Comments on economics, mystery fiction, drama, and art.

Saturday, May 25, 2013

Generating the Wealth of Nations 24: Real GDP per Capita and a Measure of Well-Being

This really goes back to a much earlier part of the semester, when we talked about measuring economic output and the relationship between real GDP per capita and well-being.  But I thought it was interesting.

Here’s a report from Forbes, on a measure of happiness:

Quantifying happiness isn't an easy task. Researchers at the Gallup World Poll went about it by surveying thousands of respondents in 155 countries, between 2005 and 2009, in order to measure two types of well-being.

“First they asked subjects to reflect on their overall satisfaction with their lives, and ranked their answers using a "life evaluation" score from 1 to 10. Then they asked questions about how each subject had felt the previous day. Those answers allowed researchers to score their "daily experiences"--things like whether they felt well-rested, respected, free of pain and intellectually engaged. Subjects that reported high scores were considered "thriving." The percentage of thriving individuals in each country determined our rankings.”
I took their rankings and charted the relationship between real GDP per capita (in 2011) and the percent of respondents ranked as “thriving,” “struggling,” and “suffering.”  The three charts below show those relationships.  The correlations between real GDP per capita and the happiness rankings are as follows:

Real GDP
per Capita and
Percent Thriving           0.69
Percent Struggling       -0.61
Percent Suffering         -0.45

This does suggest that real GDP per capita is related to well-being; on the other hand, I’m surprised by how low the correlations are—especially the correlation between real GDP per capita and the percent of the population classified as suffering.  (The second chart should be labeled "Real GDP per Capita and the percent of People Saying They Are Struggling."  My typing error.)

(Click a chart to enlarge it.)


Blogger Angostura said...

Great analysis! Thank you.

Do you think you would get a higher correlation coefficient in the last graph if you stratified it? Eg look at GDP pc below $10k separately? I think this could separate dependence on life necessities (Herzberg's hygiene factors) from discretionary and luxury spend.

4:24 AM

Blogger Angostura said...

Thank you for the insightful comments.
Do you think your correlation coefficient could be more meaningful if you stratified the data? Say, in the last graph, looking at countries below $10k GDP pc separately could distinguish between necessities and discretionary or luxury spend. Similar to Herzberg's hygiene factors in management.

4:29 AM

Blogger doc said...

Actually, I did that. The correlation between GDP per capita and "struggling" in countries with GDP per capita below $10,000 was *lower* than that for countries with GDP per capita above $10,000 (about -0.40 and about -0.60). This strikes me as something of a conundrum, and one I'm not sure how to pursue.

11:27 AM


Post a Comment

<< Home