NEWS.md

papaja 0.1.2

• Degrees of freedom are now (as a new default) reported without a comma as a big mark (reported by @Fritz-theCat, #559).
• Fixes incorrect running head in DOCX documents on pages > 2 (reported by @LisaSpitzer, #561).

papaja 0.1.1

• Maintenance update: Adjusted unit tests to avoid removal from CRAN

papaja 0.1.0

Initial CRAN release

• Some improvements to documentation, i.e. documentation of return values, removed examples for internal functions (as per CRAN request), use \donttest{} instead of \dontrun{}.
• add_equals() is now an exported function.

papaja 0.1.0.9999

Template

• Fixed bugs that occurred when using unicode characters in an appendix file (reported by @mschrein, #446 and @kalenkovich, #353).
• Level 4 and 5 headings are now displayed correctly (reported by @jamesrrae, #292).
• E-mail addresses are no longer required metadata (requested by @sbw78, #437).
• Special characters are now displayed correctly in appendices (reported by @mschrein, #446)
• Theorems are now numbered correctly (reported by @shirdekel, #447).
• The Lua-filter that parses the document metadata has been improved to be more robust (see #225).
• Markdown formatting now works for shorttitle and leftheader.
• It is now possible to specify bibliography, appendix, and CSL files (reported by @shirdekel, #482).
• Appendices can and should now be specified using the standard bookdown syntax (see #496 by @pablobernabeu and #484 by @shirdekel)
• Abstract is no longer indented (reported by @AdrienWitt, #510)

Existing functions

• apa_print()
  • Output now always contains a $table element (as suggested in #324 by @Kalif-Vaughn)
  • Column names of $table element are now standardized following the broom glossary) (e.g., term, estimate, conf.int, statistic, df, df.residual, p.value).
    • To maintain backwards compatibility, it is possible to access columns with aliases (for instance, it is possible to write output$table$F and obtain output$table$statistic, see #364).
    • Table columns are labelled (e.g., "\\hat{\\eta}^2_G", "$F$", "$p$, ...) for creating tables via apa_table().
  • The arguments ci, conf_level, and args_confint used for some methods are now depreceated and has been replaced by conf.int, which is now used consistently across methods.
  • Returned tables now consistently have a column for degrees of freedom if applicable. If the degrees of freedom are constant, the column can be removed and degrees of freedom added to the variable label of the test statistic column using the new function transmute_df_into_label().
  • apa_print.aov() and friends
    • Added a new argument estimate to ANOVA-methods, which accepts a character string ("ges", "pes", and "es"; primarily for backward compatibility), a data frame (with columns estimate, conf.low, and conf.high), or a function to determine which measure of effect size is to be reported. If a function is supplied, it should take the ANOVA object as input and return a data frame with columns estimate, conf.low, and conf.high. Currently, output from the effectsize package is supported. By default, if effectsize is installed, generalized eta-squared (point estimate and confidence interval) will be calculated. Otherwise, only the point estimate will be calculated as before.
    • It is now ensured for all cases that names of reporting strings and ordering of terms in table is consistent (reported by @pthomasi)
    • Argument es is now deprecated, but because it is a partial match with estimate, old code using es = "ges", es = "pes" or es = "es" will still work and yield the same results.
  • apa_print.list(): Added progress bars for bootstrapped model comparisons.
  • apa_print.emmGrid() and friends
    • Extended support for additional link functions and output, e.g. (emmeans::joint_tests(), #200, PR #476)
    • Adjusted p values, confidence intervals, and families of tests are now marked as such (see ?apa_print.emmGrid; #200, PR #476)
    • Fixed wrong estimate labels for interaction contrasts (reported by @shirdekel, #456).
  • apa_print.BFBayesFactor() and friends were previously experimental and have now been implemented properly. In the process the arguments scientific, min and max, auto_invert, ratio_subscript, escape, evidential_boost, and hdi have been deprecated and the apa_print.BFBayesFactorList() has been removed for now. The new methods support all BayesFactor analyses and interval null hypotheses. They also grant users more flexibility to specify what information to report.
  • apa_print.lm(): The argument observed_predictors is now deprecated and has been replaced by observed to improve consistency of argument names across methods.
• apa_num() and apa_p() are now the preferred name for the function previously called printnum() and printp().
• variable_label(), variable_labels(), and label_variables() now reside in CRAN package tinylabels, which is automatically loaded when papaja is loaded.
• cite_r() now properly escapes semicolons (reported by @marklhc, #442)

