Side by Side Bar Chart Funny

Table Of Contents

• Simulated Coin Flip Data
• Side By Side Bar Graphs
• Using facet_grid()
• References

Simulated Coin Flip Data

The ggplot2 package is first loaded into R.

            library("ggplot2")          

We have two players A and B who each an unfair coin. This unfair coin has a 30% chance of getting heads and a 70% chance of getting tails. Both players flip this coin 1000 times each. This can be simulated in R using the sample() function.

            # Flip a unfair coin 1000 times   flip_results_A <- sample(c("Heads", "Tails"), size = 1000, replace = TRUE, prob = c(0.3, 0.7)) flip_results_B <- sample(c("Heads", "Tails"), size = 1000, replace = TRUE, prob = c(0.3, 0.7))          

We can see the results of the coin flips using the table() function.

            # Get counts of the die rolls:  table(flip_results_A) ; table(flip_results_B)          

            ## flip_results_A ## Heads Tails  ##   294   706          

            ## flip_results_B ## Heads Tails  ##   284   716          

The next lines of code is about putting the results together in a data table format.

            # Merge the results from Player A and Player B together: # Format: Player A | Player A flip results (1000 rows) # Player B | Player B flip results (1000 rows)  column1 <- c(rep("Player A", 1000), rep("Player B", 1000)) column2 <- c(flip_results_A, flip_results_B)  outcome_data <- cbind(column1, column2)  flip_results <- data.frame(Player = factor(column1, levels = c("Player A", "Player B")),                 Result = column2)  # A peek of the merged data: head(flip_results); tail(flip_results)          

            ##     Player Result ## 1 Player A  Heads ## 2 Player A  Tails ## 3 Player A  Heads ## 4 Player A  Tails ## 5 Player A  Tails ## 6 Player A  Tails          

            ##        Player Result ## 1995 Player B  Tails ## 1996 Player B  Tails ## 1997 Player B  Tails ## 1998 Player B  Tails ## 1999 Player B  Tails ## 2000 Player B  Tails          

These coin flip results can be converted into a two by two table as a different visual. For graphing purposes, this 2 by 2 table is converted into a data frame.

            table(flip_results)          

            ##           Result ## Player     Heads Tails ##   Player A   294   706 ##   Player B   284   716          

            # The two by two table into a data frame for plotting:  results <- data.frame(table(flip_results))          

Side By Side Bar Graphs

To obtain side by side bar graphs in ggplot2, we need a lot of parts on top of the ggplot() command.

• geom_bar(stat = "identity", position = position_dodge(), alpha = 0.75) gives the side by side bar graphs

• ylim(0, 800) gives limits on the y-axis values

• The geom_text() line adds labels to the bar graphs. Note that position_dodge is needed as we used position dodge was used in geom_bar().

• labs() gives labels depending on what is specified

• The theme() function allows for additional aesthetic options such as a centered title and font sizes.

            # Plotting side by side bar graphs:  # http://www.cookbook-r.com/Graphs/Bar_and_line_graphs_(ggplot2)/ # R Graphics Cookbook by Winston Chang Reference # Result of heads or tails in x = axis, Counts as y axis, diff colours for each player. # Put labels:  ggplot(data = results, aes(x = Result, y = Freq, fill = Player)) +   geom_bar(stat = "identity", position = position_dodge(), alpha = 0.75)  +   ylim(0,800) +   geom_text(aes(label = Freq), fontface = "bold", vjust = 1.5,              position = position_dodge(.9), size = 4) +   labs(x = "\n Coin Flip Outcome", y = "Frequency\n", title = "\n Coin Flip Results \n") +   theme(plot.title = element_text(hjust = 0.5),          axis.title.x = element_text(face="bold", colour="red", size = 12),         axis.title.y = element_text(face="bold", colour="red", size = 12),         legend.title = element_text(face="bold", size = 10))          

Using facet_grid()

In the above plot, we have four bars in one graph where the Player A bars are beside the Player B bars. Adding facet_grid(. ~Player) will turn this bar graph plot in a way where there is one bar graph plot for Player A and another bar graph plot for player B right beside it.

            # Facet_grid:  ggplot(data = results, aes(x = Result, y = Freq)) +   geom_bar(stat = "identity", alpha = 0.7) +   facet_grid(. ~Player)  +   ylim(0,800) +   geom_text(aes(label = Freq), fontface = "bold", vjust = 1.5, colour = "white", size = 4) +   labs(x = "\n Coin Flip Outcome", y = "Frequency\n", title = "\n Coin Flip Results \n") +   theme(plot.title = element_text(hjust = 0.5),          axis.title.x = element_text(face="bold", colour="darkgreen", size = 12),         axis.title.y = element_text(face="bold", colour="darkgreen", size = 12),         legend.title = element_text(face="bold", size = 10),         strip.background = element_rect(fill="lightblue", colour="black", size=1),         strip.text = element_text(face="bold", size=rel(1.2)))          

Note that the position_dodge() options were removed as they are not needed. To highlight the Player A and Player B font text and the bars, the strip.background and strip.text options are used.

gaineydening.blogspot.com

Source: https://dk81.github.io/dkmathstats_site/rvisual-sidebyside-bar.html

Share this post