R Project
Annabelle Zhu
2023-09-17
Exploring the Relationship Between IQ Scores, Economic Development, and Geographical Factors
1. Abstract
This study is an in-depth exploration of the relationships between IQ, economic factors, and geography on a global scale. The research was conducted using advanced statistical techniques on international datasets in order to gain insight into the distribution of global intelligence and its association with productivity and progress.
The study was guided by five key research questions that aimed to uncover the intricate relationships between IQ, economic factors, and geography. The findings of the study produced some fascinating insights, including disparities in average IQ between Asia, Europe, and other regions, as well as a moderate tendency for national GDP to develop in parallel with IQ.
While the study cannot definitively establish causality between these factors, the insights gained can inform evidence-based approaches for optimizing societal potentials and developing targeted learning investments tailored to specific populations. The study acknowledges the complexity involved in tracing influences and offers a quantitative profile of human intelligence as an aggregate resource. The findings hold relevance across multiple domains, including research, policy, business, and broader society. They pave the way for continued exploration of the relationship between IQ and changing global circumstances, incorporating additional socioeconomic indicators to gain a deeper understanding of this dynamic interface.
In short, this study is an important contribution to our understanding of the relationships between IQ, economic factors, and geography on a global scale. The insights gained from this research offer new perspectives for optimizing societal potentials and developing evidence-based approaches to education and strategic planning. The study’s findings hold significant relevance across multiple domains and provide a basis for continued exploration of this dynamic field.
2. Introduction
Human intelligence has been a subject of interest and analysis for a long time, and the correlation between IQ and societal outcomes is a topic that continues to intrigue researchers. The question of how variations in human intellect correlate with divergences in economic standing between populations worldwide remains an open question. Understanding the relationship between IQ, education, prosperity, and societal development is of crucial importance for policymakers, researchers, and citizens alike.
IQ scores provide a standardized metric for a comparative examination of intellectual abilities, albeit an imperfect singular assessment acknowledging cultural influences. This study aims to analyze links between average IQ scores, GDP trends, and population-level factors using comprehensive international datasets. Population IQ testing data were compiled from over 1.5 million samples internationally, and GDP statistics from the World Bank allow consideration of how cognitive factors interface with productivity and material conditions. Current population sizes factor in the comparison to full national distributions.
By bringing these diverse sources into a single analytical framework, this research aims to offer an improved understanding of benefiting future education strategies, economic planning, and initiatives maximizing each population’s human potential given local environments and realities. The study spans over 60 years of data, and patterns are identified across nations and inhabited continents. This research provides new perspectives supporting evidence-based approaches to nurturing progress for all.
The ultimate goal of this study is to offer insights applicable to strengthening human potential on scales, paving the way for informed decision-making that promotes social, economic, and cultural development. By identifying correlations between IQ, education, prosperity, and societal development, researchers and policymakers can gain a better understanding of how to maximize human potential within their local environments and realities. The study’s findings have significant implications for future education strategies, economic planning, and initiatives aimed at promoting progress and improving the quality of life for all.
clean and annotated code
During the data wrangling process, the first thing I did is to remove the countries where their country names are empty, so I used the code”filter(country != ““)” and also make sure that the iq are not empty. Then, I encountered challenges related to inconsistent country names across the datasets, necessitating manual standardization to ensure uniformity. This involved rectifying formatting variations, such as adjusting capitalization and removing leading or trailing spaces. Additionally, addressing numerous missing values in the GDP dataset required the implementation of robust data imputation techniques, including mean imputation, regression-based imputation, and multiple imputation. The data cleaning process involved meticulous attention to detail, including the removal of leading or trailing spaces, conversion of uppercase to lowercase, and ensuring consistent formatting across all datasets. Outlier detection and data validation were also integral parts of the data cleaning process, ensuring the accuracy and reliability of the datasets.