New functions

• apa_print.lme(), apa_print.merMod(), apa_print.mixed()
• glue_apa_results() and add_glue_to_apa_results(), see the new vignette on extending apa_print() (vignette("extending_apa_print", package = "papaja"))
• apa_table.apa_results()
• apa_interval(), apa_confint(), and apa_hdint()
• apa_df()
• transmute_df_into_label() and df_into_label()

Misc

• papaja now depends on
  • tinylabels
  • rmdfiltr (>= 0.1.3; see #427 by @jrennstich)

papaja 0.1.0.9997

Template

• New author metadata field role automatically adds contributor ship roles (e.g. CRedit) to the author note. #375
• apa6_pdf()
  • Now relies on a Lua-filter to process metadata, which should resolve many encoding related problems (e.g., reported by @arcaldwell49, @dcbrh, and @DominiqueMakowski among others; #350, #357, #360)
  • Numbering of title is now disabled when numbersections: true
• apa6_docx()
  • Minor improvements to typesetting (e.g., capitalize running head)
  • docx_fixes.lua now tolerates missing figure captions (reported by @jvcasillas, #365)
• revision_letter_pdf()
  • Defaults to APA6 citation style (also see #369 by @jvcasillas)
  • Adds word count filter

Existing functions

• apa_print()
  • Fixed method dispatch that sometimes failed (#361, #250)
• apa_print.list()
  • Each column of a model comparison table used to have names, these have now been removed to enable unit testing across different R releases and improve consistency across output objects from apa_print().
• printnum.integer()
  • Now always returns a character
• printp()
  • Fixed examples in documentation
• variable_labels.data.frame()
  • When trying to assign labels to columns that do not exist, as of now the function not only throws an error, but also returns the names of the missing columns.
• wsci(), within_subjects_conf_int()
  • Experimental: Returned object now has class c("papaja_wsci", "data.frame") and returns an attribute means for post processing (see new function summary.papaja_wsci).

New functions

• apa_print.manova()
  • Experimental: Added an apa_print method for MANOVA models from the stats package.
• summary.papaja_wsci()
  • Experimental: Returns means and confidence limits calculated from the output of wsci() (suggested by @jstevens5, #362)
• apa_print.papaja_wsci()
• quote_from_tex()
• remove_comments()

Misc

• papaja now depends on
  • R >= 3.6
  • pandoc >= 2.0
  • rmdfiltr (>= 0.1.2) explicitly
• New label_quotes.lua filter to insert excerpts in revision letter; also see quote_from_tex()

papaja 0.1.0.9942

Template

• Lua filters for pandoc 2.0 or later
  • New ampersand filter that's faster and robust to encoding issues (see #225 and #262)
  • Word count filter is automatically applied (see R Markdown tab, #63)
• LaTeX
  • Fixed bug that printed NA at the top of the appendix in doc mode (reported by @paul-grassl, #295)
  • Fixed bug that caused problems with special characters in appendices (reported by @AlexChristensen, #283)
  • In jou mode, tables now span both columns unless requested otherwise (reported by @saladspices, #52, also see #298 by @jstbcs)
  • Level 4 and 5 headings are now properly runin (reported by @jamesrrae, #292)
  • Abstract heading is no longer numbered, when setting number_sections: true (reported by @tcgriffith, #332)
  • Fixed LaTeX warning Command \@parboxrestore has changed. (fix suggested by @CrumpLab, #336)
• Word
  • Overall improved template styles to reduce the need for manual fixes ()

