Bölümün yazarı: Sebastian Jentschke
Use the R-version of the PROCESS-macro from within jamovi¶
In preparation: You need to install Rj and to download the most recent version
of the PROCESS-macro.
Open the ZIP-file that you downloaded, go into the folder PROCESS v... for
R
and extract the process.R
-file into your Documents
-directory.
Open a data file that you want to use for your analyses. Afterwards, open
Rj
using the R
-symbol in the Analyses
-icon-bar (Rj
is a
jamovi module; if you have not installed it yet, you may check
Install modules in jamovi).
Now you are ready to write R-code inside jamovi. Run the following code in the
Rj
text box for commands. You may just copy-and-paste the following code.
on Windows
source(file.path(Sys.getenv('HOMEDRIVE'), Sys.getenv('HOMEPATH'), 'Documents', 'process.R'))
on Mac and Linux
source(file.path(Sys.getenv('HOME'), 'Documents', 'process.R'))
Run this code (source…
) by pressing the green triangle. Please be patient,
it might take a moment (up to a minute, depending on how fast your computer
is). You should see an output like this
******************** PROCESS for R Version 4.2 beta ****************
Written by Andrew F. Hayes, Ph.D. www.afhayes.com
Documentation available in Hayes (2022). www.guilford.com/p/hayes3
***********************************************************************
PROCESS is now ready for use.
Copyright 2022 by Andrew F. Hayes ALL RIGHTS RESERVED
Workshop schedule at http://haskayne.ucalgary.ca/CCRAM
Afterwards, the PROCESS-macro is initialized and you can comment the line out
(by putting a #
at the start of the line) → # source(…
Now you are set to run analyses. Please note that the PROCESS-macro for
R uses a different random number generator than SPSS and SAS[1] and that
therefore the bootstrapping confidence intervals for the Indirect
effect(s) of X on Y are different from what the output shown in the book.
Furthermore, does the current version of the PROCESS-macro for R accept
data only in numeric format.[2] Thus, factors must be converted to numeric
form (e.g., 0
and 1
) prior to their use in a PROCESS command. This can
be done using the following command in Rj
(just copy-and-paste it).
for (C in names(data)[sapply(data, is.factor)]) { data[[C]] = as.numeric(data[[C]]) - min(as.numeric(data[[C]])) }
Once this is done, you may just write (or copy-and-paste if you own the e-book)
the commands that are shown in the book. Please note that you have to change
the name of the data set: in this example, taken from p. 188 of Hayes (2022),
the dataset pmi
is required (to download the data sets). Thecommand in the book has
to be adjusted by changing data = pmi
into data = data
(data
refers
to the currently opened dataset in jamovi).
process(data = data, y = "reaction", x = "cond", m = c("import", "pmi"), total = 1, contrast = 1, model = 6,seed = 31216)
Please remember that you have to run the source…
command again whenever you
open a new dataset / a new jamovi session. If you want to run several analyses
with the same dataset / within the same jamovi session, this is not required.
[1] | “The default random number generator in R is different than the default random number generator in SPSS and SAS. Thus, bootstrap confidence intervals generated by R will be different than those produced by SPSS and SAS even when the same seed is used when estimating the same model using the same data.” (Hayes, 2022, p. 613) |
[2] | “PROCESS for R accepts data only in numeric format. Thus, for example, if a variable named sex were coded M and F in the data, these alphabetic codes must be converted to numeric form (e.g., 0 and 1) prior to their use in a PROCESS command.” (Hayes, 2022, p. 612) |