The results of the data wrangling process yielded a unified and standardized dataset, facilitating seamless analysis and interpretation. The meticulous data cleaning and imputation techniques employed ensured the integrity and reliability of the datasets, laying a robust foundation for subsequent analysis and interpretation. Eventually, the merged dataset would be 124 countries obtained average GDP with total sample size of 1.5 million.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 4.2.3## Warning: package 'ggplot2' was built under R version 4.2.3## Warning: package 'tibble' was built under R version 4.2.3## Warning: package 'tidyr' was built under R version 4.2.3## Warning: package 'readr' was built under R version 4.2.3## Warning: package 'purrr' was built under R version 4.2.3## Warning: package 'dplyr' was built under R version 4.2.3## Warning: package 'stringr' was built under R version 4.2.3## Warning: package 'forcats' was built under R version 4.2.3## Warning: package 'lubridate' was built under R version 4.2.3## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(ggplot2)
library(maps)## Warning: package 'maps' was built under R version 4.2.3##
## Attaching package: 'maps'
##
## The following object is masked from 'package:purrr':
##
## map# load the dataset
iq_data <- read.csv("Data/International_IQ.csv")
gdp_data <- read.csv("Data/gdp.csv")
population_by_country <- read.csv("Data/population_by_country_2020.csv")iq_data <- iq_data %>%
rename(IQ = `Average.IQ`)
colnames(iq_data) <- tolower(colnames(iq_data))# Data cleaning - remove the NA value of the country name
iq_data_cleaned <- iq_data %>%
filter(country != "") %>%
filter(iq != "")gdp <- gdp_data %>%
rename(country = `Country.Name`)population_by_country <- population_by_country %>%
rename(country = `Country..or.dependency.`) %>%
rename(country_population = `Population..2020.`)population <- population_by_country[, c("country", "country_population")]total_sample <- sum(iq_data_cleaned$participants)
total_sample## [1] 1562167iq_data_cleaned <- left_join(iq_data_cleaned, population)## Joining with `by = join_by(country)`iq_data_cleaned <- iq_data_cleaned %>%
filter(country_population != "")Research Question 1: Is there a correlation between the proportion of participants in IQ tests and the average IQ score?
I hypothesize that there is a positive correlation between the proportion of participants in IQ tests and the average IQ score. This suggests that regions with a higher proportion of participants relative to their population size may have more accurate representations of their IQ distribution.
To answer this research question, I calculated the correlation coefficient between the proportion of participants and the average IQ score. Additionally, I visually find the relationship using a scatter plot, where the x-axis represents the proportion of participants and the y-axis represents the average IQ score.
But I get an error because the number of participants can be affected by the sheer population size of the region, so I’ll try to calculate the proportion of participants by dividing the number of participants (participants) by the total population. This gives us a normalized metric that represents the proportion of participants relative to the population size.]
iq_data_cleaned <- iq_data_cleaned %>%
mutate(participant_proportion = participants / country_population)
correlation <- cor(iq_data_cleaned$participant_proportion, iq_data_cleaned$iq)
correlation## [1] 0.3442114This means that as the number of participants increases, there is a tendency for the average IQ score to also increase.
# Visualize the relationship
ggplot(iq_data_cleaned, aes(x = participant_proportion, y = iq)) +
geom_point() +
geom_smooth(method = "lm", col = "red", formula = y ~ x) +
labs(title = "Relationship between Proportion of Participants and Average IQ Score",
x = "Proportion of Participants",
y = "Average IQ")
Research Question 2: What is the distribution of average IQ scores across most countries?
Understanding the distribution of average IQ scores across different countries provides valuable insights into the range and concentration of intelligence levels globally. I create a histogram to visualize the distribution.
By examining the shape, spread, and central tendency of the distribution, we can gain insights into the prevalence of high and low IQ scores, as well as the concentration of scores around the mean.
This information can help us understand the overall intelligence levels across different countries and identify potential areas for further investigation.
iq_data_cleaned %>%
ggplot(aes(x=iq)) +
geom_histogram(binwidth=1, fill="blue", color="black") +
labs(x="Average IQ", y="Frequency", title="Distribution of Average IQ Scores")
Research Question 3: Which continents have the highest average IQ scores? And do the average IQ scores differ significantly among the continents?
My hypothesis is there are significant differences in average IQ scores among the continents. Specifically, we expect to find varying average IQ scores across different continents, indicating that geographical factors play a significant role in shaping cognitive abilities.
By examining these differences, we seek to gain a deeper understanding of the complex interplay between geographical location and cognitive abilities, providing valuable insights for future research and policy considerations.