Existing functions

• apa_table()
  • Returns an object of class knit_asis; it's no longer necessary to set the chunk option results = "asis" for tables to be rendered (#280)
  • Column headings are repeated when long tables span multiple pages (suggested by @mischavk, #289)
  • Fixed bug that cause issues with the rendering of tables in Word documents (reported by @mdanka, #301)
  • In PDF files & in cells are automatically escaped
• apa_print()
  • Output is now of class apa_results
• apa_print-htest()
  • An estimate is now reported for 2x2 contingency tables
• printnum.integer()
  • Now prints a zero_string instead of "" when x is 0 (suggested by @e-schott, #311)
  • Improved input validation fixes error when NAis passed (reported by @JanaJarecki, #317)
• printnum.numeric()
  • New add_equals argument (suggested by @mvuorre, #313)

New functions

• label_variable()/label_variables(): Pipable methods to set variable labels
• simple_codebook(): Creates a simple code book from variable labels
• apa_print.manova()/apa_print.summary.manova(): Support for standard manova output

papaja 0.1.0.9837

Template

• LaTeX template
  • Complete rewrite of template as adaptation of rmarkdown::pdf_document() template to enable all options available in the standard template (e.g., A4 paper as requested by @wuschLOR, #193; #106)
  • Adds new appendix fields in YAML front matter (#168)
  • Enables markdown formatting in abstract, author_note, and note fields (#232)
  • Renames lineno fields in YAML front matter to linenumbers
  • Fixes bug that caused citation_package specification to be ignored (reported by @doomlab, #188)
  • Fixes encoding bug in endfloat package (reported by @TobiasKoeln, #133)
  • Adds longtable as endfloat environment (reported by @tormodb, #194)
  • Fixes option clash for geometry package (reported by @PenelopeLQ, #218)
• DOCX reference file
  • New style for tables
  • Applies non-bold style to abstract and references title
• Skeleton
  • Adds RNG seed for caching to first chunk
  • Includes Nature abstract template

Existing functions

• printnum.numeric()
  • New option use_math to control whether to insert $ into the output so that Inf or scientific notation is rendered correctly. Fixes bug that caused an error when printing confidence intervals for one-sided t-tests (#192)
• apa_print.htest()
  • Corrected incorrect sign in calculation of mean differences for t-tests (reported by @jstbcs)
• apa_table()
  • Is now an S3 generic function
  • New method to merge multiple tables in a list using table spanners (#62)
  • Adds new font_size option, which extends small (#223)
  • Fixed bug that caused digits to be applied by row rather than by column (reported by Matti Heino, #173)
  • escape = TRUE is now a visible (in documentation) default (reported by @baguadoramirez, #180)
  • Fixes error due to partial matching of format argument (reported by @mischavk, #223)
• apa_factorial_plot()
  • Fixed bug that caused jit parameter to be ignored (#177)
  • Adds new option use to improve and explicate handling of NAs (#190)
• render_appendix()
  • Preparations for pandoc 2.0 support (reported by @dhagmann, #204)
  • Reenables placement options for figures and tables in appendices (reported by @TobiasKoeln, #172)
  • Enabled rendering appendices in subdirectories (reported by @jvcasillas, #168)
  • Adds support for multiple bibliography files (reported by @jvcasillas, #168)
  • No longer fails when bibliography file specified in the YAML front matter is not found (#189)
  • Enforces UTF-8 encoding
• apa6_docx()/apa6_word()
  • Accepts custom DOCX reference file (suggested by Flo)
• apa6_pdf()
  • Utilizes raw_attributes extension in pandoc > 2.0 (reported by @mischavk, #227)
• cite_r()
  • Omits version numbers if package is not installed
• wsci()
  • Prevents incorrect calculation of confidence intervals in the presence of implicitly or explicitly missing data (#191)

New functions

• apa6_docx()

Experimental:

• apa_print.emmGrid()
• apa_print.summary_emm()
• revision_letter_pdf()

Misc

• Improved support for labelled variables in printnum()-methods
• Renamed labelled class to papaja_labelled to avoid problems due to conflicting implementations of labelled in the haven and Hmisc packages (see tidyverse/haven#329; first reported by @andreifoldes, #199)
• Author affiliations can now also be provided without id
• New experimental revision letter template
• Fixes various encoding problems on Windows (reported by @SongchaoChen, #139, @peter1328, #159, and @mdingemanse, #209)
• Utilizes native R filter to fix in-text citations in pandoc > 2.0 (#186)

papaja 0.1.0.9709

Template

• Reference section is now placed at the end explicitly and allows users to add contents after the reference section. This enables proper appendices in DOCX format.
• Fixed bug that resulted in unwanted "x"s on first page of Word documents (reported by @sjystu, #174)
• Adds proper pagebreaks before abstract and first page of the introduction in Word documents
• Fixes bug in the inclusion of appendices that lead to LaTeX errors when tables were included (reported by @ericpgreen, #179)

