Comparing Visualization Systems in R: Base Graphics, Lattice, and ggplot2


How do the syntax and workflow differ between base, lattice, and ggplot2?

Base R has a simple function depending on the graph; you just need the specific variables and can label them to create an easy-to-read basic graph. Lattice is good for comparing multiple variables by group, two graphs side by side, using a formula-based syntax. ggplot2 uses multiple functions to create a more advanced graph. Can customize a wider range of aspects to the visualizations, including background, color, or shape in a much easier way.


Which system gave you the most control or produced the most “publication‑quality” output with minimal code?

The ggplot2 system gave me the most control and produced a higher quality output compared to the other two. It gives clearer visuals with color comparison and automatically adds legends with minimal code.


Any challenges or surprises you encountered when switching between systems?

My main challenge was with ggplot2, which has a lot more functions to memorize compared to Base R and lattice, which rely heavily on just the variables. It was interesting to see how the different visuals compare to each other and how graphs that we are used to drawing translate to different types of code.

 

Using the dataset MurderRates, representing Determinants of Murder Rates in the United States, I created two visuals each for Base R, Lattice, and ggplot2. 


Task 1 Base R Graphics

# Scatter plot
plot(MurderRates$income, MurderRates$rate,
     main   = "Base: Income vs. Murder Rate",
     xlab   = "Income",
     ylab   = "Rate")
# Histogram
hist(MurderRates$rate,
     main   = "Base: Distribution of Murder Rates",
     xlab   = "Rate")

Task 2 Lattice Graphics

# Conditional scatter plot (small multiples)
xyplot(rate ~ income | factor(southern),
       data = MurderRates,
       main = "Lattice: Murder Rate vs. Income by the South")

# Box-and-whisker plot
bwplot(rate ~ factor(southern),
       data = MurderRates,
       main = "Lattice: Murder Rate by the South")

Task 3 ggplot2

# Scatter plot with smoothing
ggplot(MurderRates, aes(x = income, y = rate, color = factor(southern))) +
  geom_point() +
  geom_smooth(method = "lm") +
  labs(title = "ggplot2: Murder Rate vs. Income with trend by Southern Group")

# Faceted histogram
ggplot(MurderRates, aes(rate)) +
  geom_histogram(binwidth = 1) +
  facet_wrap(~ southern) +
  labs(title = "ggplot2: Murder Rate distribution by Southern Group")






Comments

Post a Comment

Popular posts from this blog

myMean Error and Correction

Using the following vector and function in RStudio, I ran myMean(assignment2). assignment2 <- c(16, 18, 14, 22, 27, 17, 19, 17, 17, 22, 20, 22) myMean <- function(assignment2) { return(sum(assignment) / length(someData)) }    The output was "Error in myMean(assignment2) : object 'assignment' not found".   The function fails because assignment and someData are not assigned variables. These objects do not exist, so trying to pass them to the function myMean(assignment2) results in an error. To correctly return the mean of assignment2, the myMean function needs the variable assignment2 to replace assignment and someData. Corrected version of myMean, which returns 19.25. myMean <- function(assignment2) { return(sum(assignment2) / length(assignment2)) }   
Read more

Election Data Frame

Image
Errors in the original data set: a missing comma between Marco and Carly in Name, I fixed the spelling of Berine to Bernie, a missing comma between 62 and 51 in ABC, quotation marks around names had to be changed to " ", the variable names for ABC and CBS political poll results had spaces, which I changed to _. After fixing all the errors in the data set, I used the three variables Name, ABC_political_poll_results,  and CBS_political_poll_results to create the data frame. Since the data set is based on the 2016  presidential election, I named the data frame election_2016. Here is my corrected code in R: Name <- c("Jeb", "Donald", "Ted", "Marco", "Carly", "Hillary", "Bernie") ABC_political_poll_results <- c(4, 62, 51, 21, 2, 14, 15) CBS_political_poll_results <- c(12, 75, 43, 19, 1, 21, 19)  election_2016 <- data.frame(Name, ABC_political_poll_results, CBS_political_poll_results) election_2016 Af...
Read more

Doing Math on Matrices

I used the following values to find the inverse and determinants of two matrices. A = matrix(1:100, nrow=10) B = matrix(1:1000, nrow=10) First, I assigned the matrices in R.  A <- matrix(1:100, nrow=10) B <- matrix(1:1000, nrow=10) To obtain the determinant of each matrix, I used the det function. det(A) Result:  0 det(B) Result: Error in determinant.matrix(x, logarithm = TRUE, ...) :    'x' must be a square matrix Then, to obtain the inverse of each matrix, I used the solve function. solve(A) Result:  Error in solve.default(A) :    Lapack routine dgesv: system is exactly singular: U[3,3] = 0 solve(B) Result: Error in solve.default(B) : 'a' (10 x 100) must be square Looking at the matrices, we can see that A is a square matrix because it has the same number of rows and columns. There are 100 values in the matrix; dividing that by 10 rows gives me 10 columns. This makes matrix A a 10x10 square matrix. With B, there are 1000 values in the matrix, a...
Read more