update and edit paper

paper/paper.Rmd

@@ -33,57 +33,60 @@ plots that complies with the grammar of graphics [@wickham2010]. Further, `gglm`
 utilizes the `broom` and `broom.mixed` R packages to provide support for
 diagnostic plots produced from a variety of model object classes across a wide
 variety of R packages [@broom; @broom.mixed]. A quartet of diagnostic plots can 
-be quickly created using `gglm`'s homonymous function, or through instructive
-and intuitive layer functions added to a `ggplot2` object [@ggplot2].
+be quickly created using `gglm`'s homonymous function, or plots can be created
+individually through instructive and intuitive layer functions added to a 
+`ggplot2` object [@ggplot2].
 
-# Statement of need
+# Statement of Need
 
 When scientists, statistical practitioners, students, and others implement
 statistical models, it is of the utmost importance that the modeling assumptions
 are verified through visual diagnostics in order to ensure valid statistical 
 inference. The R statistical software language provides a method for producing
 diagnostic plots for linear model objects created with `stats::lm`, however 
 these plots are visually unappealing, inconsistent with diagnostic plots 
-produced for other R packages and model types, and out of place in modern 
-statistics and data science courses focused on learning R with the `tidyverse`
+across other R packages and model types, and out of place in modern statistics
+and data science courses focused on learning R with the `tidyverse` 
 [@tidyverse]. 
 
 `gglm` addresses the described issues with current diagnostic plots in R by
 providing a consistent interface for producing beautiful and publication-ready
-diagnostic plots for a large variety of R packages and model types (linear 
+diagnostic plots across a large variety of R packages and model types (linear 
 models, linear mixed models, generalized linear mixed models, etc.). `gglm` 
 provides functionality to quickly produce four common diagnostic plots, similar
 to `stats::plot.lm`, but produced by `ggplot2`. Further, `gglm` provides a suite
 of layer functions adhering to the grammar of graphics which allow the user to
 create and fine-tune their diagnostic plots through `ggplot2`'s intuitive 
 interface. The layer functions are particularly applicable in modern courses 
-teaching linear regression where students have already learned `ggplot2`, and in
-particular they are used in Harvard University's introductory statistics course
-[@mcconville2023]. Outside of educational benefits, `gglm` has potential to
-allow researchers to more easily publish elegant diagnostic plots. `gglm` has
-been downloaded from CRAN over 23,000 times as of January 2024. 
+teaching linear regression where students have already learned `ggplot2`. For 
+example, `gglm` and its layer functions are used in Harvard University's
+introductory statistics course [@mcconville2023]. Outside of educational 
+benefits, `gglm` has potential to allow researchers to more easily publish
+elegant diagnostic plots. `gglm` has been downloaded from CRAN over 23,000 times
+as of January 2024. 
 
-# Usage and Philosophy
+# Usage and Features
 
-`gglm` has a simple philosophy for usage of the package: "be easy, intuitive, 
-and customizable". This philosophy comes about from the understanding that an 
-individual producing a diagnostic plot will be in one of two camps: 1) the
-individual who wants an *easy* to use tool that allows them to quickly check
-their model diagnostics, or 2) the individual who wants an *intuitive and 
-customizable* tool that allows them to look closely at their diagnostics for the
-purposes of education, fine-tuning for publication, or other reasons. `gglm` 
-satisfies the individuals in both camps.
+`gglm` achieves a balance in functionality by being both as easy to use as the 
+built-in `stats::plot.lm` method, yet still highly intuitive and customizable 
+for the curious user. `gglm` is designed with these traits in mind due to the
+understanding that an individual producing a diagnostic plot will most likely be 
+in one of two camps: 1) the individual who wants an *easy* to use tool that
+allows them to quickly check their model diagnostics, or 2) the individual who
+wants an *intuitive and customizable* tool that allows them to look closely at
+their diagnostics for the purposes of education, fine-tuning graphics for
+publication, or other reasons. `gglm` satisfies the members of both camps.
 
