Variable labels

Never mind what I said about "Clickable". All I meant was I created an
item "definitions" that appears after I load the binary file, and that I
can "click" (don's ask me what I mean by "click") the item in RStudio to
read its contents -- variable definitions.

All I want to know at this pointis, is whether my way of getting the
definitions in the environment "clumsy" and whether there are better
ways to do it. Yes, you mention "attr..." but that is not as simple as
viewing it in RStudio's environment pane. Thank you!

On 2021/5/14 ?? 06:37, PIKAL Petr wrote:

Hm. What do you mean by "clickable".

#I can save any objects to a file
save(mydata,definitions, file="test.R")
rm("mydata", "definitions")

#load them back
load("test.R")

#but it does not make them "clickable". Point and click is something I am familiar with in Excel or similar programs byt not in R.

#objects are back in the environment and one can inspect them by regular way (print, str, head, ...)
mydata
   id age yrmarry
1  1  35       4
2  2  31       6
3  3  21       4
4  4  20       3
5  5  19       7
6  6  24       5
definitions
                var.labels
id          Individual ID
age          Age in Years
yrmarry Years of marriage

If you want definitions to be part of the data file just use attr.

attr(mydata, "var.labels") <- definitions$var.labels

  attributes(mydata)
$names
[1] "id"      "age"     "yrmarry"

$class
[1] "data.frame"

$row.names
[1] 1 2 3 4 5 6

$var.labels
[1] "Individual ID"     "Age in Years"      "Years of marriage"

Cheers
Petr

-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Steven Yen
Sent: Friday, May 14, 2021 11:20 AM
To: Jim Lemon <drjimlemon at gmail.com>
Cc: R-help Mailing List <r-help at r-project.org>
Subject: Re: [R] Variable labels

Thanks to all, for bearing with me.

Now I realize expss may not be what I need. I have now written a self-
runnable, replicable set of codes (listed below). Perhaps that gives an idea of
what I need. Question is, whethet this is the right way to do this (to have a
clickable object to learn about variable
definitions) or whether there are better ways. Thanks!

Steven

rm(list=ls())
n<-6
mydata<-data.frame(id=1:n,
                     age=floor(rnorm(n,25,10)),
                     yrmarry=floor(rnorm(n,5,2))) var.labels<-c(id  = "Individual ID",
                age = "Age in Years",
                yrmarry = "Years of marriage")
definitions<-as.data.frame(var.labels) # declare definitions as a data frame
save.image("c:/temp/a/try1.RData")     # save binary .RData file
rm(list=ls())                          # clean environment
load("c:/temp/a/try1.RData") # now load .RData file and definitions are
clickable
                               # all I need is for user to be able to click
                               # and read the variable definitions

On 2021/5/14 ?? 05:15, Jim Lemon wrote:

Hi Steven,
I just happened to scan Petr's message to you and wondered if you were
looking for something related to the "describe" function in the
prettyR package (and a few others). For instance, if you do this:

library(prettyR)
describe(mtcars)

you get this:

Description of mtcars

Numeric
        mean median      var     sd valid.n
mpg   20.09  19.20    36.32   6.03      32
cyl    6.19   6.00     3.19   1.79      32
disp 230.72 196.30 15360.80 123.94      32
hp   146.69 123.00  4700.87  68.56      32
drat   3.60   3.70     0.29   0.53      32
wt     3.22   3.33     0.96   0.98      32
qsec  17.85  17.71     3.19   1.79      32
vs     0.44   0.00     0.25   0.50      32
am     0.41   0.00     0.25   0.50      32
gear   3.69   4.00     0.54   0.74      32
carb   2.81   2.00     2.61   1.62      32

However, you can call almost any summary function as an argument to
describe. Suppose I wrote a function "fackey" that produced this
output on a factor variable "city":

fackey(city)

label          numeric    count
New York   10            30
London       15            23
Paris          16            22
Rome         20            25

So if you ran "describe" on your data frame, you would get a list of
summary data frames that could be saved with the data frame in an
.Rdata file. Is this what you are looking for?

Jim

On Fri, May 14, 2021 at 4:59 PM PIKAL Petr <petr.pikal at precheza.cz>

wrote:

Hallo Steven

You probably need to be more specific what is your intention. I still

wonder what is the real problem you want to solve.

You loaded binary file and it resulted to 2 data frames. So far so good. But

now I am lost.

You want to merge info from data frame "desc" to data frame "data"? You

can use attr.

You want to make binary file which behaves like the one you get?  Use

save/load.

You want to do something different? So please explain what exactly.

Cheers
Petr

-----Original Message-----
From: Steven Yen <styen at ntu.edu.tw>
Sent: Thursday, May 13, 2021 5:53 PM
To: PIKAL Petr <petr.pikal at precheza.cz>
Subject: Re: [R] Variable labels

Petr

Those attachments (1.jpg, 2.jpg) I sent earlier were just screen
captures (with a third-party program) of what I saw in the
Environment pane right after loading the data. Sorry I cannot explain my

questions well enough.

All I was showing you was, right after loading the binary data file,
I saw two data frames---data which contain the data, and desc which
contains definitions of all variables (as shown in 2.jpg). This is a
data file from the publisher and I wanted to know what it takes to
create a binary data files along with definitions of variables, both in the

environment.

