Trying to make a bar chart with uniform column widths that plots the count of each "User Experi-CodePudding

The goal is to make a bar chart from the below data. Data should be clustered into each Customer Journey Area by User Experience Framework. For example, the "Browsing" Journey Area should have an individual bar for each User Experience Framework count. For example, Browsing = Execution (5), Systems (3). Where this gets tricky is that we want to have 0 in the middle and the "Positive hits" going to the right and the "negative hits" going to the left. All bars should be uniform in size.

I can't quite make this happen. Below is my head(df):

> head(df,30)
          Journey.Area Experience.Framework Postive.or.Negative
1           Installing     People/Associate                   1
2        Using Product            Execution                  -1
3           Installing     People/Associate                   1
4             Delivery            Execution                  -1
5           Installing     People/Associate                  -1
6             Delivery     People/Associate                   1
7           Installing            Execution                  -1
8        Using Product            Execution                  -1
9             Browsing     People/Associate                  -1
10            Browsing     People/Associate                  -1
11            Browsing     People/Associate                   1
12              Buying              Systems                  -1
13            Delivery            Execution                  -1
14            Delivery     People/Associate                   1
15          Installing            Execution                  -1
16 Deciding How to Buy       Process/Policy                   1
17          Installing     People/Associate                  -1
18          Installing     People/Associate                   1
19            Delivery            Execution                  -1
20              Buying       Process/Policy                   1
21            Delivery            Execution                  -1
22          Installing            Execution                   1
23            Browsing     People/Associate                   1
24          Installing            Execution                  -1
25            Delivery            Execution                  -1
26          Installing            Execution                  -1
27          Installing            Execution                  -1
28 Deciding How to Buy       Process/Policy                  -1
29          Installing            Execution                  -1
30            Browsing              Systems                  -1

Here is my dput(df):

> dput(df)
structure(list(Journey.Area = c("Installing", "Using Product", 
"Installing", "Delivery", "Installing", "Delivery", "Installing", 
"Using Product", "Browsing", "Browsing", "Browsing", "Buying", 
"Delivery", "Delivery", "Installing", "Deciding How to Buy", 
"Installing", "Installing", "Delivery", "Buying", "Delivery", 
"Installing", "Browsing", "Installing", "Delivery", "Installing", 
"Installing", "Deciding How to Buy", "Installing", "Browsing", 
"Delivery", "Installing", "Browsing", "Installing", "Browsing", 
"Deciding How to Buy", "Installing", "Anticipating", "Delivery", 
"Delivery", "Deciding How to Buy", "Installing", "Using Product", 
"Installing", "Delivery", "Installing", "Installing", "Deciding How to Buy", 
"Delivery", "Delivery", "Delivery", "Browsing", "Using Product", 
"Deciding How to Buy", "Delivery", "Installing", "Installing", 
"Deciding How to Buy", "Installing", "Installing", "Anticipating", 
"Installing", "Deciding How to Buy", "Deciding How to Buy", "Installing", 
"Browsing", "Delivery", "Pickup", "Anticipating", "Deciding How to Buy", 
"Using Product", "Installing", "Anticipating", "Deciding How to Buy", 
"Browsing", "Deciding How to Buy", "Buying", "Delivery", "Installing", 
"Installing", "Installing", "Deciding How to Buy", "Anticipating", 
"Delivery", "Installing", "Anticipating", "Delivery", "Delivery", 
"Delivery", "Anticipating", "Browsing", "Deciding How to Buy", 
"Deciding How to Buy", "Anticipating", "Deciding How to Buy", 
"Delivery", "Delivery", "Deciding How to Buy", "Deciding How to Buy", 
"Deciding How to Buy"), Experience.Framework = c("People/Associate", 
"Execution", "People/Associate", "Execution", "People/Associate", 
"People/Associate", "Execution", "Execution", "People/Associate", 
"People/Associate", "People/Associate", "Systems", "Execution", 
"People/Associate", "Execution", "Process/Policy", "People/Associate", 
"People/Associate", "Execution", "Process/Policy", "Execution", 
"Execution", "People/Associate", "Execution", "Execution", "Execution", 
"Execution", "Process/Policy", "Execution", "Systems", "Execution", 
"Execution", "People/Associate", "Execution", "Execution", "Process/Policy", 
"Execution", "Systems", "Execution", "Process/Policy", "Process/Policy", 
"Execution", "Execution", "Execution", "Execution", "People/Associate", 
"Execution", "Execution", "Execution", "Execution", "Execution", 
"People/Associate", "Execution", "Process/Policy", "Execution", 
"People/Associate", "People/Associate", "People/Associate", "Execution", 
"People/Associate", "Process/Policy", "Execution", "Execution", 
"Execution", "Execution", "Systems", "Execution", "Execution", 
"Execution", "Systems", "Execution", "People/Associate", "Execution", 
"Execution", "People/Associate", "People/Associate", "Systems", 
"Execution", "Execution", "People/Associate", "Execution", "People/Associate", 
"Systems", "Execution", "Execution", "Execution", "Execution", 
"Execution", "Execution", "Execution", "Execution", "People/Associate", 
"People/Associate", "Execution", "Systems", "Execution", "Execution", 
"People/Associate", "People/Associate", "Execution"), Postive.or.Negative = c(1L, 
-1L, 1L, -1L, -1L, 1L, -1L, -1L, -1L, -1L, 1L, -1L, -1L, 1L, 
-1L, 1L, -1L, 1L, -1L, 1L, -1L, 1L, 1L, -1L, -1L, -1L, -1L, -1L, 
-1L, -1L, -1L, -1L, 1L, -1L, 1L, -1L, -1L, -1L, -1L, 1L, 1L, 
-1L, -1L, -1L, 1L, -1L, -1L, 1L, -1L, -1L, -1L, -1L, -1L, -1L, 
-1L, 1L, 1L, 1L, -1L, 1L, -1L, -1L, 1L, 1L, -1L, -1L, -1L, -1L, 
-1L, -1L, 1L, 1L, -1L, 1L, -1L, -1L, -1L, 1L, -1L, -1L, 1L, -1L, 
-1L, -1L, -1L, -1L, -1L, 1L, -1L, -1L, -1L, 1L, -1L, -1L, 1L, 
-1L, -1L, 1L, 1L, 1L)), class = "data.frame", row.names = c(NA, 
-100L))

Again, in the plot all columns should be the same width. Negative counts should go left and Positive counts should go right. Attached is a photo that is close...but you'll notice the bars are different widths.

CodePudding user response：

I think you might want geom_col(position = position_dodge2(preserve = "single")).

EDIT:

and you can count the data first to get single bars for each category:

library(tidyverse)
df %>%
  count(Journey.Area, Experience.Framework, 
        wt = Postive.or.Negative, name = "Positive.or.Negative") %>%
  ggplot(aes(Positive.or.Negative, Journey.Area, fill = Experience.Framework))  
  geom_col(position = position_dodge2(preserve = "single"))

Compare:

data.frame(Journey.Area = c("Installing", rep("Using Product", 3)),
           Experience.Framework = c("A","A","B","C"),
           Positive.or.Negative = c(-5, 4:6)) %>%
  ggplot(aes(Positive.or.Negative, Journey.Area, fill = Experience.Framework))  
  geom_col(position = "dodge")                               # (1)
  geom_col(position = position_dodge())                      # (1)
  geom_col(position = position_dodge2())                     # (2)
  geom_col(position = position_dodge2(preserve = "single"))  # (3)

(1)

(2)

(3)