-The `gglm::gglm` function is made for folks in the first camp who are looking
-a more aesthetically pleasing alternative to `stats::plot.lm`. In practice, the
-process of using `gglm::gglm` is as simple as and more general than using 
+The `gglm::gglm` function is made for folks in the first camp who are looking 
+for a more aesthetically pleasing alternative to `stats::plot.lm`. In practice,
+the process of using `gglm::gglm` is as simple as and more general than using 
 `stats::plot.lm`, with steps as follows:
 
 + fit a model of any class listed in `gglm::list_model_classes`,
 + call `gglm::gglm` on the saved model object. 
 
-The `gglm::stat_*` functions are thus for the individual in the second camp. 
-`gglm` provides seven functions of this sort, including those that produce the 
+The `gglm::stat_*` functions are thus for those in the second camp. `gglm` 
+provides seven functions of this sort, including those that produce the 
 following plots: Cook's distance by leverage, Cook's distance by observation
 number, fitted values by residual values, normal QQ, residual histogram, 
 residual values by leverage, and scale by location. The steps to produce a 
@@ -103,9 +106,12 @@ Functionality similar to that of `gglm`'s is provided by a variety of R
 packages. As mentioned throughout, `stats` provides a `plot` method for 
 producing diagnostic plots for `lm` objects with base R graphics [@R]. Further,
 `lindia` produces diagnostic plots for `lm` objects with `ggplot2` graphics, but
-does not include functions that adhere with the grammar of graphics. Finally, 
-many packages provide methods for plotting diagnostics based on their own model 
-classes (see, e.g. `lme4::plot.merMod`), however these methods are do not have
-consistent usage across packages [@lme4]. 
+does not include functions that adhere with the grammar of graphics [@lindia]. 
+Finally, many packages provide methods for plotting diagnostics based on their
+own model classes (see, e.g. `lme4::plot.merMod`), however these methods are do
+not have consistent usage across packages [@lme4]. `gglm` hence addresses a
+significant gap in functionality by creating a consistent framework for
+producing diagnostic plots across R packages and model types while adhering to 
+the grammar of graphics. 
 
 # References

paper/paper.log

@@ -1,4 +1,4 @@
-This is XeTeX, Version 3.141592653-2.6-0.999995 (TeX Live 2023) (preloaded format=xelatex 2024.1.15)  15 JAN 2024 17:30
+This is XeTeX, Version 3.141592653-2.6-0.999995 (TeX Live 2023) (preloaded format=xelatex 2024.1.15)  15 JAN 2024 21:40
 entering extended mode
  restricted \write18 enabled.
  %&-line parsing enabled.
@@ -924,8 +924,8 @@ Package hyperref Warning: Suppressing link with empty target on input line 271.
 
 Package hyperref Warning: Suppressing link with empty target on input line 271.
 
-File: /home/grayson/R/x86_64-pc-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png Graphic file (type bmp)
-</home/grayson/R/x86_64-pc-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png>
+File: /home/grayson/R/aarch64-unknown-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png Graphic file (type bmp)
+</home/grayson/R/aarch64-unknown-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png>
 
 Package fancyhdr Warning: \headheight is too small (62.59596pt): 
 (fancyhdr)                Make it at least 63.55022pt, for example:
@@ -934,12 +934,12 @@ Package fancyhdr Warning: \headheight is too small (62.59596pt):
 (fancyhdr)                \addtolength{\topmargin}{-0.95425pt}.
 
 LaTeX Font Info:    Font shape `TU/lmss/m/it' in size <8> not available
-(Font)              Font shape `TU/lmss/m/sl' tried instead on input line 331.
+(Font)              Font shape `TU/lmss/m/sl' tried instead on input line 334.
 [1
 
 ]
-File: /home/grayson/R/x86_64-pc-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png Graphic file (type bmp)
-</home/grayson/R/x86_64-pc-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png>
+File: /home/grayson/R/aarch64-unknown-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png Graphic file (type bmp)
+</home/grayson/R/aarch64-unknown-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png>
 
 Package fancyhdr Warning: \headheight is too small (62.59596pt): 
 (fancyhdr)                Make it at least 63.55022pt, for example:
@@ -948,8 +948,8 @@ Package fancyhdr Warning: \headheight is too small (62.59596pt):
 (fancyhdr)                \addtolength{\topmargin}{-0.95425pt}.
 
