CODEWITHSUNDEEP
rggplot2plotreshape
DYtto
DYtto

Reputation: 11

Lineplot by group over time

I am trying to build a line plot showing the change of a certain variable over time by group. My data has values from a single person stored in each row and I am struggling to plot the mean and deviation of the said value over time. Here goes an example of a dataset.

ID Group Value Time
1 MDD 14.2 8
1 MDD 15.3 9
2 NO-MDD 11.2 8
2 NO-MDD 12.3 9
2 NO-MDD 12.8 5
3 NO-MDD 9.1 3
3 NO-MDD 11.3 1
4 MDD 10.8 5
4 MDD 14.6 18
4 MDD 15.3 37

The time variable stores values from 0 to 37 but not for each participant. It is a large dataset though, so there are enough measurements to calculate SD for each time point. My main problem is the way the data is arranged. I was thinking on reshaping the data but not quite sure if that would be helpful. Here is the expected output drawn by hand, sorry!

enter image description here

Thank you in advance for any suggestions you can provide!

Upvotes: 1

Views: 64

Answers (1)

stefan
stefan

Reputation: 123783

One option to achieve your desired result with ggplot2 would be to use multiple stat_summary layers to add the mean line and points and the errorbars.

As you example data did not allow to compute a standard deviation by time and group I use some fake random example data:

set.seed(123)

df <- data.frame(
  Time = sample(1:37, 100, replace = TRUE),
  Group = sample(c("MDD", "NO-MDD"), 100, replace = TRUE),
  Value = rnorm(100)
)

df$Value <- df$Value + ifelse(df$Group == "MDD", 2, 0)

library(ggplot2)

ggplot(df, aes(Time, Value, color = Group)) +
  stat_summary(fun = mean, geom = "line") +
  stat_summary(fun.data = mean_se, geom = "errorbar") +
  stat_summary(fun = mean, geom = "point")

Upvotes: 1

Related Questions