Is the distribution of this year’s August ice-cream purchases different from last year’s expected distribution of 20% chocolate, 20% strawberry, 20% mango, and 40% vanilla?
library(readxl)
A4Q1 <- read_excel("C:/Users/nehab/OneDrive/A5221/Assignment 4/A4Q1.xlsx")
observed <- table(A4Q1$flavor)
observed
##
## Chocolate Mango Strawberry Vanilla
## 87 32 57 74
barplot(
observed,
main = "Ice Cream Purchases",
xlab = "Ice Cream Flavor",
ylab = "Count",
col = rainbow(length(observed))
)
expected <- c(.20, .20, .20, .40)
chi_result <- chisq.test(x = observed, p = expected)
chi_result
##
## Chi-squared test for given probabilities
##
## data: observed
## X-squared = 41.6, df = 3, p-value = 4.878e-09
w <- sqrt(as.numeric(chi_result$statistic) / sum(observed))
w
## [1] 0.4079216
# I updated the interpretation of the effect size.
# My original answer reported Cohen's W = .41 but did not explain that it was moderate.
A Chi-Square Goodness of Fit test showed that the distribution of this year’s ice-cream purchases was significantly different from last year’s expected distribution, X-squared(3, N = 250) = 41.60, p < .001. The difference was moderate, Cohen’s W = .41.