 [2]
-File: /home/grayson/R/x86_64-pc-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png Graphic file (type bmp)
-</home/grayson/R/x86_64-pc-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png>
+File: /home/grayson/R/aarch64-unknown-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png Graphic file (type bmp)
+</home/grayson/R/aarch64-unknown-linux-gnu-library/4.3/rticles/rmarkdown/templates/joss/resources/JOSS-logo.png>
 
 Package fancyhdr Warning: \headheight is too small (62.59596pt): 
 (fancyhdr)                Make it at least 63.55022pt, for example:
@@ -963,18 +963,18 @@ LaTeX2e <2023-11-01>
 L3 programming layer <2024-01-04>
  ***********
 Package rerunfilecheck Info: File `paper.out' has not changed.
-(rerunfilecheck)             Checksum: 0B3BF1C9D2BC7F6B8E6DB44814E9B8B7;651.
+(rerunfilecheck)             Checksum: 36E8A3BFD73D3865102046825CE75FCD;641.
 Package logreq Info: Writing requests to 'paper.run.xml'.
 \openout1 = `paper.run.xml'.
 
  ) 
 Here is how much of TeX's memory you used:
  34349 strings out of 476822
- 703677 string characters out of 5804165
- 1944174 words of memory out of 5000000
+ 703729 string characters out of 5804165
+ 1943174 words of memory out of 5000000
  55827 multiletter control sequences out of 15000+600000
  564925 words of font info for 82 fonts, out of 8000000 for 9000
  14 hyphenation exceptions out of 8191
- 84i,12n,87p,678b,850s stack positions out of 10000i,1000n,20000p,200000b,200000s
+ 84i,13n,87p,678b,850s stack positions out of 10000i,1000n,20000p,200000b,200000s
 
 Output written on paper.pdf (3 pages).

paper/paper.md

@@ -33,57 +33,60 @@ plots that complies with the grammar of graphics [@wickham2010]. Further, `gglm`
 utilizes the `broom` and `broom.mixed` R packages to provide support for
 diagnostic plots produced from a variety of model object classes across a wide
 variety of R packages [@broom; @broom.mixed]. A quartet of diagnostic plots can 
-be quickly created using `gglm`'s homonymous function, or through instructive
-and intuitive layer functions added to a `ggplot2` object [@ggplot2].
+be quickly created using `gglm`'s homonymous function, or plots can be created
+individually through instructive and intuitive layer functions added to a 
+`ggplot2` object [@ggplot2].
 
-# Statement of need
+# Statement of Need
 
 When scientists, statistical practitioners, students, and others implement
 statistical models, it is of the utmost importance that the modeling assumptions
 are verified through visual diagnostics in order to ensure valid statistical 
 inference. The R statistical software language provides a method for producing
 diagnostic plots for linear model objects created with `stats::lm`, however 
 these plots are visually unappealing, inconsistent with diagnostic plots 
-produced for other R packages and model types, and out of place in modern 
-statistics and data science courses focused on learning R with the `tidyverse`
+across other R packages and model types, and out of place in modern statistics
+and data science courses focused on learning R with the `tidyverse` 
 [@tidyverse]. 
 
