QM1 2025 Lab Final
Fall 2025
Assignment policy: Students may discuss but must complete their work independently. Submit a single Excel file with clearly separated answer sheets for each section and back-end work. Clarity and presentation count! The final lab project is due Friday, Dec 5, 2025, 2PM PT.
Grading:
Section 1 – 10 points
Section 2 – 30 points
Section 3 – 30 points
Section 4 – 20 points
Overall Presentation – 10 points
Introduction
How we spend our time each day is fundamental to economic and social life. In this project, you will explore international time use data – how people allocate their day among work, leisure, personal care, and other activities – and analyze how these patterns relate to factors like gender and income. Time use is a key lens for understanding labor supply, work-life balance, and gender equality. For instance, differences in paid work and unpaid household work between men and women can shed light on persistent gender roles, while the amount of leisure time people enjoy may correlate with a country’s level of economic development and well-being. You will work with real data from the Organization for Economic Co-operation and Development (OECD) and the World Bank to gain insights into questions such as: How do people spend their daily time? Who enjoys the most leisure time, and how is it measured? Is there a link between a country’s income (GDP per capita) and how much its people work? And do gender gaps exist in time use? By the end of this project, you will have practiced essential data skills – including data cleaning, aggregation, merging datasets, creating informative charts – and drawn meaningful interpretations about the interplay between time use, labor, leisure, and economic factors.
Datasets Provided
- OECD Time Use by Gender (oecd time use by gender.csv), as of July 29, 2024
- Time Use Diaries (Detailed) (time use diaries.xlsx), produced by ourworldindata.org
- GDP per Capita by Country (gdp per capita.csv), in constant 2015 US$
Use the provided datasets for all analysis. Document how you handle missing values, correct spelling issues, or merge datasets.
Data Sources
- OECD Time Use Database. Time spent in paid/unpaid work, personal care, and leisure by sex. Organisation for Economic Co-operation and Development. Retrieved from: https://www.oecd.org/en/data/datasets/time-use-database.html
- World Bank. GDP per capita (constant 2015 US$). World Development Indicators. Retrieved from: https://data.worldbank.org/indicator/NY.GDP.PCAP.KD
- Esteban Ortiz-Ospina, Bastian Herre, Tuna Acisu, Charlie Giattino, and Max Roser (2020). “Time Use.” OurWorldInData.org. Retrieved from: https://ourworldindata.org/time-use
1. Data Exploration 一 Time Use in OECD Countries
1.1 Open the OECD time use by gender dataset (oecd time use by gender.csv).
1.2 Count the number of countries in this dataset.
1.3 Create a pivot table showing, at the country level, the share of total time spent on the five broad categories: Paid work or study, Unpaid work, Personal care, Leisure, and Other.
1.4 Briefly describe how people spend their time across OECD countries (1-2 sentences).
2. Data Wrangling 一 Leisure Time & Its Composition
2.1 Create a new variable such that time use is measured in hours per day.
2.2 Compute the mean and median of total leisure time across all countries.
2.3 Plot a bar chart of total leisure time (in hours), with countries sorted from highest to lowest. Include the mean you compute in part 2.2 as an additional bar on the graph. Describe the chart in 1–2 sentences.
2.4 Use the time use diaries dataset (time use diaries.xlsx) to construct two new measures of leisure time:
• Series 1: For each country, add the minutes spent on the following activities:
o Seeing friends
o Shopping
o Sports
o TV and Radio
o Attending events
o Eating and drinking
o Other leisure activities
• Series 2: For each country, add the minutes spent on the following activities:
o Seeing friends
o Sports
o TV and Radio
o Attending events
o Other leisure activities
2.5 Compare the OECD measure of total leisure time with Series 1 and Series 2. You would need to merge the two datasets at the country level to compare.
2.6 Which of the two new series is most closely aligned with the measure of total leisure time in the OECD time use dataset? Discuss whether “eating and drinking” and “shopping” should count as leisure. Justify your answer based on how time use surveys group activities into different categories.
3. Data Analysis 一 Relationship between Work Hours and Income
3.1 Create two pivot tables showing the total time spent on paid work or study time for the top 5 and bottom 5 countries. Report time in hours per day.
3.2 Is there a difference in hours worked between these groups? Should we expect a difference? Why or why not?
3.3 Merge GDP per capita data (gdp per capita.csv) with time use data for OECD countries (oecd time use by gender.csv). Choose a GDP year (between 1960–2014) for analysis. The choice of GDP year should be consistent with when the OECD data was updated and the availability of information for all countries. (Hint: Use VLOOKUP function in Excel with country codes to match.)
3.4 Create a graph showing the relationship between total paid work hours and the log of GDP per capita. Describe whether the relationship appears positive or negative.
3.5 Split countries into above- and below-median GDP per capita. Compute mean and standard deviation of total paid work time for both groups.
3.6 Conduct a difference in means test on paid work time across the two groups. State your hypothesis and interpret the p-value.
4. Data Interpretation 一 Gender Gap in Time Use
4.1 Create a chart that compares the average time spent by men and women across the five broad time-use categories.
4.2 Do you notice gender gaps in time use? Write 1–2 sentences about what you learn from this graph. Briefly explain why a gender gap in time use might exist.
4.3 Create a table showing the ratio of time spent on unpaid work by men vs. women for each country.
4.4 Test whether the ratio is statistically less than 1 (i.e., whether women spend more time on unpaid work). Describe your hypothesis and run an appropriate statistical test.