R help - Dec 2011 - convert variable types when creating data frame from cor.test results

Dear list,

The below dataset and code creates a new dataset with the results from
the function cor.test being performed on each individual
('Individual_ID') from my original dataset. How do I convert each
variable from the cor.test results to a numeric data type, as it is
passed into the new dataframe? For example, 'estimate',
'p.value', and
'conf.int' should be numeric not character. Second, I would like
variable 'conf.int' to be two variables, 'lowlim' and
'uplim'.

Many Cheers,
Keith

## Create sample dataset
WW_Wing_SI <-
structure(list(Individual_ID = c("WW_08A_02", "WW_08A_02",
"WW_08A_02",
"WW_08A_02", "WW_08A_02", "WW_08A_02",
"WW_08A_02", "WW_08A_02",
"WW_08A_02", "WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03"
), FeatherPosition = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4,
5, 6, 7, 8, 9), Delta13C = c(-18.67, -19.16, -20.38, -20.96,
-21.61, -21.65, -21.31, -20.8, -21.28, -20.06, -20.3, -21.21,
-22.9, -22.87, -21.13, -20.68, -20.58, -20.69)), .Names =
c("Individual_ID",
"FeatherPosition", "Delta13C"), row.names = c(NA, 18L),
class = "data.frame")

# split data frame according the the individual IDs
wing.list <- split(WW_Wing_SI, WW_Wing_SI$Individual_ID)

# apply cor.test() with extract to each element of the list
test <- as.data.frame(t(sapply(wing.list, function(temp)
                                cor.test(temp$Delta13C, temp$FeatherPosition,
                               
method="pearson")[c("estimate",
"p.value", "conf.int")])))

*******************************************************************************************
Keith Larson, PhD Student
Evolutionary Ecology, Lund University
S?lvegatan 37
223 62 Lund Sweden
Phone: +46 (0)46 2229014 Mobile: +46 (0)73 0465016 Fax: +46 (0)46 2224716
Skype: sternacaspia FB: keith.w.larson at gmail.com

Hi
> 
> Dear list,
> 
> The below dataset and code creates a new dataset with the results from
> the function cor.test being performed on each individual
> ('Individual_ID') from my original dataset. How do I convert each
> variable from the cor.test results to a numeric data type, as it is
> passed into the new dataframe? For example, 'estimate',
'p.value', and
> 'conf.int' should be numeric not character. Second, I would like
> variable 'conf.int' to be two variables, 'lowlim' and
'uplim'.
> 
> Many Cheers,
> Keith
> 
> ## Create sample dataset
> WW_Wing_SI <-
> structure(list(Individual_ID = c("WW_08A_02",
"WW_08A_02", "WW_08A_02",
> "WW_08A_02", "WW_08A_02", "WW_08A_02",
"WW_08A_02", "WW_08A_02",
> "WW_08A_02", "WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03",
> "WW_08A_03", "WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03"
> ), FeatherPosition = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4,
> 5, 6, 7, 8, 9), Delta13C = c(-18.67, -19.16, -20.38, -20.96,
> -21.61, -21.65, -21.31, -20.8, -21.28, -20.06, -20.3, -21.21,
> -22.9, -22.87, -21.13, -20.68, -20.58, -20.69)), .Names = 
c("Individual_ID",
> "FeatherPosition", "Delta13C"), row.names = c(NA, 18L),
class =
"data.frame")
> 
> # split data frame according the the individual IDs
> wing.list <- split(WW_Wing_SI, WW_Wing_SI$Individual_ID)
> 
> # apply cor.test() with extract to each element of the list
> test <- as.data.frame(t(sapply(wing.list, function(temp)
>                                 cor.test(temp$Delta13C, 
temp$FeatherPosition,
>                                
method="pearson")[c("estimate",
> "p.value", "conf.int")])))
The problem is that transposing a sapply object transforms it to matrix 
and matrix can have only one type of values.

Not very sophisticated but this

result <- unlist(sapply(wing.list, function(temp) cor.test(temp$Delta13C, 
temp$FeatherPosition, method="pearson"))[c(3,4,9),])

gives you a numeric vector of desired values. You can change it to numeric 
matrix

dim(result)<-c(4,2)

Give the propper name to rows and columns and/or transpose according to 
your wish.

Regards
Petr


> 
> 
*******************************************************************************************
> Keith Larson, PhD Student
> Evolutionary Ecology, Lund University
> S?lvegatan 37
> 223 62 Lund Sweden
> Phone: +46 (0)46 2229014 Mobile: +46 (0)73 0465016 Fax: +46 (0)46 
2224716
> Skype: sternacaspia FB: keith.w.larson at gmail.com
> 
> ______________________________________________
> 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.

Keith Larson wrote on 12/28/2011 04:02:39 AM:
> Dear list,
> 
> The below dataset and code creates a new dataset with the results from
> the function cor.test being performed on each individual
> ('Individual_ID') from my original dataset. How do I convert each
> variable from the cor.test results to a numeric data type, as it is
> passed into the new dataframe? For example, 'estimate',
'p.value', and
> 'conf.int' should be numeric not character. Second, I would like
> variable 'conf.int' to be two variables, 'lowlim' and
'uplim'.
> 
> Many Cheers,
> Keith
> 
> ## Create sample dataset
> WW_Wing_SI <-
> structure(list(Individual_ID = c("WW_08A_02",
"WW_08A_02", "WW_08A_02",
> "WW_08A_02", "WW_08A_02", "WW_08A_02",
"WW_08A_02", "WW_08A_02",
> "WW_08A_02", "WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03",
> "WW_08A_03", "WW_08A_03", "WW_08A_03",
"WW_08A_03", "WW_08A_03"
> ), FeatherPosition = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 4,
> 5, 6, 7, 8, 9), Delta13C = c(-18.67, -19.16, -20.38, -20.96,
> -21.61, -21.65, -21.31, -20.8, -21.28, -20.06, -20.3, -21.21,
> -22.9, -22.87, -21.13, -20.68, -20.58, -20.69)), .Names = 
c("Individual_ID",
> "FeatherPosition", "Delta13C"), row.names = c(NA, 18L),
class =
"data.frame")
> 
> # split data frame according the the individual IDs
> wing.list <- split(WW_Wing_SI, WW_Wing_SI$Individual_ID)
> 
> # apply cor.test() with extract to each element of the list
> test <- as.data.frame(t(sapply(wing.list, function(temp)
>                                 cor.test(temp$Delta13C, 
temp$FeatherPosition,
>                                
method="pearson")[c("estimate",
> "p.value", "conf.int")])))
> 
> 
*******************************************************************************************
> Keith Larson, PhD Student
> Evolutionary Ecology, Lund University
> Sölvegatan 37
> 223 62 Lund Sweden
> Phone: +46 (0)46 2229014 Mobile: +46 (0)73 0465016 Fax: +46 (0)46 
2224716
> Skype: sternacaspia FB: keith.w.larson@gmail.com

If you unlist() the call to cor.test, you will end up with numeric data. 
You will also end up with two variables for the confidence intervals.

test <- data.frame(t(sapply(wing.list, function(temp) 
        unlist(cor.test(temp$Delta13C, temp$FeatherPosition, 
        method="pearson")[c("estimate", "p.value",
"conf.int")]))))

test
          estimate.cor    p.value  conf.int1  conf.int2
WW_08A_02  -0.76706752 0.01585509 -0.9481677 -0.2098478
WW_08A_03  -0.02320767 0.95274294 -0.6768966  0.6509469


Jean

	[[alternative HTML version deleted]]