Stat 240: Exploring the Relationship between Economic Growth and Unemployment 2025-04-08 Connor Barry, Vivian O’Brien, Ethan Peterson, David Jozwik Introduction The unemployment rate and GDP growth rate are both critical indicators of a nation’s current economic performance and reflect the efficiency of the labor market and size and growth of the economy. The unemployment rate measures the share of workers in the labor force who do not currently have a job but are actively looking for work. The GDP growth rate measures annual percentage increase (or decrease) in GDP, representing the total value of goods and services produced within a country during a specific period, with positive growth indicates expansion, while negative growth indicates contraction. Research Question We are interested in exploring the relationship between both of these variables, as they are both key indicators of macroeconomic health and can offer insights into broader economic dynamics.This leads into our research question: “Does a higher unemployment rate correspond to lower GDP growth among countries in the year 2016?” What we found in our data was that a simple linear regression revealed a statistically significant negative relationship: for every one percentage point increase in the unemployment rate, GDP growth decreased by approximately 0.53 percentage points. The one-sided p-value was 0.0082, providing strong evidence against the null hypothesis and supporting our claim that higher unemployment is associated with weaker economic growth. This aligns with economic theory and visual patterns observed in the plot, which shows a clear downward trend between the two variables. Background Data Sources Raw data: The owner of the data is Adil Shamim whereas the provider of the data is the IMF, World Bank, OECD, Government Economic Reports, UN, and World Economic Forum. Data was sourced from the above- mentioned organizations, compiled and normalized for each country. For years where data gaps exist, estimates were made based on available historical data trends and projections. Variables Country: The name of the country for which the economic data is reported. Year: The year for which the data is recorded, we looked at the year 2016. EG (Economic Growth Rate) (%) : The percentage of the country’s labor force that is not currently employed. It is a key indicator of economic health and labor market efficiency. UR (Unemployment Rate) (%): The percentage of the country’s labor force that is unemployed but actively seeking employment. It is a key indicator of economic health and labor market efficiency. Data Structure: Each row in our dataset represents a specific country in the year 2016 and includes key economic indicators such as GDP, GDP growth rate (EG), unemployment rate (UR), and inflation rate (IR). Our sample consists of approximately 20 countries, providing a small but diverse cross-section of national economies. Despite the limited sample size, each row aggregates data at the country level, reflecting nationwide economic performance and labor market conditions. This makes the data highly informative for macroeconomic comparisons. In our analysis, we focus on the relationship between unemployment rate and GDP growth, using a simple linear regression to assess whether higher unemployment is associated with slower economic growth Data Cleaning and Processing: To ensure relevance and consistency in our analysis, the dataset was filtered to include only observations from the year 2016, providing a cross-sectional snapshot of global economic performance. We excluded years outside this time frame to focus our analysis on contemporaneous relationships between macroeconomic indicators across countries. Additionally, countries with missing or incomplete values for key variables such as GDP growth (EG) or unemployment rate (UR) were removed from the dataset to maintain accuracy in regression modeling. The cleaned dataset includes observations for countries with fully reported economic indicators, such as those displayed in the reference table. This subset includes countries like the USA, Australia, Canada, and others, each with complete values for GDP, GDP growth rate, unemployment rate, and inflation rate. Statistical Analysis To explore our hypothesis that higher unemployment is associated with lower GDP growth, we used a linear regression model with GDP growth rate as the outcome variable and unemployment rate as the predictor. The resulting scatterplot (see graph below) visualizes this relationship, with each point representing a country and the blue regression line indicating the estimated negative trend. Our final regression showed a negative slope (–0.5305), meaning that for every one-point increase in unemployment, GDP growth tends to fall by about half a percentage point. The p-value for a one-sided test (0.0082) suggests this result is statistically significant, supporting our hypothesis. By focusing only on countries in 2016 with fully reported data, we improve the reliability of this conclusion while acknowledging the limited sample size. The parameter of interest in this study is the true population slope ( β 1) from the linear regression equation: Where EG represents the GDP growth rate, UR represents the unemployment rate, and β 1 is the slope, which indicates the change in GDP growth for a one-unit change in unemployment rate. We are examining , the relationship between UR and EG through a one - tailed hypothesis test. Our chosen confidence interval is 95%, corresponding to an alpha of 5%, or .05. Whether our p-value is above or below this value will determine if we reject or fail to reject our null hypothesis. In this test, we assume a null and alternate hypothesis as such: We assumed a one - tailed test as we are investigating whether economic growth has an inverse relationship with unemployment rate, as EG indicates an improving economy, while UR indicates a more ailing economy. To test our hypothesis, we constructed our test statistic based on this equation: EG = + × UR + β 0 β 1 ε EG β 1 : = 0 || : < 0 H 0 β 1 H α β 1 Where is our point estimate, is the value of the point estimate assumed by our null hypothesis (0), and is the standard error of the point estimate. our statistic test used 17 degrees of freedom, which is our number of data points minus two. The 95% confidence interval for the slope of the regression line is [ − 0.944, − 0.117]. This interval suggests that the true slope is likely between − 0.944 and − 0.117. Since zero isn’t included in this range, it provides further support for the conclusion that there is a significant negative relationship between the unemployment rate and GDP growth. Discussion Further Interpretation The results of our regression analysis indicate that the p-value (0.00817) is less than the significance level of 0.05. This provides sufficient evidence to reject the null hypothesis and conclude that there is a statistically significant negative relationship between economic growth and the unemployment rate in different countries. Specifically, the negative slope (–0.5305) suggests that for every one-point increase in the unemployment rate, economic growth tends to decline by approximately half a percentage point. For non-statistical readers, the p- value represents the probability of observing a strong inverse relationship between economic growth growth and unemployment rate. A p-value as small as 0.0082 suggests it is highly likely that economic growth growth and unemployment rate are related. By limiting the analysis to the economy in 2016, we increase the reliability of this conclusion, although we acknowledge the limitation of a smaller sample size. The direction and significance of the relationship support our hypothesis that higher unemployment tends to relate to a weaker economic growth performance. Shortcomings of the Analysis Economic Events: The analysis may be influenced by major economic disruptions such as the COVID-19 pandemic, which significantly affected both economic growth and unemployment rates in 2020. These events can create anomalies in the data, potentially affecting the assumption of a consistent relationship across years. Unaccounted Variables: This analysis does not include other important economic factors that may influence economic growth, such as inflation, interest rates, government spending, or international trade. Including these variables in future models would capture a better representation of different countries’ economic growth. Time Span: The study focuses only on data from 2016 and includes specific countries. While this enhances reliability, it limits generalizability. A broader time frame or more countries could reveal whether the negative relationship between economic growth and unemployment is consistent for other economies. Additional Question and Recommendations for Future Research Generalizing to Other Countries: The methodology and regression approach used in this analysis can be easily adapted to explore the relationship between economic growth and unemployment in countries beyond those included in the 2016 dataset. Applying the same model to other years could help assess whether the negative relationship holds consistently across different economies. t = pe − p H 0 se ( pe ) pe p H 0 se ( pe ) Analyzing Changes Across Years: How did the economic growth and unemployment behave during previous global recessions or recovery periods? Expanding the Dataset: Incorporating additional variables such as inflation, interest rates, labor, and government spending could help us understand how economic growth is affected further. References link to data: https://www.kaggle.com/datasets/adilshamim8/economic-indicators-and-inflation? resource=download (https://www.kaggle.com/datasets/adilshamim8/economic-indicators-and-inflation? resource=download)