Getting Started with psych350lab

Installation, core concepts, and a quick tour

What This Package Does

psych350lab helps instructors and TAs build statistics lab materials in R that replicate SPSS/jamovi output. The package provides four types of output for each statistical test:

1. Worksheet output — markdown-formatted results with a KEY/BLANK toggle for answer keys vs. student fill-in-the-blank worksheets (rendered via Quarto to .docx)
2. APA tables — publication-ready tables as flextable objects
3. Statistical write-ups — complete APA-style interpretation text
4. Homework checkers — interactive HTML widgets (via webexercises) where students enter answers and get immediate feedback

Installation

Install both packages from GitHub using pak:

install.packages("pak")

pak::pak("emmarshall/psych350data")
pak::pak("emmarshall/psych350lab")

GitHub Authentication (First Time Only)

install.packages(c("gitcreds", "usethis"))

usethis::create_github_token()
gitcreds::gitcreds_set()
usethis::git_sitrep()

Core Workflow

Every lab section follows the same five-step pattern, regardless of the statistical test:

library(psych350lab)
library(psych350data)
library(dplyr)
library(ggplot2)

# ── Step 1: Load and prepare data ────────────────────────
data(superman, package = "psych350data")

# ── Step 2: Run the analysis ─────────────────────────────
result <- corr_answers(
  data = superman,
  var1 = "clark_height_in",
  var2 = "rt_audience_score"
)

# ── Step 3: Format for worksheets (KEY or BLANK) ─────────
key_output <- format_corr_results(
  rh_name = "RH1",
  vars = c("clark_height_in", "rt_audience_score"),
  corr_results_list = result,
  Key = TRUE
)

# ── Step 4: Create APA table ─────────────────────────────
create_apa_corr_table(
  data = superman,
  vars = c("clark_height_in", "rt_audience_score"),
  var_labels = c("1. Actor Height (in)", "2. Audience Score"),
  KEY = TRUE
)

# ── Step 5: Generate write-up ────────────────────────────
writeup <- apa_corr_writeup(
  corr_results_list = result,
  var1_label = "actor height",
  var2_label = "audience score",
  hypothesis = list(
    direction = "positive",
    rh_text = "taller actors would receive higher audience scores"
  ),
  include_descriptives = TRUE
)

The KEY/BLANK System

Most display functions have a KEY argument (sometimes called Key) that toggles between two modes:

• KEY = TRUE — fills in all values (instructor answer key)
• KEY = FALSE — shows blanks for students to fill in

This lets you generate both versions from the same Quarto source file using a params toggle:

# In your .qmd YAML header:
params:
  hide_answers: true

# In a code chunk:
format_corr_results(
  rh_name = "RH1",
  vars = c("clark_height_in", "rt_audience_score"),
  corr_results_list = result,
  Key = !params$hide_answers
)

Available Analyses

The package follows the lab course progression. Each analysis type has its own detailed guide:

Lab Topic	Analysis	Guide
Bivariate	Pearson’s correlation	Correlations
Bivariate	Chi-square (2×2)	Chi-Square (2×2)
Group differences	One-way ANOVA (BG & WG)	One-Way ANOVA
Group differences	K-group one-way ANOVA	K-Group ANOVA
Categorical	K-group chi-square (k×2)	K-Group Chi-Square
Factorial	2×2 factorial ANOVA (BG, MG, WG)	Factorial ANOVA
Multivariate	Multiple linear regression	Regression

Function Naming Conventions

Functions follow a consistent naming pattern:

Pattern	Purpose	Example
*_answers()	Run analysis, return raw results	corr_answers(), bg_anova_answers()
format_*_results()	Markdown for worksheets (KEY/BLANK)	format_corr_results()
create_*_table()	APA flextable	create_apa_corr_table()
create_*_checker()	Interactive HTML checker	create_corr_checker()
apa_*_writeup()	Full APA write-up text	apa_corr_writeup()
apa_inline_*()	Inline APA statistic	apa_inline_r()
format_*()	Format a single value	format_p_value(), format_stat()

Formatting Utilities

These helper functions format individual values for display:

format_p_value(0.023)                         # ".023"
format_p_value(0.0004)                        # "< .001"
format_p_value(0.023, include_p = TRUE)       # "p = .023"
format_stat(3.456)                            # "3.46"
format_stat(0.456, remove_leading_zero = TRUE) # ".46"
format_r(0.456)                               # ".46"
format_effect(0.35)                           # ".35"
format_int(25.4)                              # "25"
p_to_stars(0.003)                             # "**"
key_or_blank(0.45, KEY = TRUE, format_fn = format_r) # ".45"
key_or_blank(0.45, KEY = FALSE)               # "______"

SPSS-Style Plots

theme_SPSS() and number_SPSS replicate the SPSS chart appearance:

ggplot(superman, aes(x = clark_height_in, y = rt_audience_score)) +
  geom_point(
    shape = 21, fill = "#1192E8", color = "black",
    size = 2.5, stroke = 0.75, alpha = 0.7
  ) +
  theme_SPSS() +
  scale_x_continuous(labels = number_SPSS) +
  scale_y_continuous(labels = number_SPSS) +
  labs(x = "Actor Height (inches)", y = "Audience Score")

Typical Lab File Structure

A complete lab consists of three files:

labs/lab-07/
├── lab07-setup.R          # Data prep + all analyses
├── lab07_worksheet.qmd    # Quarto → Word worksheet
└── lab07_checker.qmd      # Quarto → HTML homework checker

The setup file runs all analyses and stores results in objects. The worksheet and checker files source the setup file and display the results using the appropriate output functions.

See the Correlations guide for a complete setup file example.