crsh / papaja Goto Github PK

There's a minor bug in apa_barplot(): If lower limit of y-axis is not 0, bars are drawn even below the x-axis

Default should probably be true.

I'm having trouble with getting tables to be rescaled to fit the page. Below is the table I'm using (and I've also tried using the xtable and ztable packages as well as basic markdown tables):

\begin{table}[htbp]
\center
\caption{Original DAST-10}
\label{table1}
\begin{tabular}{cl}
\toprule
& Item
\toprule
1 & Have you used drugs other than those required for medical reasons?
2 & Do you use more than one drug at a time?
3 & Are you always able to stop using drugs when you want to? (-)
4 & Have you ever had blackouts or flashbacks as a result of drug use?
5 & Do you ever feel bad or guilty about your drug use?
6 & Does your spouse (or parents) ever complain about your involvement with drugs?
7 & Have you neglected your family because of your use of drugs?
8 & Have you engaged in illegal activities in order to obtain drugs?
9 & Have you ever experienced withdrawal symptoms (felt sick) when you stopped taking drugs?
10 & Have you had medical problems as a result of your drug use (e.g., memory loss, hepatitis, convulsions, bleeding)?
\bottomrule
\end{tabular}
\end{table}

This would be necessary for anova, aov, etc. and lm objects.

When I pass an aov object from repeated-measures ANOVA to apa_print, it fails now:

Fehler in UseMethod("arrange_anova", x) :
nicht anwendbare Methode für 'arrange_anova' auf Objekt der Klasse "c('summary.aov', 'listof')" angewendet

Why did you do that?

I would like to see a collection of small example manuscripts demonstrating different things that can be done using papaja and related packages, such as DiagrammR (https://github.com/rich-iannone/DiagrammeR).

apa_print() provides confidence intervals for R² (lm) but as of now no confidence intervals are calculated for ANOVA objects.

If we define classes with attributes for the tables produced by apa_print() it would be possible to automatically create sensible table notes containing, e.g., R² for regression models etc.

Standard knitr fig.cap is printed in all italics and without a leading "Figure x", where x is the automatic number of the figure. Not sure if or how this could be done.

?papaja opens the documentation in papaja.R. This text should help new users of the package getting started.

Currently, cite_r() has a white list of packages to cite (pkgs), but it would be probably be much more useful to have a black list of packages not to cite. This would prevent situations in which a package is later added to the analysis script but not to the call of cite_r.

Currently, model comparison anova-objects are not supported by apa_print() but they really should be.

My understanding is that current APA style requires the bibliography to use hanging indentation (i.e. the first line is normal, all the following lines are indented, for each reference). I didn't see an option to change this in the documentation.

I experienced so problems with image scaling when knitting MS Word documents. Embedded images ![](...) are scaled down and distorted (PDF) while plot generated from R are not scaled to fit the page with the custom settings. Maybe we can adjust the chunk defaults depending on the output format?

If I do not specify a note in my .Rmd template, I erroneously get two commas on the first page of the manuscript:

Correspondence concerning this article should be addressed to Donald Duck**, ,**
Gansstr. 2, 12345 Entenhausen, Germany.

I can replicate this in the example file example.Rmd of this repository if I simply remove the note:

note: |

Currently, a development version of papaja is available at https://github.com/crsh/papaja.

I do not think that this is the desired behaviour, if I do not want to insert a note?

Maybe use printnum() parameters that take vectors in apa.table() where each vector element represents on column of the table.

The references do not come out according to APA 06 (hanging indent), but rather they are indented on the first line. How would I fix this issue?

While at it: Document that this option has no effect for MS Word documents, yet.

Do I need to do anything in addition to installing papaja to get the apa_table function to work? I get the following error when I try and run your code from the example document (i.e., cars descriptives):

Error in apa_table.latex(...) : could not find function "validate"

It is for this reason that I've been trying to use LaTeX code to create a table. A second problem of mine, however, is I don't know where to add \usepackage{}. For example, with the table I've tried \usepackage{adjustbox} or tabularx, as suggested by you, but I get an error.

This is important for the first release.

longtable doesn't work in two-column layouts, which are produced by apa6 when the class is set to jou. We should try if we can use supertabular instead.

Is there any possibilty to add landscape pages (i.e., for large tables) to the document? I understood that I need the pdflscape environment in TeX to do so, but including this package in the yaml header doesn't change anything in papaja documents, while these features are accepted in standard RStudio pdfs. Is there a convenient way to load additional TeX environments?

My code:
includes:
in_header: \usepackage{pdflscape}

As of now there are no sanity checks for function inputs. In order to receive meaningful warnings and errors basic sanity checking of input would be nice.

Let's see if we can create a MS Word reference-file to make Word documents look like apa6-documents when the class is set to doc. Also: Is it possible to set reference-docx in the yaml-header conditional on the selected document class?

When using class: jou I get the error:

! LaTeX Error: Something's wrong--perhaps a missing \item. See the LaTeX manual or LaTeX Companion for explanation. Type H <return> for immediate help. ... l.141 pandoc.exe: Error producing PDF from TeX source Error: pandoc document conversion failed with error 43

This happens with newest papaja, R 3.2.1 and newest RStudio 0.99.442 under Windows 10 with MikTeX and Ubuntu 14.04 with TeX Live. All documents I tried work fine with class: doc and class: man. This error is reproducible with the standard example document. Please correct me if I'm missing something here.