Steven

On 2021/5/13 ?? 09:51, PIKAL Petr wrote:

Hi Steven

I probably do not understand your question correctly. In 1 you show
two

objects "data" 14x42 data frame and "desc" which is 2x14 data frame,
both residing in global environment.

In 2 you show contents of data frame desc where variable are
probably

variable names which are also in data object.

names(data)

and label which is some more elaborate description of the variable.

If you want to move this label into your data object you probably
could use attr

attr(data, "label") <- desc$label

If the order of "variable" is same as the order of data columns.

I do not understand what do you mean by - how to get that "desc" in
there in the environment? It is already part of global environment.
You

want to create some new environment and move you desc there?

Beside, your images are not familiar to me, this is plain R or some
kind of

special GUI like R studio?

Cheers
Petr

-----Original Message-----
From: Steven Yen <styen at ntu.edu.tw>
Sent: Thursday, May 13, 2021 1:37 PM
To: PIKAL Petr <petr.pikal at precheza.cz>
Subject: Re: [R] Variable labels

Petr

Thanks. I am sending this to you privately as I am sending attachment.

1. I load the binary file and see the data frame and what appears
to be description (desc) alongside it (1.jpg).

2. Expanding "desc", I get to read the documentation (contents of

desc).

(2.jpg).

#2 is all I need. I do not need to do anything fancy with the
variable label. I just like my students to have a simple ways of
learning the variables is the data file I provide to them.

Again, my main question is, how to get that "desc" in there in the
environment. Thanks.

Steven

On 2021/5/13 ?? 06:31, PIKAL Petr wrote:

Hi.

Maybe you could use attributes.

dput(vec.m)
structure(list(Group.1 = c(2003, 2021, 2003, 2021, 2003, 2021,
2003, 2021, 2003, 2021, 2003, 2021, 2003, 2021, 2003, 2021, 2003,
2021), variable = structure(c(1L, 1L, 2L, 2L, 3L, 3L, 4L, 4L, 5L,
5L, 6L, 6L, 7L, 7L, 8L, 8L, 9L, 9L), .Label = c("s6", "s5", "s4",
"s3", "s2", "s1.5", "s.7", "s.5", "pod"), class = "factor"),
        value = c(3.29, 0.525, 5.01, 1.385, 16.38, 7.67, 5.535, 3.28,
        25.49, 24.41, 10.285, 12.79, 8.905, 12.92, 1.68, 3.67, 2.595,
        5.06)), row.names = c(NA, -18L), class = "data.frame")

attr(vec.m, "some.kind.of.value") <- c("some specialvector",
"another special vector", "just ordinary vector")

You can access them by attributes or attr.

     attributes(vec.m)
$row.names
     [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18

$names
[1] "Group.1"  "variable" "value"

$class
[1] "data.frame"

$some.kind.of.value
[1] "some specialvector"     "another special vector" "just ordinary

vector"

attr(vec.m, "some")

[1] "some specialvector"     "another special vector" "just ordinary

vector"

Cheers
Petr

-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Steven

Yen

Sent: Thursday, May 13, 2021 10:07 AM
To: Fredrik Karlsson <dargosch at gmail.com>
Cc: R-help Mailing List <r-help at r-project.org>
Subject: Re: [R] Variable labels

Thanks. What I need ?appears? simple. The .RData file is
provided by a third party (likely converted from a different
data format such as SAS in which variable labels (not value
labels) are common). When I load the binary file, in the
?environment? I see, as expected, a data frame showing how many
observations for how many variables. In addition, there is also
an item (in the
environment) (say ?desc?) containing a list of variable labels
(definitions).  I simply like to know how to get ?desc? in the
environment?-it is a convenient way to show definitions of all
variables when you send a binary data file to a third party. Thank

you.

On May 13, 2021, at 2:57 PM, Fredrik Karlsson
<dargosch at gmail.com>

wrote:

Hi,

I am sorry but I don't understand your question, Generally,
"clicking" is not

something you can assume to be implemented for anything in R.

However, if you read the manual for the package

     https://gdemin.github.io/expss/

you get an example at the bottom where an illustration of how
the package

can be used to create Excel tables which would then be easy to
interact with through clicking.

Is that what you wanted?

Fredrik

On Thu, May 13, 2021 at 4:49 AM Steven Yen

<styen at ntu.edu.tw>

wrote:

I insert variable with the expss function as shown below. No
error message. My question is, how to save the variable labels
in the data frame so that I can click to read the labels. Thank you.

mydata<-read_excel("data/Excel/hseinv.xlsx",na=".")
library(expss)
mydata=apply_labels(mydata,
                         year   ="1947-1988",
                         inv    ="real housing inv, millions $",
                         pop    ="population, 1000s",
                         price  ="housing price index; 1982 = 1",
                         linv   ="log(inv)",
                         lpop   ="log(pop)",
                         lprice  ="log(price)",
                         t       ="time trend: t=1,...,42",
                         invpc   ="per capita inv: inv/pop",
                         linvpc  ="log(invpc)",
                         lprice_1="lprice[_n-1]",
                         linvpc_1="linvpc[_n-1]",
                         gprice  ="lprice - lprice_1",
                         ginvpc  ="linvpc - linvpc_1")

code.

--
"Life is like a trumpet - if you don't put anything into it,
you don't get

anything out of it."

   [[alternative HTML version deleted]]