How to create a (100%) complex histogram in R?

Question

My dataset:

I have data in the following format (here, imported from a CSV file). You can find an example dataset like CSV here .

PAIR PREFERENCE 1 5 1 3 1 2 2 4 2 1 2 3 

... etc. There are 19 pairs in total, and PREFERENCE ranges from 1 to 5 as discrete values.

---

What I'm trying to achieve:

I need a histogram with accumulation, for example. 100% column for each pair, indicating the distribution of PREFERENCE values.

Something similar to "100% stack columns" in Excel or (although not quite the same thing, the so-called "mosaic plot"):

---

What I tried:

I realized that this would be easiest with ggplot2 , but I don’t even know where to start. I know that I can create a simple histogram with something like:

 ggplot(d, aes(x=factor(PAIR), y=factor(PREFERENCE))) + geom_bar(position="fill") 

... this, however, is not very far. So I tried this, and it brought me closer to what I'm trying to achieve, but it still uses the PREFERENCE count, I suppose? Note that ylab is "count" here, and values ​​range from 19.

 qplot(factor(PAIR), data=d, geom="bar", fill=factor(PREFERENCE_FIXED)) 

Results in:

• So what do I need to do to overlay the columns on the histogram?
• Or do they really do it already?
• If so, what do I need to change to get the correct labels (for example, have percentages instead of "count")?

_{By the way, this is not related to this issue and is only slightly related to it (i.e. probably the same idea, but not continuous values, are instead grouped into columns).}

Answer 1

Maybe you want something like this:

 ggplot() + geom_bar(data = dat, aes(x = factor(PAIR),fill = factor(PREFERENCE)), position = "fill") 

where i read your data in dat . This outputs something like this:

The y icon is still "counted", but you can change it manually by adding:

 + scale_x_discrete("Pairs") + scale_y_continuous("Votes")