Summary in R for frequency tables?

Question

I have a set of user recommandations

review=matrix(c(5:1,10,2,1,1,2), nrow=5, ncol=2, dimnames=list(NULL,c("Star","Votes")))

and wanted to use summary(review) to show basic properties mean, median, quartiles and min max.

But it gives back the summary of both columns. I refrain from using data.frame because the factors 'Star' are ordered. How can I tell R that Star is a ~~ordered list of factors~~ numeric score and votes are their frequency?

I just saw the tag frequency-analysis. Are you looking for table()? Or contingency tables? — Matt Bannert, Feb 05 '11 at 13:34
i tried table didn't work. I somhow need to do mean, median, quartiles. and i dont want to do it by hand. thats a minimum I expect from a statistical framework — Roland Kofler, Feb 05 '11 at 13:51
Note that the weighted mean of an _ordered factor_ is not defined, because the whole point of not calling it numeric is that the between-level intervals are not known. You have to assign numeric scores to take means. — Aniko, Feb 05 '11 at 15:37
Thanks Aniko, I understand your argument. To name the beast. I want to do it with Amazon reviews. I think they see ratings as cardinal since they do compute the mean — Roland Kofler, Feb 05 '11 at 15:40
What you are really looking for is the mean, and quantiles of the Stars *weighted* by the Votes. And Al3xa's approach is a good way to get this. There is really no need to make the Star column a factor. — Prasad Chalasani, Feb 05 '11 at 15:56

score 5 · Accepted Answer · answered Feb 05 '11 at 14:30

5

I'm not exactly sure what you mean by taking the mean in general if Star is supposed to be an ordered factor. However, in the example you give where Star is actually a set of numeric values, you can use the following:

library(Hmisc)

R> review=matrix(c(5:1,10,2,1,1,2), nrow=5, ncol=2, dimnames=list(NULL,c("Star","Votes")))

R> wtd.mean(review[, 1], weights = review[, 2])
[1] 4.0625

R> wtd.quantile(review[, 1], weights = review[, 2])
  0%  25%  50%  75% 100% 
1.00 3.75 5.00 5.00 5.00

answered Feb 05 '11 at 14:30

brentonk

1,308
1
13
14

I really like this approach better than al3xa's because i don't need to "decompress" the scores with their weights. But its a shame that is not in the base package. this certainly is daily bread for statistics – Roland Kofler Feb 05 '11 at 16:06
3

Use `weighted.mean` in base R. – hadley Feb 05 '11 at 17:38
thats great, i would give you the "right answer" if its was not a comment – Roland Kofler Feb 05 '11 at 20:03

aL3xa · Answer 2 · 2011-02-05T15:55:25.230

2

I don't understand what's the problem. Why shouldn't you use data.frame?

rv <- data.frame(star = ordered(review[, 1]), votes = review[, 2])

You should convert your data.frame to vector:

( vts <- with(rv, rep(star, votes)) )
 [1] 5 5 5 5 5 5 5 5 5 5 4 4 3 2 1 1
Levels: 1 < 2 < 3 < 4 < 5

Then do the summary... I just don't know what kind of summary, since summary will bring you back to the start. O_o

summary(vts)
 1  2  3  4  5 
 2  1  1  2 10

EDIT (on @Prasad's suggestion)

Since vts is an ordered factor, you should convert it to numeric, hence calculate the summary (at this moment I will disregard the background statistical issues):

nvts <- as.numeric(levels(vts)[vts])  ## numeric conversion
summary(nvts)  ## "ordinary" summary
fivenum(nvts)  ## Tukey's five number summary

edited Feb 05 '11 at 15:55

answered Feb 05 '11 at 13:32

aL3xa

35,415
18
79
112

I want to see mean, median, quartiles and min/max, this is what I mean by summary – Roland Kofler Feb 05 '11 at 13:48
`summary(as.numeric(vts))` or even better `summary(as.numeric(levels(vts)[vts]))`. The latter approach is recommended. – aL3xa Feb 05 '11 at 14:27
1

+1 I would recommend putting this in the main answer just to make it clear. – Prasad Chalasani Feb 05 '11 at 15:25
Damn, my mental registry betrayed me this time... I'll try to recall where did I read about this one. Good question, though. =) – aL3xa Feb 05 '11 at 16:24
Conversion to vector is not possible if frequencies are huge! – meolic Mar 28 '18 at 18:19

score 0 · Answer 3 · answered Feb 05 '11 at 14:15

Just to clarify -- when you say you would like "mean, median, quartiles and min/max", you're talking in terms of number of stars? e.g mean = 4.062 stars? Then using aL3xa's code, would something like summary(as.numeric(as.character(vts))) be what you want?

Summary in R for frequency tables?

3 Answers3