psyteachr/glossary

Glossary of terms

glossary

The goal of {glossary} is to provide a glossary of statistical and R coding terms used in the PsyTeachR books that can be accessed offline.

Installation

You can install the development version of glossary from GitHub with:

# install.packages("devtools")
devtools::install_github("psyteachr/glossary")

Access the book offline

glossary::open_glossary()

You can use the glossary to define terms in a markdown document. By default, the term links to the online glossary and shows a short definition when you hover over the link. You can change what text displays in the link, what the short definition displays, whether the link opens the online glossary, and whether it’s added to the glossary table.

• glossary("alpha"): alpha
• glossary("beta", display = "β"): β
• glossary("power", def = "custom definition of power"): power
• glossary("probability", def = ""): probability
• glossary("p-value", link = FALSE): p-value
• glossary("replicability", add_to_table = FALSE): replicability

You can include a glossary table at the end of a document that includes all of the terms you defined above.

glossary_table()

term	definition
alpha	(stats) The cutoff value for making a decision to reject the null hypothesis; (graphics) A value between 0 and 1 used to control the levels of transparency in a plot
beta	The false negative rate we accept for a statistical test.
p value	The probability of seeing an effect at least as extreme as what you have, if the real effect was the value you are testing against (e.g., a null effect)
power	custom definition of power
probability	A number between 0 and 1 where 0 indicates impossibility of the event and 1 indicates certainty

Compilation

If you add or change any terms in the glossary, you will need to recompile it. Here are two ways to do this, either through a running R session or from a terminal prompt.

Recompile using R

From the R console, with the working directory set to the root directory of the glossary repository, type

source("_makefile.R")

Recompile from the terminal

In a terminal window, navigate to the directory containing the glossary files and type

make

This will invoke the commands in Makefile.

glossary's Issues

NEW: pivot functions

Could be useful to have an explanation about pivot functions pivot_longer() and pivot_wider(), linking to the definitions for long and wide.

NEEDS: probability

1. Add .Rmd
2. TBD

NEW: scientific notation

See https://courses.lumenlearning.com/waymakerintermediatealgebra/chapter/read-writing-scientific-notation-2/

link to Ch 4 in ADS

NEW: compile

link to knit

NEW: double brackets

pseudoreplications

NEW: Outlier

Would be good to include maybe a paper on detecting outliers as well

NEW: aspect ratio

New: static

in the context of Shiny content

EDIT: Q-Q plot

Typo: "noraml"

NEW: Environment

Would be useful to define the environment - maybe linking to definitions for objects and variables.

preregistration term

TDB

NEW: Conflict

TBD - would be good to add something about conflicts in function names

NEW: between-subjects and within-subjects

NEW: tied ranks

TBD

Suggested: "when two or more values/observations within a variable are identical and assume the same rank order value"

NEW: Cronbach's alpha

Link from "alpha" also

pseudoreplications

NEEDS: replicability

Needs more detail and links to papers/resources

NEW: directory

NEW: joins

Overview of joins and link to :

• binding joins
• filtering joins
• mutating joins

Also separate entries for semi_join, anti_join, left_join, right_join, inner_join, outer_join that link to their join type.

continuous data

1. Add .Rmd
2. TBD

NEW: permutation tests

Could be useful to define permutation tests and maybe link to Monte Carlo?

NEEDS: alpha

Secondary definition for alpha in the context of ggplot

Secondary definition for Cronbach's alpha

idea: function to display definition in text rather than term

Might be an idea to have a function that displays the definition of a word in the actual body of text in a paragraph, rather than the word. Say for example:

"Degrees of Freedom is defined as r glossary_show_definition("degrees of freedom")

Which would knit as:

"Degrees of Freedom is defined as The number of observations that are free to vary to produce a known outcome."

though might need to put a tolower() around it. That way people would only need to update the definition in the glossary and it would update in other places it is used and remove any inconsistencies. Might not be frequently used but could be handy.

NEEDS: hypothesis

1. Add RMD
2. TBD

Normal Distribution Term

1. to add .Rmd file
2. TBD

NEW: model

A description of what a model does/its aims. Could link to examples like glm and glmm.

NEEDS: binding joins

Example needs explanatory text

NEEDS: R markdown

Needs an example

NEEDS: list

Needs an example

NEW: container

Definition of term "list" uses the term "container" (container data type) which links to C category in Glossary but definition for it doesn't exist yet.

NEEDS: bivariate normal

Needs an example

NEW: formula

Define formula in the sense of R's object type?

NEEDS: beta

Example or extended definition

Tibble term

1. Rmd Needed
2. TBD

NEW: NA

explain how to remove NAs for calcs like mean, sd, cor

NEW: Boolean Expression

TBD - add a section on what is a boolean expression and give examples.

New: Degrees of Freedom

To be Added:

Propose: the number of observations that are free to vary to produce a known outcome

Basic example: If you run 5 people and ask them their age, and you know the mean age of those 5 people is 20.1, then four of those people can have any age but the 5th person must have a specific age to maintain the mean of 20.1. So the mean age here is the known outcome and four peoples age can vary freely (so your degrees of freedom is 4) but one person's age must be a specific number to maintain the known outcome.

Alternatively we could use something from the t-distribution and show how changing the df changes the probability of the outcome on the t-distribution.

EDIT: whitespace

Additional comment or example about using whitespace in APA write-up. e.g.

with whitespace: t(df) = t-value, p = p-value, d = d-value

without whitespace: t(df)=t-value, p=p-value, d=d-value

NEW: Data Visualisation

ToBeAdded.

Suggestion:

Data Visualisation is the skill of creating and interpreting plots and figures from data.

two-tailed term

1. add Rmd
2. TBD

NEW: literal

in the context of backslashes creating literal versions of symbols that have a special meaning.

NEW: html tag

NEW: dynamic

In the context of Shiny content

discrete data

1. Add .Rmd
2. TBD

NEW: dependency

Packages that are required to run another package. R should automatically install these for you when you install a new package.

NEW: library

TBD - would be good to add something about the idea of the library and loading packages into the library.

NEEDS: power

1. add rmd
2. TBD

EDIT: pseudoreplication

Typo: "the conventional a t-test" should be "a conventional t-test"

NEW: directory