Keep this open while you work

Cheat sheet

Which test?

One group vs a number

One-sample t-test

t_test(y ~ 1, mu = 4)

Two groups of different people

Two-group t-test

t_test(y ~ group)

Same people, twice

Paired t-test

t_test(y ~ Time, paired = TRUE)

Two numbers together?

Correlation

cor.test(x, y)

Two categories related?

Chi-square test

chisq.test(table(a, b))

Task	Code
Install a package (once)	`install.packages("rstatix")`
Load a package (each session)	`library(tidyverse)`
Read CSV / Excel / SPSS	`read_csv("data/file.csv")`, `read_excel()`, `read_sav()`
Look at data	`glimpse(d)`, `head(d)`, `View(d)`
Count missing values	`sum(is.na(d$x))`

Task	Code
Keep columns	`select(ID, Age)`, `select(-ID)`
Keep rows	`filter(Age >= 21, Gender == "Female")`
Sort	`arrange(desc(ExamScore))`
New column	`mutate(Change = T2 - T1)`
Two groups	`mutate(G = if_else(Age >= 21, "Older", "Younger"))`
Recode	`mutate(S = case_when(x == 0 ~ "Visual", x == 1 ~ "Auditory"))`
Order groups	`mutate(M = factor(M, levels = c("Low", "Medium", "High")))`
Summary per group	`group_by(G) %>% summarise(n = n(), m = mean(y, na.rm = TRUE))`
Count values	`count(G)`
Wide to long	`pivot_longer(cols = c(T1, T2), names_to = "Time", values_to = "Score")`
Keep the change	`d <- d %>% mutate(...)`

Task	Code
Middle	`mean(x, na.rm = TRUE)`, `median(x, na.rm = TRUE)`
Spread	`sd()`, `IQR()`, `min()`, `max()`
All at once	`describe()` (psych), `get_summary_stats(y, type = "mean_sd")` (rstatix)
z-scores	`mutate(z = (x - mean(x)) / sd(x))`
z-test	`z.test(x, mu = , sigma.x = )` (BSDA)

Chart	Code
Histogram	`ggplot(d, aes(x = y)) + geom_histogram(binwidth = 5)`
Bar chart	`ggplot(d, aes(x = g, fill = h)) + geom_bar(position = "dodge")`
Boxplot	`ggplot(d, aes(x = g, y = y)) + geom_boxplot()`
Scatter + line	`ggplot(d, aes(x = x, y = y)) + geom_point() + geom_smooth(method = "lm")`
Labels and look	`+ labs(title = , x = , y = ) + theme_classic()`
Save	`ggsave("chart.png", width = 6, height = 4, dpi = 300)`

Task	Code
Sample size	`pwr.t.test(d = 0.5, sig.level = .05, power = .80, type = "two.sample")`
Normality	`shapiro_test(y)`, `ggplot(d, aes(sample = y)) + stat_qq() + stat_qq_line()`
Outliers	`identify_outliers(y)`
Effect size (t-test)	`cohens_d(y ~ group)`
Chart label	`labs(subtitle = get_test_label(result, detailed = TRUE))`
Spearman	`cor.test(x, y, method = "spearman")`
Expected counts	`chisq.test(t)$expected`
Cramér's V	`cramer_v(t)`