 `gglm` addresses the described issues with current diagnostic plots in R by
 providing a consistent interface for producing beautiful and publication-ready
-diagnostic plots for a large variety of R packages and model types (linear 
+diagnostic plots across a large variety of R packages and model types (linear 
 models, linear mixed models, generalized linear mixed models, etc.). `gglm` 
 provides functionality to quickly produce four common diagnostic plots, similar
 to `stats::plot.lm`, but produced by `ggplot2`. Further, `gglm` provides a suite
 of layer functions adhering to the grammar of graphics which allow the user to
 create and fine-tune their diagnostic plots through `ggplot2`'s intuitive 
 interface. The layer functions are particularly applicable in modern courses 
-teaching linear regression where students have already learned `ggplot2`, and in
-particular they are used in Harvard University's introductory statistics course
-[@mcconville2023]. Outside of educational benefits, `gglm` has potential to
-allow researchers to more easily publish elegant diagnostic plots. `gglm` has
-been downloaded from CRAN over 23,000 times as of January 2024. 
+teaching linear regression where students have already learned `ggplot2`. For 
+example, `gglm` and its layer functions are used in Harvard University's
+introductory statistics course [@mcconville2023]. Outside of educational 
+benefits, `gglm` has potential to allow researchers to more easily publish
+elegant diagnostic plots. `gglm` has been downloaded from CRAN over 23,000 times
+as of January 2024. 
 
-# Usage and Philosophy
+# Usage and Features
 
-`gglm` has a simple philosophy for usage of the package: "be easy, intuitive, 
-and customizable". This philosophy comes about from the understanding that an 
-individual producing a diagnostic plot will be in one of two camps: 1) the
-individual who wants an *easy* to use tool that allows them to quickly check
-their model diagnostics, or 2) the individual who wants an *intuitive and 
-customizable* tool that allows them to look closely at their diagnostics for the
-purposes of education, fine-tuning for publication, or other reasons. `gglm` 
-satisfies the individuals in both camps.
+`gglm` achieves a balance in functionality by being both as easy to use as the 
+built-in `stats::plot.lm` method, yet still highly intuitive and customizable 
+for the curious user. `gglm` is designed with these traits in mind due to the
+understanding that an individual producing a diagnostic plot will most likely be 
+in one of two camps: 1) the individual who wants an *easy* to use tool that
+allows them to quickly check their model diagnostics, or 2) the individual who
+wants an *intuitive and customizable* tool that allows them to look closely at
+their diagnostics for the purposes of education, fine-tuning graphics for
+publication, or other reasons. `gglm` satisfies the members of both camps.
 
-The `gglm::gglm` function is made for folks in the first camp who are looking
-a more aesthetically pleasing alternative to `stats::plot.lm`. In practice, the
-process of using `gglm::gglm` is as simple as and more general than using 
+The `gglm::gglm` function is made for folks in the first camp who are looking 
+for a more aesthetically pleasing alternative to `stats::plot.lm`. In practice,
+the process of using `gglm::gglm` is as simple as and more general than using 
 `stats::plot.lm`, with steps as follows:
 
 + fit a model of any class listed in `gglm::list_model_classes`,
 + call `gglm::gglm` on the saved model object. 
 
-The `gglm::stat_*` functions are thus for the individual in the second camp. 
-`gglm` provides seven functions of this sort, including those that produce the 
+The `gglm::stat_*` functions are thus for those in the second camp. `gglm` 
+provides seven functions of this sort, including those that produce the 
 following plots: Cook's distance by leverage, Cook's distance by observation
 number, fitted values by residual values, normal QQ, residual histogram, 
 residual values by leverage, and scale by location. The steps to produce a 
@@ -103,9 +106,12 @@ Functionality similar to that of `gglm`'s is provided by a variety of R
 packages. As mentioned throughout, `stats` provides a `plot` method for 
 producing diagnostic plots for `lm` objects with base R graphics [@R]. Further,
 `lindia` produces diagnostic plots for `lm` objects with `ggplot2` graphics, but
-does not include functions that adhere with the grammar of graphics. Finally, 
-many packages provide methods for plotting diagnostics based on their own model 
-classes (see, e.g. `lme4::plot.merMod`), however these methods are do not have
-consistent usage across packages [@lme4]. 
+does not include functions that adhere with the grammar of graphics [@lindia]. 
+Finally, many packages provide methods for plotting diagnostics based on their
+own model classes (see, e.g. `lme4::plot.merMod`), however these methods are do
+not have consistent usage across packages [@lme4]. `gglm` hence addresses a
+significant gap in functionality by creating a consistent framework for
+producing diagnostic plots across R packages and model types while adhering to 
+the grammar of graphics. 
 
 # References