[R] Problem with Median

[David Winsemius]

On May 7, 2012, at 5:16 PM, Suhaila Haji Mohd Hussin wrote:

>
> Hello.
> Now the median is solved but then I'm still figuring out how to put  
> the updated column back to replace the original column of the whole  
> data. I'll show you what I meant:
> Continuing from the previous commands you guys helped out I  
> continued as followed:
> Original Data: http://i1165.photobucket.com/albums/q585/halfpirate/data.jpg
>
> Before column a: http://i1165.photobucket.com/albums/q585/halfpirate/1.jpg

That fine for sending pictures to the family but NOT for communicating  
R code.
>
> 1) a.AC [is.na(a.AC)] <- median(a.AC$a, na.rm= TRUE)
> 2) a.ACAfter column a: http://i1165.photobucket.com/albums/q585/halfpirate/2.jpg
> 3) data$a <- a.ACGives me error that the original data has 6 rows  
> yet the one I'm replacing is 1 row.
> I understand the error and I don't want to completely replace the  
> column if I really want to.

You were not attempting to replace the entire  column, but you were  
trying to replace 6 entries. So why not... :

a.AC [is.na(a.AC)] <- rep( median(a.AC$a, na.rm= TRUE) , 6)


> Basically I'm expecting the result to be like this. After you  
> compare the original data, the final answer should be like this http://i1165.photobucket.com/albums/q585/halfpirate/3.jpg

What part of "I can't do anything with your jpeg to answer your  
questions about your dataframe" was difficult to understand? We do NOT  
want pictures from which we need to repeat data entry that you have  
already done.  The first virtue of a programmer being laziness.  
Console output is also very ambiguous.  Learn to use str() and dput()  
or dump().

?dput
?dump

And run the examples. please.

-- 
David.


> Please help.Suhaila
> From: bell_beauty12 at hotmail.com
> To: jholtman at gmail.com
> CC: r-help at r-project.org
> Subject: RE: [R] Problem with Median
> Date: Tue, 8 May 2012 07:22:19 +1200
>
> Thank you so much!
> Suhaila.
>
>> Date: Mon, 7 May 2012 15:08:47 -0400
>> Subject: Re: [R] Problem with Median
>> From: jholtman at gmail.com
>> To: bell_beauty12 at hotmail.com
>>
>> Your problem is that a.AC is a dataframe:
>>
>>
>>> x <- read.table(text = "   a     b     c      class
>> + 1   12   0      90     A-B
>> + 2   3     97    11     A-B
>> + 3   78   NA    123   A-C
>> + 4   NA   NA    12    A-C
>> + 5   8     33     2     A-B
>> + 6   12   NA     0     A-D", header = TRUE)
>>> a.AC <- subset(x, class == "A-C", select = a)
>>> # same error
>>> median(a.AC)
>> Error in median.default(a.AC) : need numeric data
>>
>>> # now look a the structure of a.AC (its a dataframe)
>>> str(a.AC)
>> 'data.frame':   2 obs. of  1 variable:
>> $ a: int  78 NA
>>> # now do it right
>>> median(a.AC$a)
>> [1] NA
>>> median(a.AC$a, na.rm = TRUE)
>> [1] 78
>>
>>
>>
>> On Mon, May 7, 2012 at 2:52 PM, Suhaila Haji Mohd Hussin
>> <bell_beauty12 at hotmail.com> wrote:
>>>
>>> Hello.
>>> I'm trying to compute median for a filtered column based on other  
>>> column but there was something wrong. I'll show how I did step by  
>>> step.
>>> Here's the data:
>>>    a     b     c      class
>>>
>>> 1   12   0      90     A-B2   3     97    11     A-B3   78   NA     
>>> 123   A-C4   NA   NA    12    A-C5   8     33     2     A-B6    
>>> 12   NA     0     A-D
>>> On the command I typed:
>>> 1) data = read.csv("data.csv")
>>>
>>> 2) a.AC <- subset(data, class == "A-C", select = a)
>>> 3) median(a.AC)Error in median.default(a.AC) : need numeric data
>>> 4) is.numeric(a.AC)FALSE
>>> 5) as.numeric(a.AC)Error: (list) object cannot be coerced to type  
>>> 'double'
>>> How can I fix this? Please help.
>>> Cheers,Suhaila
>>>       [[alternative HTML version deleted]]
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>
>>
>>
>> -- 
>> Jim Holtman
>> Data Munger Guru
>>
>> What is the problem that you are trying to solve?
>> Tell me what you want to do, not how you want to do it.
>
>
> I might be silly but if I was going to type in dput() then how  
> should I send the data over here?
> Instead, I've just uploaded the image online, you can access it via  
> the link below.
> http://i1165.photobucket.com/albums/q585/halfpirate/data.jpg
>
>> Date: Mon, 7 May 2012 14:55:24 -0400
>> Subject: Re: [R] Problem with Median
>> From: sarah.goslee at gmail.com
>> To: bell_beauty12 at hotmail.com
>> CC: r-help at r-project.org
>>
>> Please use dput() to give us your data (eg dput(data) ) rather than
>> simply pasting it in.
>>
>> Sarah
>>
>> On Mon, May 7, 2012 at 2:52 PM, Suhaila Haji Mohd Hussin
>> <bell_beauty12 at hotmail.com> wrote:
>>>
>>> Hello.
>>> I'm trying to compute median for a filtered column based on other  
>>> column but there was something wrong. I'll show how I did step by  
>>> step.
>>> Here's the data:
>>>    a     b     c      class
>>>
>>> 1   12   0      90     A-B2   3     97    11     A-B3   78   NA     
>>> 123   A-C4   NA   NA    12    A-C5   8     33     2     A-B6    
>>> 12   NA     0     A-D
>>> On the command I typed:
>>> 1) data = read.csv("data.csv")
>>>
>>> 2) a.AC <- subset(data, class == "A-C", select = a)
>>> 3) median(a.AC)Error in median.default(a.AC) : need numeric data
>>> 4) is.numeric(a.AC)FALSE
>>> 5) as.numeric(a.AC)Error: (list) object cannot be coerced to type  
>>> 'double'
>>> How can I fix this? Please help.
>>> Cheers,Suhaila
>>
>>
>> -- 
>> Sarah Goslee
>> http://www.functionaldiversity.org 		 	   		
>
> 		 	   		
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius, MD
West Hartford, CT