numerical data required

[Maxime12]

Hi ! I have this data, but when I want to calculate the median of "Score.on.Alumni" or any other column it says to me: "Error in median.default(rows$Score.on.Alumni): numerical data needed
Thank you in advance!

[siddharthprabhu]

The data type of those columns must be non-numeric. Try median(as.numeric(rangs$Total.Score)).

[Maxime12]

Thanks for the reply !
Here, in my answer sheet we used median(rangs$Score.on.Alumni) and they get "28.8". (with my console I get an error)
When I tried with yours median(as.numeric((rangs$Score.on.Alumni)), I got "15.5".
Do you know why ?

[siddharthprabhu]

Hard to say without seeing the data. Can you show us the complete contents of the vector rangs$Score.on.Alumni?

Also, please familiarize yourself with our homework policy since this appears to be an assignment.

FAQ: Homework Policy

Can I ask questions from a course I am taking here? General questions are always welcome! Please do ask general questions about things like: How to use R How to use the RStudio IDE or RStudio Cloud How to work with tidyverse packages Where to find resources to help you learn or solve problems Specific questions can be OK, if you follow these rules: Never copy-paste instructions from an assignment (even for online courses). Explicitly mention the course you are taking and use the #homework tag. Ask your question as a reproducible example (reprex), preferably prepared for posting using the reprex package. Want to maximize your chances of getting the help you need? Keep readin…

[siddharthprabhu]

Please do not post screenshots. It's much easier if you generate the code required to reproduce the data. For example, try running dput(rangs$Score.on.Alumni)) and copying the output here.

Also, as I mentioned before, the data type of your column is definitely not numeric even though it appears to contain numbers.

[Maxime12]

Sorry I deleted my screenshots, I get this exactly
structure(c(2L, 25L, 33L, 36L, 34L, 29L, 35L, 31L, 32L, 30L,
28L, 26L, 13L, 6L, 17L, 15L, 24L, 1L, 27L, 18L, 24L, 20L, 7L,
14L, 16L, 23L, 21L, 11L, 8L, 19L, 9L, 7L, 18L, 5L, 1L, 7L, 12L,
8L, 22L, 10L, 7L, 16L, 3L, 1L, 1L, 16L, 4L, 13L, 11L, 1L), .Label = c("0",
"100", "13,2", "14,4", "15,6", "16,6", "19,5", "20,4", "21,2",
"22,8", "23,5", "24,3", "25,6", "26,3", "27", "28,8", "33,3",
"33,8", "35,8", "37,2", "37,7", "38,4", "39", "40,3", "42", "43,6",
"48,1", "50,9", "55,5", "60,3", "62,3", "70,8", "72,5", "74,6",
"76", "93,6"), class = "factor")

[siddharthprabhu]

EDIT: My explanation here is incorrect. Please see andresrcs' post below.

Thank you. As you can see from the output, the variable's data type is factor. You can cast it to numeric and make the change permanent as follows:

rangs$Score.on.Alumni <- as.numeric(rangs$Score.on.Alumni)

As for why your answer sheet gives a value of 28.8 for the median, I cannot say. The median value of this set of numbers is 15.5.

[andresrcs]

@siddharthprabhu

The median value of this set of numbers is 15.5.

Actually, using as.numeric() on a factor variable would give you the value of the levels, not the actual numbers, there are many ways of solving this but I like to use readr::parse_number()

library(readr)

rangs <- data.frame(
  Score.on.Alumni = as.factor(c("100","42",
                                "72,5","93,6","74,6","55,5","76","62,3",
                                "70,8","60,3","50,9","43,6","25,6","16,6",
                                "33,3","27","40,3","0","48,1","33,8","40,3",
                                "37,2","19,5","26,3","28,8","39","37,7",
                                "23,5","20,4","35,8","21,2","19,5","33,8",
                                "15,6","0","19,5","24,3","20,4","38,4",
                                "22,8","19,5","28,8","13,2","0","0","28,8",
                                "14,4","25,6","23,5","0"))
)

median(parse_number(as.character(rangs$Score.on.Alumni),
                  locale = locale(decimal_mark = ",")))
#> [1] 28.8

^{Created on 2020-04-26 by the reprex package (v0.3.0)}

[siddharthprabhu]

Ah, I hadn't realized that! Thank you for pointing it out.

[Maxime12]

Thanks all for the reply ! It's helped me a lot.