Empirical Project 8 Measuring the non-monetary cost of unemployment

Learning objectives

In this project, you will:

  • practise working with fairly large datasets
  • detect and correct entries in a dataset (Part 8.1)
  • recode variables to make them easier to analyse (Part 8.1)
  • calculate percentiles for subsets of the data (Part 8.1)
  • use column charts, line charts, and scatterplots to visualize data (Part 8.2)
  • calculate and interpret correlation coefficients (Part 8.2)
  • calculate and interpret confidence intervals for the difference in mean between two groups (Part 8.3).

Key concepts

  • Concepts needed for this project: mean, standard deviation, range, percentile, correlation, correlation coefficient, confidence interval.
  • Concepts introduced in this project: confidence interval for the difference in means.


CORE projects

This empirical project is related to material in:

Unemployment is a problem for society, not only because of the lost output and wages, but also because it can have a direct impact on individuals’ well-being and life satisfaction. If societies have a conviction that able people of a working age should be working, then being unemployed could result in a fear of being stigmatized, a sense of shame from being unemployed, and feeling inferior to others who are employed, all of which would reduce an individual’s life satisfaction.

Disutility from unemployment is a concept that we cannot measure directly, so instead we will use self-reported life satisfaction. This measure has its limitations but is widely used to quantify costs that we cannot observe, such as the effect of becoming chronically ill or other life-changing events.

We will be using an approach and data that is similar to the study ‘Employment status and subjective well-being’, which used the European Values Study (EVS), a cross-country survey, to investigate the differences in life satisfaction between people of different employment statuses. The hypothesis was that unemployed people would, on average, be less satisfied with life than employed people, and that this relationship between employment status and life satisfaction would vary depending on social norms. There is evidence to support this hypothesis in specific countries, so we will explore this relationship in EU countries.123

As discussed in the study, one explanation for a relationship between employment status and reported life satisfaction is social norms regarding a work ethic. If social norms are an important determinant of life satisfaction, then we expect the gap in life satisfaction between employed and unemployed to be larger in countries with a stronger self-reported work ethic.

While the main focus of this project is on people who are full-time employed and people who are unemployed, we will also consider whether life satisfaction differs for other employment statuses, such as people who are retired. Norms of working may not be as strong for the elderly, so the lack of formal employment would have less of an effect on life satisfaction compared with working-age people who are unemployed.

Working in Excel

Working in R

Working in Google Sheets

Working in Python

  1. Winkelmann, L., and Winkelmann, R. (1998). ‘Why are the unemployed so unhappy? Evidence from panel data’. Economica, 65(257), 1–15. 

  2. Clark, A. E. (2003). ‘Unemployment as a social norm: Psychological evidence from panel data’. Journal of Labor Economics, 21(2), 323–351. 

  3. Clark, A. E., and Oswald, A. J. (1994). ‘Unhappiness and unemployment’. The Economic Journal, 104(424), 648–659.