To do this, we can create a bar chart and a world map so that we can see the difference between continents more clearly; on the other hand, we can preform a Oneway ANOVA test, to determine if there exists a significant IQ difference between continents.
iq_data_mutated <- iq_data_cleaned %>%
mutate(continent = case_when(
country %in% c("Japan", "South Korea", "China", "Iran (Islamic Republic of)", "Singapore", "Mongolia", "Viet Nam",
"Sri Lanka", "Russian Federation", "Nepal", "India", "Malaysia", "Lebanon", "Egypt", "Thailand", "Myanmar", "Bangladesh", "Turkey", "United Arab Emirates", "Israel", "Jordan", "Iraq", "Cambodia", "Philippines", "Oman", "Lao People's Democratic Republic", "Uzbekistan", "Pakistan", "Saudi Arabia", "Kazakhstan", "Qatar", "Kuwait", "Indonesia", "Tajikistan", "Kyrgyzstan")
~ "Asia",
country %in% c("Germany", "Slovenia", "Switzerland", "Austria", "Bulgaria", "Belgium", "Netherlands", "Luxembourg", "Italy", "Czechia", "Russia", "France", "Spain", "Georgia", "Poland", "Armenia", "Hungary", "Cyprus", "Norway", "Iceland", "Croatia", "Montenegro", "Estonia", "Denmark", "Latvia", "Greece", "Finland", "Sweden", "Portugal", "Ireland", "United Kingdom", "Serbia", "Belarus", "Andorra", "North Macedonia", "Bosnia and Herzegovina", "Lithuania", "Albania", "Moldova (Republic of)", "Ukraine", "Romania", "Slovakia") ~ "Europe",
country %in% c("United States", "Canada", "Mexico", "Puerto Rico", "Cuba", "Trinidad and Tobago", "Jamaica", "Costa Rica") ~ "North America",
country %in% c("Australia", "New Zealand") ~ "Oceania",
country %in% c("Brazil", "Argentina", "Chile", "Venezuela (Bolivarian Republic of)", "Colombia", "Peru", "Ecuador", "Uruguay", "Bolivia (Plurinational State of)", "Paraguay") ~ "South America",
country %in% c("South Africa", "Nigeria", "Kenya", "Cameroon", "Namibia", "Senegal", "Angola", "Cote d'Ivoire", "Ghana", "Uganda", "Democratic Republic of the Congo", "Mali", "Burkina Faso", "Togo", "Madagascar", "Gabon", "Benin", "Guatemala", "Panama", "Dominican Republic", "El Salvador", "Honduras", "Nicaragua", "Tunisia", "Algeria", "Morocco", "Ethiopia", "Congo (Democratic Republic of the)", "Congo") ~ "Africa"
))world_map <- map_data("world")
merged_data <- merge(world_map, iq_data_cleaned, by.x = "region", by.y = "country", all.x = TRUE)
ggplot() +
geom_polygon(data = merged_data, aes(x = long, y = lat, group = group, fill = iq)) +
scale_fill_gradient(low = "blue", high = "red", na.value = "white") +
labs(title = "Distribution of Average IQ Scores Across most Countries",
fill = "IQ")
avg_iq_by_continent <- iq_data_mutated %>%
group_by(continent) %>%
summarize(Average_IQ = mean(iq)) %>%
ggplot(aes(x = continent, y = Average_IQ, fill= continent)) +
geom_bar(stat = "identity") +
labs(x = "Continent", y = "Average IQ", title = "Average IQ by Continent")
avg_iq_by_continent
anova_result <- aov(iq ~ continent, data = iq_data_mutated)
summary(anova_result)## Df Sum Sq Mean Sq F value Pr(>F)
## continent 5 1384 276.74 27.18 <2e-16 ***
## Residuals 116 1181 10.18
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1The results of the one-way ANOVA revealed a significant difference in average IQ scores among the continents (p < 0.001). This indicates that there are significant variations in average IQ scores across different continents.
Question 4: What are the top 10 countries with the highest average IQ scores? And is there a correlation between average GDP and average IQ in the top 10 countries with the highest IQ scores?
Understanding the potential correlation between intelligence levels and economic productivity is essential for gaining insights into the factors that contribute to a country’s economic development.
We hypothesize that there exists a significant relationship between the average IQ scores and the GDP of the top 10 countries with the highest IQ scores.
