CODEWITHSUNDEEP
rggplot2
Gautam Mainkar
Gautam Mainkar

Reputation: 576

Why is my graph's Y-axis not in the right order?

I am trying to plot a graph for a data frame that looks like this:

year week cases
2003    1     0
2003    2     0
2003    3    12
2003    4    23
2003    5    12
2003    6    16
2003    7    20
2003    8    13
2003    9     0
2003   10     0
2003   11    21
2003   12   133
2003   13     9
2003   14    22

Carrying data for 52 weeks running from 2003-2012.

Here's what running dput(head(df,20) gives me:

structure(list(year = c(2003L, 2003L, 2003L, 2003L, 2003L, 2003L, 
2003L, 2003L, 2003L, 2003L, 2003L, 2003L, 2003L, 2003L, 2003L, 
2003L, 2003L, 2003L, 2003L, 2003L), week = 1:20, cases = c(2, 
2, 26, 146, 26, 70, 115, 37, 2, 2, 124, 41, 245, 135, 146, 163, 
26, 26, 92, 92)), .Names = c("year", "week", "cases"), row.names 1925:1944, class = "data.frame")

I want my Y-axis to be simply the range of the variable 'cases', and the X-axis to run from week 1 through 52. I want to plot every year's data points in a different color.

Here's my ggplot2 code:

ggplot(df, aes(x=week, y=cases, col=year)) + geom_point()

This is the graph it's generating:

graph

Why is this happening? I see no reason why my Y-axis shouldn't just be the range of 'cases' in ascending order.

Upvotes: 1

Views: 7051

Answers (1)

Cath
Cath

Reputation: 24074

To sum up what was said in the comments :

Your y-axis is indeed sorted but according to the character values (or rather the factor levels, as your variable was imported as factor) and not the numeric ones (so 1, 10, 11, ..., 2, 20, ...)

There is 2 problems that need to be solved:
the first one is that you have to understand why the variable wasn't imported as numeric. You probably have a "strange" value (like 1,2 for example, ie a comma instead of a point as decimal separator)
The second one is you need numeric values to plot your data correctly. For that, you can transform your factor with df$cases <- as.numeric(as.character(df$cases)). Note that the strange value(s) will be converted to NAs, you may not want that.

Just a final note, if you don't want your character variables to be imported as factors, you can use the parameter stringsAsFactors=FALSE in the import step.

Upvotes: 4

Related Questions