1
votes

I am trying to create a bar graph. When I set the limits as (0,7), the bars appear. However, I would like the lower limit to be 1, not 0. When I set the lower limit to 1, the bars do not appear. I get the following error message:

Removed 8 rows containing missing values (geom_bar).

It doesn't matter how I set the limits. I have used both of the following options:

ylim(1, 7)

scale_y_continuous(limits = c(1, 7))

Does anyone know how I can fix this?

I'd like a graph that looks like this, but with 1 as the lower y-axis label, which would mean all the bars would be shifted down by 1.

enter image description here

Here's the full code for the graph:

full %>%
 ggplot(aes(x = order, y = mean)) + 
 geom_bar(stat = "identity", fill = "003900", width = 0.5, position = position_dodge()) +
  geom_errorbar(aes(ymin = mean - se, ymax = mean + se), width = .2, position = position_dodge(.9)) +
  geom_text(aes(label = round(mean, digits =1)), position = position_dodge(width=1.0), vjust = -4.0, size = 3) +
  #facet_wrap(~names) +
  labs(title = "Behavioral intentions in response to each message") +
 # ylim(0, 7) + 
  scale_y_continuous(limits = c(1, 7)) + 
  theme(axis.text = element_text(size = 7)) + 
  xlab("Message") + 
  ylab("Behavioral intentions")

Here's reproducible data:

structure(list(message = c("a", "e", "h", "m", "convince_animals", 
"convince_environment", "convince_health", "convince_money"), 
    mean = c(3.1038961038961, 3.21052631578947, 3.56, 2.7972972972973, 
    4.19512195121951, 4.18536585365854, 5.65365853658537, 4.93658536585366
    ), se = c(0.208814981196227, 0.204609846510406, 0.220760356801522, 
    0.20542415978608, 0.121188432228325, 0.11075110910238, 0.0896896391724367, 
    0.120394657272105), type = c("Behavioral Intentions", "Behavioral Intentions", 
    "Behavioral Intentions", "Behavioral Intentions", "Expected Behavior", 
    "Expected Behavior", "Expected Behavior", "Expected Behavior"
    ), names = c("Animals", "Environment", "Health", "Money", 
    "Animals", "Environment", "Health", "Money"), order = c(1, 
    3, 5, 7, 2, 4, 6, 8)), row.names = c(NA, -8L), class = c("tbl_df", 
"tbl", "data.frame"))
1

1 Answers

1
votes

Looks like you're losing data by setting limits and it's screwing up your plot. You can use coord_cartesian() instead of ylim() to 'zoom in' on your data; see https://stackoverflow.com/a/25685952/12957340 and/or page 160 of the ggplot2 book for further info.

full %>%
  ggplot(aes(x = order, y = mean)) + 
  geom_bar(stat = "identity", fill = "003900", width = 0.5, position = position_dodge()) +
  geom_errorbar(aes(ymin = mean - se, ymax = mean + se), width = .2, position = position_dodge(.9)) +
  geom_text(aes(label = round(mean, digits =1)), position = position_dodge(width=1.0), vjust = -4.0, size = 3) +
  #facet_wrap(~names) +
  labs(title = "Behavioral intentions in response to each message") +
  coord_cartesian(ylim = c(1, 7)) + 
  theme(axis.text = element_text(size = 7)) + 
  xlab("Message") + 
  ylab("Behavioral intentions")

example.png