Existing functions

• Switched implementation of variable_labels() and related functions from S4 to S3 class. The transition was necessary because the S4 class was incompatible with at least one primitive function (#176). This change may unfortunately break backward compatibility with the previous release.
• apa_table()
  • Now supports for column spanners that span only one column
• render_appendix()
  • Now supports for the application of pandoc-citeproc

New functions

• printnum.integer() (suggested by @LorcanKenny, #178)

Misc

• papaja now depends on bookdown 0.6 to resolve compatibility issues with pandoc 2.0 (reported by @ericpgreen, #170)

papaja 0.1.0.9655

Template

• Improved generation of the author note (#126)
• Author note skeleton text clarified to stress the need for indentation (reported and fixed by @alexholcombe, #132)
• Name of author does not include an ampersand in the correspondence line of the author note (reported by @jacob-long, #110)
• Author note is now omitted if empty in Word
• Formats throw error if name, postal address or e-mail address of corresponding author is missing.
• New rendering option mask removes author name, affiliations, and author note from the title page (suggested by @mcbeem, #114)
• New rendering option replace_ampersands controls whether ampersand_filter() is applied during rendering (suggested by Julia Haaf, Jeff Rouder, and Anne Scheel)
• Rendering option lang removed because it is not compatible with current pandoc versions (reimplementation needed)

Existing functions

• apa_barplot()
  • Now prints ylab on every plot as do apa_lineplot() and apa_beeplot() (#124)
• theme_apa()
  • Added support for right y-axis.
• render_appendix()
  • The function now removes the placement options of figures and tables in appendix (reported by @TobiasKoeln, #113)
• inline_hook()
  • Added support for difftime objects
• r_refs()
  • Now enforces UTF-8 encoding (reported by @JimGrange, #123)
• apa_print.anova() and related
  • Now correctly italicize MSE
  • Corrected bug in the calculation of eta^2 and denote as estimate using ^.
• within_subjects_conf_int()
  • Calculates confidence intervals in presence of NAs (reported by @TobiasKoeln, #137)
• apa_print()
  • Returned tables are of class apa_results_table, column names are easy to address, and each column is labelled for pretty printing.
• apa_table()
  • Takes into account additional options passed to format.args parameter (reported by @lnalborczyk, #131)
• apa_print.htest()
  • Supports oneway.test() objects.
• apa6_word()
  • Enforces UTF-8 encoding while augmenting the YAML front matter (reported by @lindsayhinzman, #156)

New functions

• apa_factorial_plot()
• apa_factorial_plot.afex()
• apa_print.glm()
• apa_print.summary.glm()
• print.apa_results_table()
• variable_label()

Experimental:

• apa_print.BFBayesFactor()
• apa_print.BFBayesFactorTop()
• apa_print.BFBayesFactorList()

papaja 0.1.0.9492

Template

• The ampersand filter is applied with --vanilla to avoid problems with customized .Rprofiles (reported by @awellis, #96)
• Fixed bug in ampersand_filter() regex, which caused some citations to be missed depending on preceding citations (reported by @awellis, #96, and @m-Py, #100)
• On windows the Batch file to apply the ampersand filter is now created at the current location and deleted on exit to avoid permission problems (reported by @stahl-c)
• Changed default graphics devices to pdf and png to improve compatibility and rendering speed; postscript and tiff are no longer used be default
• Author note is omitted if no corresponding author is defined and no author note text is provided (reported by @jacob-long, #108; fixed by @jacob-long, #109)

