Sunday, October 24, 2021
Analytics STEM

Ig Nobel Prize Winners: 2021

The winners of the 2021 Ig Nobel Prize are announced today. One of them caught my attention, which happens to be in the Economics category by Pavlo Blavatskyy for for discovering that the obesity of a country’s politicians may be a good indicator of that country’s corruption.

Very interesting indeed and although it made me chuckle at first, and immediately remembered Putin and recent huge weight-loss by North Korea’s Kim Jong-un (who until recently was guessed to weight about 300lbs). However, as with most statistical data there will always be outliers. It also brings me to think, if politicians recognize this and start to lose weight, will they become less corrupt? Of course not! At any rate, here’s a summary and my takeaways from the full publication…with links and citations to more reading if you’re so inclined.

First, a quick intro to the Ig Nobel Prize. The Ig Nobel Prize is a satiric prize awarded annually since 1991 to celebrate ten unusual or trivial achievements in scientific research. Some have actually gone on to win actual Nobel Prize after winning Ig Nobel awards. Organized by the scientific humor magazine Annals of Improbable Research (AIR), the Ig Nobel Prizes are presented by Nobel laureates in a ceremony at the Sanders Theater, Harvard University, and are followed by the winners’ public lectures at the Massachusetts Institute of Technology. [except during Pandemic time, it was done virtually]


Pavlo Blavatskyy is a member of the Entrepreneurship and Innovation Chair, which is part of LabEx Entrepreneurship (University of Montpellier, France) and is funded by the French government (Labex Entreprendre, ANR-10-Labex-11-01). His work is published at “Obesity of Politicians and Corruption in Post‐Soviet Countries,” Pavlo Blavatskyy, Economic of Transition and Institutional Change.

It’s important to note that the data used is from 2017 (not 2021) and Pavlo focused on specific countries/states (post-Soviet-era countries) for his study.
Also, when you read the tables and charts (in his plublication), remember that all the lower values of corruption index values (Transparency International Corruption Perceptions Index, and World Bank worldwide governance indicator Control of Corruption) indicate HIGHER corruption.

Here’s a part of the Abstract: “We collected 299 frontal face images of 2017 cabinet ministers from 15 post-Soviet states (Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan). For each image, the minister’s body-mass index is estimated using a computer vision algorithm. The median estimated body-mass index of cabinet ministers is highly correlated with conventional measures of corruption (Transparency International Corruption Perceptions Index, World Bank worldwide governance in-dicator Control of Corruption, Index of Public Integrity). This result suggests that physical characteristics of politicians such as their body-mass index can be used as proxy variables for political corruption…at a very local level.”

The sentence in the introduction instantly grabbed my attention: “Grand political corruption is a clandestine economic activity that is difficult (even life-threatening) to measure directly.”

Part of the explanation of the Dataset: “We collected 299 frontal face images of cabinet ministers from 15 post-Soviet states who were in office in 2017…Whenever possible, we selected a minister’s image that resembled a passport photograph—unobscured frontal face image preferably taken during an event in 2017.”

The Estimation process: “For each image in the dataset, the minister’s body-mass index is estimated using the computer vision algorithm recently developed by Kocabey (2017). This algorithm is a two-stage procedure. The first stage is a deep convolutional neural network VGG-Face developed by Parkhi, Vedaldi, and Zisserman (2015). This neural network extracts the features from a deep fully connected neuron layer fc6 for the input image. The second stage is an epsilon support vector regression (Smola & Vapnik, 1997) of the extracted features to predict body-mass indexes of 3,368 training images (with known body-mass index values) collected by Kocabey (2017).
…As a robustness check, we also estimated body-mass indexes from our dataset of 299 images using the algorithm of Wen and Guo (2013). When this algorithm converged, it produced similar results to Kocabey’s algorithm. However, for 43 out of 299 images (14.4%), the algorithm of Wen and Guo (2013) did not converge.”

The Results: “Estimated body-mass index for ministers in our dataset is generally quite high. According to the estimated body-mass index, 96 out of 299 ministers (32%) are severely obese (estimated body-mass index between 35 and 40). In particular, 13 out of 24 Uzbek ministers (54%), 8 out 18 Tajik ministers (44%) and 10 out of 24 Ukrainian ministers (42%) are estimated to be severely obese. Another 13 out of 299 ministers in our dataset (4%) are very severely obese (estimated body-mass index greater than 40). In particular, 3 out of 20 Kazakh ministers (15%) and 2 out of 24 Ukrainian ministers (8%) are estimated to be very severely obese. Only 10 out of 299 ministers in our dataset (3%) are estimated to have normal weight (body-mass index between 18.5 and 25)…None of the ministers in our dataset is estimated to be un-derweight (body-mass index below 18.5).”

Figure 1 below shows a Scatterplot of median estimated ministers’ body-mass index against Transparency International Corruption Perceptions Index 2017 (with a linear trend), where lower values of CPI indicate greater corruption.

Scatterplot of median estimated ministers’ body-mass index against Transparency International Corruption Perceptions Index 2017 (with a linear trend)
Figure 1

Pavlo ends the RESULTS section with: “High correlation between our median estimated ministers’ body-mass index…and conventional measures of corruption…is striking.”

These findings are probably intuitively known or observed by most public, however, by quantifying in a scientific way makes it a more provable hypothesis which then can be extended to broader swathes of the world’s principalities. The idea of turning images into estimations of BMI is also interesting and useful especially in cases where actual weight may be never accurately known or disclosed. Below are some additional resources.

See the complete list of 2021 Ig Nobel prize winners (all categories)

Complete list of winners (all years to-date)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back To Top