficiency. The estimated regression equation is: Highway MPG = 34.222 − 0.03733 × Horsepower
The slope is approximately −0.03733, meaning that each additional unit of horsepower is associated with a predicted decrease of about 0.037 highway MPG. This describes an average relationship in the dataset, not a guaranteed change for every individual car. The p-value for horsepower is below 0.001, which is less than the significance level of 0.05. Therefore, there is strong statistical evidence of a linear association between horsepower and highway MPG. The adjusted R-squared value is approximately 0.300. This means that the model accounts for about 30% of the observed variation in highway MPG. Horsepower is related to fuel efficiency, but it does not explain all the differences between vehicles. Other factors, such as vehicle weight, engine design, and transmission, may also contribute.
Conclusion and Reflection
I began by importing the original CSV file using the readr package and examining its dimensions, column names, and data types. The dataset contained 5,076 rows and 18 variables. I used the make.names() function to standardize the column names, replacing spaces and other characters with periods so the names could be referenced more easily in R. I then renamed seven variables used in the analysis, including horsepower, highway MPG, city MPG, fuel type, and car make. I checked for missing values using is.na() and removed incomplete observations with drop_na(). However, the original dataset did not contain missing values in the selected variables, so this step did not remove any rows. Finally, I used summary statistics and frequency counts to examine the data before creating the visualization and regression model. The scatterplot compares horsepower with highway fuel efficiency, measured in miles per gallon. Each point represents a car, and its color identifies the fuel type. The linear regression results indicate a negative association, meaning that cars with higher horsepower tend to have lower highway MPG. The graph also allows comparisons between fuel types, although gasoline accounts for most observations. Compressed natural gas has only two observations, and diesel has 27, so conclusions about these smaller groups should be made cautiously. One limitation is that the model uses only horsepower to predict highway MPG. Its adjusted R-squared value of approximately 0.300 indicates that horsepower alone explains about 30% of the variation in highway fuel efficiency. I would like to investigate additional variables, such as vehicle dimensions, torque, transmission type, and fuel type, to see whether they improve the model. A multiple linear regression could examine several predictors simultaneously. I would also consider creating separate visualizations for different fuel types to make group comparisons easier. Finally, I would examine the diagnostic plots carefully because unusual observations or violations of regression assumptions could affect the reliability of the model.