Existing functions

• apa_print.aov()
  • Fixed bug that caused incorrect calculation of eta squared
  • MSE is now typeset in regular font rather than italic as per APA guidelines
• apa_table()
  • Added proper support for automated bookdown cross-referencing of tables: Chunk label is now automatically added as \label{} tag.
  • Now supports na_string and digits options
• theme_apa()
  • Now uses ggplot::margin() to set margins because ggplot::unit() is now deprecated.
  • Changed base size default from 14 to 12 and adjusted some margin and spacing defaults
• r_refs()
  • Fixed bug that caused some existing references to be added to the bib-file (#102)
• cite_r()
  • Fixed incorrect handling of survival-book reference

New functions

• apa_print.lsmobj()
• apa_print.summary.ref.grid()
• apa_print.glht()
• apa_print.summary.glht()

Misc

• New global option papaja.na_string with default value "NA" used by apa_table() and printnum()
• Updated example file and README

papaja 0.1.0.9470

Template

• The template now automatically invokes an R-based pandoc filter that replaces & with and in in-text citations to fully adhere to APA citation guidelines; (see #53)
• Initial chunk in template now is set to use chunk option include = FALSE.

Existing functions

• apa_table.latex()
  • If landscape = TRUE tables are now always in longtable environments to improve interplay with the LaTeX package endfloat and correctly honoring the figsintext option in the YAML header (reported by @m-Py, #90)
  • Added clarification that the chunk option results = "asis" is required when apa_table() is called (requested by RMHogervorst, #93)
• apa_print.lm()
  • Calling lm() on a data.frame columns by using $ resulted in regression term names that broke apa_table() (reported by @m-Py, #87)
• render_appendix()
  • Now builds on knitr::knit_child() to fix a bug causing figures not to render in the appendix; render_appendix() now has to be called within an R Markdown document. (reported by @fdabl, #70)
• apa_*plot()
  • Expressions can now be used as legend titles (see #94)

papaja 0.1.0.9456

Template

• apa6_pdf and apa6_word now adapt pdf_document2() and word_document2() from the bookdown package. This enables the use of bookdown cross-referencing syntax including automatically generated table and figure labels as detailed here (see #2).
• LaTeX
  • If numbersections: true numbering now starts at the body of the text. The abstract is no longer a numbered heading (reported by @lnalborczyk, #85)
  • New option footnotelist to create list of footnotes at the end of the document (see #74)
• Word
  • Fixed bug that the wordcount set in the YAML header was not reported on the title page of Word documents (reported by @TobiasKoeln)

Existing functions

• apa_table.latex()
  • Fixed bug causing longtables and and landscape tables to appear in text rather than at the end of the document when figsintext: false (reported by @m-Py, #64).
  • Fixed bug causing LaTeX to break when table contains %. All unescaped % are now automatically escaped even if escape = FALSE (reported by @m-Py, #66).

Vignette

• Corrected various typos (reported by JooYoung Seo)

papaja 0.1.0.9440

Template

• Fixed bug in inline hook that omitted single numbers (reported by @hcp4715, #79)
• Chunk option default changed to results = "markup" (see #77)

Existing functions

• apa_barplot() can calculate ylim when measure of dispersion cannot be calculated (#76)
• apa_beeplot() brightens data points instead of using transparency
• apa_print()
  • Renamed returned list elements to statistic, estimate, full_result, and table (#67)
  • All methods now return lists of the same structure (table may be NULL)
• apa_print.lm() has new option observed_predictors to calculate proper R^2 confidence intervals

Misc

• MBESS package is now suggested and not imported

papaja 0.1.0.9423

Template

• Fixed a bug causing a LaTeX error in jou mode (reported by @saladspices, #48)
• Default lang is now english; fixed a pandoc-citeproc bug (reported by @mvuorre and @sebastiansauer, #56 and #57; fixed by @mvuorre, #58)
• Removed author field department, information can be supplied as part of the address field
• New option author_note; note is used to add additional information to the title page
• Changed heading "Sample" to "Participants"
• Fixed bug that supplied CSL file was ignored
• Proper author name concatination on title page (reported by Ulf Mertens)
• LaTeX
  • New options figurelist and tablelist to create lists of figure or table captions at the end of the document
  • Added upgreek package
  • Fixed bug causing figsintext to have no effect (reported by @mattsigal, #61)
  • Added page break before reference section (reported by @mattsigal, #60)
• Word
  • Fixed bug causing a deranged title page (reported by @mvuorre, #59)
  • Figure captions are now preceded by "Figure."

