Skip to content

You're working as a sports journalist at a major online sports media company, specializing in soccer analysis and reporting. You've been watching both men's and women's international soccer matches for a number of years, and your gut instinct tells you that more goals are scored in women's international football matches than men's. This would make an interesting investigative article that your subscribers are bound to love, but you'll need to perform a valid statistical hypothesis test to be sure!

While scoping this project, you acknowledge that the sport has changed a lot over the years, and performances likely vary a lot depending on the tournament, so you decide to limit the data used in the analysis to only official FIFA World Cup matches (not including qualifiers) since 2002-01-01.

You create two datasets containing the results of every official men's and women's international football match since the 19th century, which you scraped from a reliable online source. This data is stored in two CSV files: women_results.csv and men_results.csv.

The question you are trying to determine the answer to is:

Are more goals scored in women's international soccer matches than men's?

You assume a 10% significance level, and use the following null and alternative hypotheses:

: The mean number of goals scored in women's international soccer matches is the same as men's.

: The mean number of goals scored in women's international soccer matches is greater than men's.

# Start your code here!
library(tidyverse, quietly = TRUE)

#__________________________
# Exploratory data analysis
#__________________________

# Load files

men <- read_csv("men_results.csv")
women <- read_csv("women_results.csv")

str(men)
head(men)
summary(men)
paste("---------------------------------------------")
str(women)
head(women)
summary(women)
# Look at unique values in the columns naming teams and types of tournaments.
# We might find some problems.
# Men first:

for (col in names(men[,c(3,4,7)])) {
  unique_values <- unique(men[[col]])
  print(paste("Unique values in column", col, ":"))
  print(unique_values)
  print("-----------------------------")
}
# Women next:

for (col in names(women[,c(3,4,7)])) {
  unique_values <- unique(men[[col]])
  print(paste("Unique values in column", col, ":"))
  print(unique_values)
  print("-----------------------------")
}
# There do not seem to be any immediate problems.
# There are 311 teams in home, 306 teams in each dataset for away.
# There are 141 tournament names.
# the men dataset has more than 10x the rows of women due to starting in the 1800's.

#__________________________
# Filter
#__________________________

men_filtered <- men %>% 
filter(tournament == "FIFA World Cup",
	date > "2002-01-01")

women_filtered <- women %>% 
filter(tournament == "FIFA World Cup",
	   date > "2002-01-01")

str(men_filtered)
paste("-----------------------------")
str(women_filtered)
#__________________________
# Create the new variable
# for hypothesis testing
#__________________________

men_calc <- men_filtered %>% 
mutate(goals_scored = home_score + away_score)

summary(men_calc)

paste("--------------------------------------------------")

women_calc <- women_filtered %>% 
mutate(goals_scored = home_score + away_score)

summary(women_calc)
#__________________________
# Choose the correct hypothesis
# test.
#__________________________

# Plot the distribution of the outcome variable we are looking at.
# The goals scored variable in each dataset has strong positive skewness.

paste("The mean of men's goals scored is ", mean(men_calc$goals_scored), ", and the standard deviation is ", sd(men_calc$goals_scored)
	 )

paste("--------------------------------------------------")

paste("The mean of women's goals scored is ", mean(women_calc$goals_scored), ", and the standard deviation is ", sd(women_calc$goals_scored)
	 )

ggplot(men_calc, aes(goals_scored)) + 
geom_boxplot() + 
coord_flip() + 
labs(title = "Men's Soccer",
	subtitle = "Distribution of the Goals Scored Variable",
	 x = "Goals Scored"
	) + 
theme_linedraw()

ggplot(men_calc, aes(goals_scored)) + 
geom_density(color = "red",
			size = 2
			) +  
labs(title = "Men's Soccer",
	subtitle = "Distribution of the Goals Scored Variable",
	 x = "Goals Scored"
	) + 
theme_linedraw()

ggplot(women_calc, aes(goals_scored)) + 
geom_boxplot() + 
coord_flip() + 
labs(title = "Women's Soccer",
	subtitle = "Distribution of the Goals Scored Variable",
	 x = "Goals Scored"
	) + 
theme_linedraw()

ggplot(women_calc, aes(goals_scored)) + 
geom_density(color = "red",
			size = 2
			) +  
labs(title = "Women's Soccer",
	subtitle = "Distribution of the Goals Scored Variable",
	 x = "Goals Scored"
	) + 
theme_linedraw()



# The appropriate test could be the Wilcoxon rank-sum test given that the data are not normally distributed.

test_result <- wilcox.test(women_calc$goals_scored,
					  men_calc$goals_scored, 
					 alternative = "greater"
					 )

p_val <- test_result$p.value
result <- if_else(p_val <= 0.01, "reject", "fail to reject")

result_df <- data.frame(p_val, result)

result_df