First we should do the data cleaning. Calculate the average GDP from the years 1985 to 2020 for each country
gdp_avg <- gdp %>%
select(-ncol(.)) %>%
rename_with(~ gsub("X", "", .), starts_with("X"))gdp_avg <- gdp_avg%>% mutate(avg_GDP = rowMeans(select(., `1985`:`2020`), na.rm = TRUE))country_gdp_avg <- left_join(iq_data_mutated, gdp_avg, by = "country") %>%
filter(!is.na(Code))# Making a bar plot to find out which 10 countries have the highest average IQ.
top_10_countries <- iq_data_cleaned %>%
arrange(desc(iq)) %>%
head(10)
ggplot(top_10_countries, aes(x = reorder(country,iq), y = iq, fill=country)) +
geom_bar(stat = "identity") +
coord_flip() +
labs(title = "Top 10 Countries with Highest Average IQ Scores",
x = "Country",
y = "Average IQ",
caption = "Source: Kaggle: Countries and what they say")
top_10_countries_gdp <- left_join(top_10_countries, gdp_avg, by = "country") %>%
filter(!is.na(Code))
ggplot(top_10_countries_gdp,aes(x = avg_GDP, y = iq)) +
geom_point() +
geom_smooth(method = "lm", formula = y ~ x) +
labs(title = "Relationship between Average IQ and Average GDP in top 10 highest IQ countries",
x = "Average GDP",
y = "Average IQ")
And to answer the other part of this research question, we can calculate the correlation coefficient between average GDP and average IQ in the top 10 countries.
correlation <- cor(top_10_countries_gdp$avg_GDP, top_10_countries_gdp$iq)
correlation## [1] 0.4914852The correlation coefficient between average GDP and average IQ in the top_10_countries_gdp dataset. The resulting correlation coefficient is 0.4914.
We can say that there is a moderate positive correlation (0.4914) between average GDP and average IQ in the top 10 countries.
This suggests that as the average GDP increases, there is a tendency for the average IQ to also increase, although the relationship is not extremely strong.
Question 5: Is there a relationship between the average IQ and average GDP within each continent?
# Group the data by continent and calculate the correlation
correlation_by_continent <- country_gdp_avg %>%
group_by(continent) %>%
summarize(correlation = cor(avg_GDP, iq, use = "complete.obs"))
correlation_by_continent## # A tibble: 6 × 2
## continent correlation
## <chr> <dbl>
## 1 Africa 0.347
## 2 Asia 0.603
## 3 Europe 0.288
## 4 North America 0.630
## 5 Oceania 1
## 6 South America -0.281ggplot(data = country_gdp_avg, aes(x = avg_GDP, y = iq)) +
geom_point() +
geom_smooth(method = "lm", formula = y ~ x) +
facet_wrap(~continent, scales = "free") +
labs(title = "Relationship between Average IQ and Average GDP by Continent",
x = "Average GDP",
y = "Average IQ")## Warning in qt((1 - level)/2, df): NaNs produced## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
The findings of our study on the relationship between average IQ and average GDP across different continents have significant implications for policymakers, economists, educators, and researchers.
The correlations observed are nuanced and highlight the need for region-specific approaches to address IQ and GDP. These insights can inform tailored educational and economic policies, fostering development initiatives that are aligned with the specific needs of each continent.
Conlusion & Discussion
The continent-specific correlations uncovered in our study open avenues for future research endeavors, including longitudinal studies tracking the evolution of average IQ and average GDP over time within different continents. Collaborative efforts between researchers from diverse disciplines can enrich the understanding of the interplay between IQ and economic growth, fostering a holistic approach to addressing economic challenges.
The future of research in this domain holds immense potential for unraveling the intricate connections between IQ and economic prosperity. Embracing technological advancements and complex data analytics techniques can enhance the depth and breadth of research, unlocking new avenues for exploring complex interactions and patterns within the data.
The impact of our results extends beyond academia, offering insights for policymakers, educators, and researchers. The future of research in this domain is poised to shape evidence-based interventions, policy recommendations, and global initiatives aimed at fostering cognitive and economic development, ultimately contributing to the advancement of societies and economies worldwide.
Overall, our study highlights the importance of region-specific approaches in addressing IQ and economic dynamics. By tailoring policies and initiatives to the specific needs of each continent, we can ensure that development efforts are effective and sustainable. The future of research in this domain holds immense promise, and we look forward to continued collaborations and advancements in understanding the complex interplay between IQ and economic growth.