Existing functions

• apa_print()
  • Improved type setting of predictors in tables
  • New supports car::LeveneTest() output
  • Dropped redundant df-column in apa_print.lm() table
  • Fixed bug causing the omission of R^2 confidence intervalls in apa_print.lm() when F-values were small
  • Fixed bug in apa_print.list() causing an error with only two models
  • Refactored apa_print()-family for ANOVA objects
    • Only df corrected for nonsphericity are reported with two decimals
    • Added indication of sphericity correction to apa_print.Anova.mlm() table
    • print_anova() rounds effect size estimates to three decimals
    • Can now calculate eta^2 effect sizes
    • New option intercept determines if intercept tests are included in the output
    • New option mse to report mean squared error terms (thanks to @TomHardwicke for reporting early bugs, #49)
• apa_table()
  • Dropped redundant row_names option
  • New option stub_indents adds title sections to indented rows
  • Renamed option added_colnames to added_stub_head
  • apa_table.latex()
    • New option col_spanners to add column spanners
    • New option small to decrease font size for large tables
    • Improvements to long and landscape tables
• apa_barplot()
  • Now works with white space in factor names
  • Improvements to automatic setting of ylim
  • Fixed bug that led to the incorrect drawing of user-specified colors
  • Only one legend is drawn for multiple facets
  • New option args_arrows to customize error bars
  • Invisibly returns means and measures of dispersion
• cite_r()
  • New options pkgs and withhold to black- or whitelist packages (#50)
  • Output retains structure if footnote = TURE but no bib-file is found
• r_refs()
  • Now handles packages with mulitple references (reported by @mattsigal, #65)
• conf_int()
  • Fixed bug that led to negative values

New functions

• apa_print.afex_aov() (#47)
• apa_print.list() (#27)
• wsci()
• apa_lineplot()
• apa_beeplot() (thanks to @TobiasKoeln for reporting early bugs, #72)
• theme_apa()
• render_appendix()
• fetch_zotero_refs()
• in_paren()

Dropped functions

• fetch_web_refs() evolved into a clone of downloder::download()
• apa_prepare_doc() is no longer needed

Misc

• papaja sets global options papaja.mse and papaja.sphericity_correction among others
• Updated example manuscript
• Added vignette

papaja 0.1.0.9054

Template

• Added wordcount parameter to YAML header
• Changed YAML header structure to support proper author and affiliation information (requires updates of existing Rmd-files)
• LaTeX: Changed SVG device to PS
• Removed author field 'department' from the YAML header
• Added LaTeX fix for tightlist-bug in pandoc 1.14
• Fixed extra comma in automatically generated corresponding author line when no note is provided

Existing functions

• apa_print()
  • Added table output-element to apa_print.lm() containing complete regression table
  • Refactored and tested apa_print()-family for ANOVA objects
  • Added table output-element to apa_print()-family for ANOVA objects containing complete ANOVA table
  • First experimental implementation of F-test model comparisons (car::leveneTest objects)
  • Estimates in apa_print.htest() and apa_print.lm() can be formatted by passing options to printnum()
• apa_table()
  • Added midrules option for apa_table.latex() and cleaned up LaTeX output
  • Fixed vertical space between table and table notes for apa_table.latex()
  • Bug fixes
• r_refs()
  • Now attempts to create references for packages used in cached chunks
  • Creates proper references based on citation file if available instead of generating default references from description file
• apa_prepare_doc()
  • Fixed bug when lang was not set in YAML header
• apa_barplot()
  • Fixed ignorance of user specified color for one factorial plots
  • Fixed default ylim choice to consider the lower bound of confidence intervals
• apa_table()
  • Removed redundant row_name parameter
  • Fixed escaped multicolumn-headings in apa_table.latex()

New functions

• apa_print.anova.mlm()
• apa_barplot()
• cite_r()
• fetch_web_refs()

Misc

• Set up proper package structure for automatic unit testing
• Created news file
• Fixed some typos in the example document

papaja 0.1.0.9000

• Initial release on GitHub