Assignment #9: Visualization in R – Base Graphics, Lattice, and ggplot2

Github Link: https://github.com/rayhankhan-svg/r-programming-assignments


R Code: 

data <- read.csv(file.choose(), stringsAsFactors = FALSE)


head(data)

str(data)


data$rownames <- as.numeric(data$rownames)

data$education <- as.numeric(data$education)


plot(data$rownames, data$education,

     main = "Base: Education by Observation",

     xlab = "Observation",

     ylab = "Education",

     col = "blue")


hist(data$education,

     main = "Base: Distribution of Education",

     xlab = "Education",

     col = "lightgreen")


library(lattice)


xyplot(education ~ rownames | gender,

       data = data,

       main = "Lattice: Education by Observation and Gender")


bwplot(education ~ job,

       data = data,

       main = "Lattice: Education by Job Type")


library(ggplot2)


ggplot(data, aes(x = rownames, y = education, color = gender)) +

  geom_point() +

  geom_smooth(method = "lm") +

  labs(title = "ggplot2: Education by Observation with Trend")


ggplot(data, aes(x = education)) +

  geom_histogram(binwidth = 1, fill = "blue") +

  facet_wrap(~ job) +

  labs(title = "ggplot2: Education Distribution by Job Type")


Output: 

> data <- read.csv(file.choose(), stringsAsFactors = FALSE)

> 

> head(data)

  rownames    job education gender minority

1        1 manage        15   male       no

2        2  admin        16   male       no

3        3  admin        12 female       no

4        4  admin         8 female       no

5        5  admin        15   male       no

6        6  admin        15   male       no

> str(data)

'data.frame': 474 obs. of  5 variables:

 $ rownames : int  1 2 3 4 5 6 7 8 9 10 ...

 $ job      : chr  "manage" "admin" "admin" "admin" ...

 $ education: int  15 16 12 8 15 15 15 12 15 12 ...

 $ gender   : chr  "male" "male" "female" "female" ...

 $ minority : chr  "no" "no" "no" "no" ...

> 

> data$rownames <- as.numeric(data$rownames)

> data$education <- as.numeric(data$education)

> 

> plot(data$rownames, data$education,

+      main = "Base: Education by Observation",

+      xlab = "Observation",

+      ylab = "Education",

+      col = "blue")

> 

> hist(data$education,

+      main = "Base: Distribution of Education",

+      xlab = "Education",

+      col = "lightgreen")

> 

> library(lattice)

> 

> xyplot(education ~ rownames | gender,

+        data = data,

+        main = "Lattice: Education by Observation and Gender")

> 

> bwplot(education ~ job,

+        data = data,

+        main = "Lattice: Education by Job Type")

> 

> library(ggplot2)

> 

> ggplot(data, aes(x = rownames, y = education, color = gender)) +

+   geom_point() +

+   geom_smooth(method = "lm") +

+   labs(title = "ggplot2: Education by Observation with Trend")

`geom_smooth()` using formula = 'y ~ x'

> 

> ggplot(data, aes(x = education)) +

+   geom_histogram(binwidth = 1, fill = "blue") +

+   facet_wrap(~ job) +

+   labs(title = "ggplot2: Education Distribution by Job Type")

> 

























Explanation: 

For Module #9, I compared three R visualization systems, base graphics, lattice, and ggplot2, using the BankWages dataset. I chose education as the primary numerical variable for visualization because this dataset includes variables like rownames, jobs, education, and gender. I produced a scatter plot of education by observation number and a histogram illustrating the distribution of education using base R graphics. Although base visuals were simple and easy to use, each plot element had to be manually specified. Despite being straightforward, after the plot is established, they become less adaptable. I used bwplot() to compare education across job categories and xyplot() to show education by observation conditioned on gender using lattice graphics. Lattice improves cross-category comparison by making it simpler to work with grouped data and producing multi-panel charts automatically. Lastly, I made a scatter plot using ggplot2 that included a faceted histogram per job type and a regression line. Plot customization and extension are made easier by ggplot2's tiered architecture. In comparison to the other systems, it also generated graphics that were the clearest and most aesthetically pleasing. Lattice worked well for grouped and panel data, Base R was the most straightforward, and ggplot2 offered the greatest flexibility and high-quality visuals. The primary difficulty was adjusting to the various syntactic styles that each system employed.

Comments

Post a Comment

Popular posts from this blog

Assignment #1

Image
GitHub link:  https://github.com/rayhankhan-svg/r-programming-assignments R and RStudio:  I ran into a few little setup and system compatibility problems when installing R and RStudio. R was easy to install from CRAN, however at first I wasn't sure which version to obtain for my operating system. I verified the correct installer for my OS after carefully examining the CRAN download page, and I continued without any more issues. The installation of RStudio went quite smoothly as well, but in order for RStudio to properly identify R, I had to make sure R was fully installed first.  Another problem I ran into was that when I first launched RStudio, output did not appear in the Console right away. The Console started operating regularly after restarting RStudio and confirming that R was correctly installed and identified in the system path. Also I verified that both R and RStudio were running the most recent versions by looking up their versions in the Console. By paying...
Read more

Module # 6 Doing math in R part 2

Github Link: https://github.com/rayhankhan-svg/r-programming-assignments R Code:  # 1) Matrices A and B A <- matrix(c(2,0,1,3), ncol=2) B <- matrix(c(5,2,4,-1), ncol=2) A B # a) A + B A_plus_B <- A + B A_plus_B # b) A - B A_minus_B <- A - B A_minus_B # 2) diag() matrix size 4 with diagonal values 4,1,2,3 D <- diag(c(4,1,2,3)) D # 3) Generate the 5x5 matrix using diag() M <- diag(3, 5)      # start with 3s on the diagonal M[1, 2:5] <- 1       # first row (cols 2-5) become 1 M[2:5, 1] <- 2       # first column (rows 2-5) become 2 M Output:  > # 1) Matrices A and B > A <- matrix(c(2,0,1,3), ncol=2) > B <- matrix(c(5,2,4,-1), ncol=2) >  > A      [,1] [,2] [1,]    2    1 [2,]    0    3 > B      [,1] [,2] [1,]    5    4 [2,]    2   -1 >  > # a) A + B ...
Read more

Module # 3 data.frame

Code Repository: https://github.com/rayhankhan-svg/r-programming-assignments/blob/main/README.md R Code:  # Create vectors Name <- c("Jeb", "Donald", "Ted", "Marco", "Carly", "Hillary", "Bernie") ABC <- c(4, 62, 51, 21, 2, 14, 15) CBS <- c(12, 75, 43, 19, 1, 21, 19) # Create data frame poll_results <- data.frame(Name, ABC, CBS) # View data frame poll_results # Compare polls poll_results$Difference <- poll_results$CBS - poll_results$ABC poll_results Output:  > # Compare polls > poll_results$Difference <- poll_results$CBS - poll_results$ABC > poll_results      Name ABC CBS Difference 1     Jeb   4  12          8 2  Donald  62  75         13 3     Ted  51  43         -8 4   Marco  21  19         -2 5   Carly   2   1...
Read more