An alternative to delivering a hard copy of the apa6 document class, which we do right now, would be to use the version shipped with the LaTeX-distribution:

---
title: "Untitled"
author: "Author"
output: pdf_document
classoption: "man"
documentclass: "apa6"
geometry: false

---

However, this way we would need to include some additional arguments in the preamble, which could be done with includes in the yaml-header, the _output.yaml-file, or the format-function.

Create a function that uses the same parameters as apa_barplot(), but draws nice lineplots

Fehler in rmarkdown::pdf_document(template = template, fig_caption = fig_caption, :
unbenutztes Argument (keep.tex = keep.tex)
Calls: ... create_output_format -> do.call -> ->
Ausführung angehalten

Today I got papaja to run outside of RStudio in Ubuntu 14.04. It was not easy because it took me some time to install a recent version of pandoc and then call it from within R.

In the readme file of this repository you describe to use rmarkdown::draft() in order to create an APA article if you are not using RStudio [ like this: rmarkdown::draft("mymanuscript.Rmd", template = "apa6", package = "papaja") ]. This does not seem to work for me. I get this error message: The file 'mymanuscript.Rmd' already exists.

If I just use a different file name I get a weird .rmd document containing this abstract:

"abstract: In der Rede wird folgende These diskutiert: "Optozentristisch durchseucht sind Philosophie und die Wissenschaft. [...]"

It seems that draft() does not create a .pdf file, but rather tries to create a new file from a template (see http://rmarkdown.rstudio.com/developer_document_templates.html)

To create a papaja APA6 article, I needed to use the rmarkdown function render(), instead of draft(). Using rmarkdown::render(), I could replicate creating your example article as well as articles that contained changes. How to use it:

render("myExample.rmd", output_file="myExample.pdf")

tl;dr: If you are using papaja outside of RStudio, it is possible that you can only create an article by calling rmarkdown::render(), not rmarkdown::draft().

When I use apa_table() to insert a table into a papaja::apa6_pdf document, this table will have two rows containing the row names of the matrix that I passed to apa_table() (also happens if I pass a data.frame instead). If I replicate the example from the ?apa_table() documentation, it looks like this:

Descriptives speed dist

Mean Mean 15.4 42.98

SD SD 5.29 25.77

Min Min 4 2

Max Max 25 120

This does not happen if I directly call kable() instead. I am using papaja_0.1.0.9000.

In "doc" mode, when the abstract is long (~300 words), just the word "Abstract" appears on the first page and the rest appear on page 2. It makes it look like the abstract section has no title. Inserting a pagebreak in the tex file does not fix this.

Use the knitr::opts_knit$get("rmarkdown.pandoc.to") to query the output format and remove the output_format parameter in apa.table().

Found this solution at: https://trinkerrstuff.wordpress.com/2014/11/18/rmarkdown-alter-action-depending-on-document/

When using class:journal, the paragraphs do not get indented.

As soon as the new release is on CRAN in which output objects are of afex_aov S3 class.

It would be nice if there was an option to add significance indicators to a apa.table(), e.g. for correlation tables, by passing a matrix of the same dimensions that contains p-values (s. corr.test() from the psych-package) or by passing a matrix of the same dimensions with logicals to add indicators of something entirely different.

It's not a pretty solution to have users execute apa_prepare_doc() manually at the start of a document. Maybe this can be avoided by automatically executing the function when the package is loaded/attached. See http://r-pkgs.had.co.nz/r.html#r-code

Unit tests for results of printnum(), apa.stat(), and subfunctions using testthat-package.

Create a convenience wrapper form printnum() inside apa.table() to produce complete ANOVA and Regression tables.

When printing plots from a data.table in the form DT[,plot(xy)], data.table returns NULL next to the plot. This NULL is also printed in the apa manuscript.

Example:

dat <- data.table(x=1:10,y=11:20,subj=rep(c(1,2),5))
dat[subj==1,plot(x,y)]
dat[subj==2,lines(x,y)]

When knitting the standard papaja template to a .docx file non-ascii characters (I tested ÄÜÖäüö߀) in the abstract and the note sections of the word document are seemingly not encoded correctly. The same characters are displayed properly when used in the title or outside the header in the normal document text. Characters are also displayed properly in the two sections when knitting a pdf file.

I am using MikTex 2.9.5105 and the latest development version of papaja. To view the .docx file I use MS Word 2013.

Many journals like to have all figures as seperate high resolution files. It would be nice if it was possible to automatically save all plots.

Find a way to create automatic and referenceable numbering of figures and tables in MS Word.

A solution for tables will probably have to be incorporated into apa.table.word().

Currently abstract, affiliations, etc. are not inserted into Word documents.

Unless the align parameter is specified in a call to apa_table.latex(), it seems to me that aligning columns with numbers by decimal point is a sensible default. The siunitx package seems like a good option to achieve this (see example here).

The tables packages provides a powerful interface to produce tables from the data. It would be nice if we found a way to make this work with apa.table(). As it is, tabular objects can be used to produce pretty tables in LaTeX but we would need to find a way to make it work with Markdown/Pandoc.

This may be a good place to start looking for solutions: jgm/pandoc-templates#20

A problem, as of yet, seems to be that figure captions don't work in .odt: jgm/pandoc#570