The Ramadan Experiment

Name: The Ramadan Experiment
Author: Abdul Zahir Sheikh

Science · Statistics · Faith

What Thirty Days of Fasting Reveal About Human Behaviour

What does a month of fasting actually change? Courts, hospitals, police files and markets from more than thirty countries, graded on one scale, with costs counted beside benefits.

Paperback: 978-9914-776-56-0
Hardcover: 978-9914-776-55-3
eBook: 978-9914-776-54-6

Bookshop.org Kobo eBook Apple Books Google Play

The book in ten slides

About three minutes; each slide stays up long enough to read

The Ramadan Experiment. What does a month of fasting actually change? Courts, hospitals, police files and markets from more than thirty countries, graded on one scale.

Two sets of claims. Those who keep the fast speak of calm, restraint and generosity. Those who watch from outside speak of short tempers, slow offices and crowded roads at sunset. Almost every claim arrives without its evidence.

Poorer, but happier. Lengthen the average fast from 12 hours to 13, and output growth falls by roughly 0.6 to 0.9 of a percentage point, while the chance of reporting being happy rises by about 4 points and satisfied with life by about 9.

A natural experiment. The lunar calendar carries the fast through every season, and latitude lengthens or shortens the day. A longer fast can be compared with a shorter one in the same place.

Two judgements for every finding. Is the change real: Established, Probable, Unresolved or Unsupported? And did the fast and the faith cause it? The Attribution Ladder answers the second in four rungs.

Four tests, set first. A change is attributed to the fast only if it passes all four. The thesis fails if fewer than three domains, from three independent research teams, pass them.

Costs on the same ladder. Sleep falls by about an hour a night. Household spending and food prices rise. Fasting in early pregnancy is followed by costs that can last a lifetime. Weight loss has the most studies and stops at Observed.

Real, but not the fast. Crime falls in Ramadan in Switzerland and Türkiye: the fall is Established. But longer fasts did not deepen it, and in Türkiye it fell even where fewest people fast. Crime fails two of the four tests.

Three domains pass. The decisions of judges, reported well-being, and fewer terrorist attacks after longer fasts. Three domains, from three independent teams: exactly the number the book asked for, and not one more.

Where the method stops. Each passing domain rests on one study; a failed replication of any one would leave two. And no dataset can show intention, sincerity, or whether a fast is accepted.

About the book

Ramadan arrives each year carrying two sets of claims. Until recently, almost no one could check either.

Those who keep the fast speak of calm, restraint and generosity. Those who watch from outside speak of short tempers, slow offices and crowded roads at sunset. This book asks for the record: what changes measurably in human behaviour during Ramadan, and how much of that change belongs to the fast and the faith rather than to season, community or schedule?

The month turns out to be a natural experiment. The lunar calendar carries the fast through every season, and latitude lengthens or shortens the day, so a longer fast can be compared with a shorter one in the same place. The book reads that evidence — court records, police files, hospital data, tax returns, stock exchanges and surveys — and grades every finding on one ladder: is the change real, and how far can it be attributed to the fast?

The four tests, and the result that would prove the book wrong, are set out before a single study is examined.

Costs are graded exactly as benefits are. Sleep falls by about an hour a night. Household spending and food prices rise. Fasting in early pregnancy is followed by costs that can last a lifetime. Crime does fall in Ramadan, and the book shows that the fast is not what lowers it.

The verdict is stated with its margin. The book required three domains, from three independent research teams, to pass all four tests. Three do: the decisions of judges, reported well-being, and terrorist violence. Each rests on a single study, and the book says plainly what one failed replication would do to the result.

It also knows where its method stops. No dataset can show intention, sincerity, or whether a fast is accepted.

Written by Abdul Zahir Sheikh, a Certified Fraud Examiner and Senior Partner at BDO East Africa with four decades of investigative experience across fourteen countries, the book ends with The Ramadan Record: pages for readers to keep through their own month, and to read again after Eid.

Thirty days. More than a billion observations. One verdict, stated with its margin.