Ad Here  
Culture of suave, gentle and British becoming extinct... The other side of the Olympic divide RaGa bowls a decent over Sehwag...No soft edges Psephologists fail the wisdom of crowds People vs. Collins The case of the suspicious husband Lovely Banking experience What a judgment Sir-ji Noise on Social Networks Britain lobs a bomb. But it won’t explode. 500/500 The flawed “Pakistan policy” The loser is … CRICKET The right to reject Waylaid on the National Highway Walking into the sunset Modi sells a Dream? Of mercy petition and mercy killing Brick gets pricked Why we get a compromised deal The death of cash Editor unplugged Throwing ink is fine The wily old fox Of discrimination and rights Bleeding Blue WAR-MONGERING and a national pastime The return of the muffler man Rajiv killers escape the gallows Nitish swords Modi The broken window PM Watch Being taken for a ride The curious case of the missing maid FITTING farewell The KING is a fugitive The fight along the border Two leaves merged? Resolving disputes: Scotland, Ajit, Dhoni A judgment that shocked a nation’s conscience Educational shame End of the world and the moving finger... A troubled fortnight A ride around the city Flight into danger Greater than the greatest ? Odd-Even Plan Privatise the Railways the lost generation... From Gungi Gudiya to Iron Lady Two more for Modi Cracking the CA code The future is here The car dealer and the cabbie stories…
The case of the suspicious husband
“The average gestation period is 266 days. Some babies come early. Others come late. Yours was late. In the United States there are about twelve thousand babies born this late each year. Yours is just one of them.”

It happened a good 40 years ago; in fact, this year (2013) we celebrate the 40th anniversary of this celebrated story that has made its way into statistic folklore. It’s worth recounting it here.

In 1973, a San Diego reader complained to a newspaper columnist:

“You wrote in your column that a women is pregnant for 266 days…. I carried my baby for ten months and five days… there is no doubt about it because I know the exact date my baby was conceived… My husband is in the Navy and it couldn’t have been conceived any other time because I saw him only once for an hour, and I didn’t see him again until the day the baby was born… there is no way this baby isn’t his baby… please print a retraction about that 266-day carrying time because otherwise I will be in a lot of trouble.”

The columnist wrote back saying: “Dear Reader: The average gestation period is 266 days. Some babies come early. Others come late. Yours was late.”

While the columnist was bang on, he could have been more quantitative. He should have marshaled math logic to provide greater assurance to the couple. This is what he could have done.

That the kid was born late is a fact. At approximately 310 days it was way above 266, which the columnist rightly says is the global average. The point is, what is the probability of this 310 happening. Meaning, what is the probability of a kid getting born 44 days behind the average? Meaning, is it so unusual that we could conclude that the woman was lying? This is where statistics comes in handy.

Let’s assume that the gestation period follows a normal distribution curve; that is, it is bell shaped. In that case it would be possible, by using the Z value table, to find the probability that a pregnancy would last at least 310 days. Now, current data says that the average gestation period is 266 days with a standard deviation (movement around the mean) of 16 days. This is an information to which the columnist was privy. Statistics’ normal distribution table suggests that under the circumstances (average = 310, standard deviation =16), the probability that a pregnancy would last at least 10 months and 5 days (about 300days or longer) is 0.003.

This means that three babies out of a thousand are born this late. With 40 lakh babies born each year in America,about twelve thousand babies are born this late. The columnist could have replied saying:“the average gestation period is 266 days. Some babies come early. Others come late. Yours was late. In the United States there are about twelve thousand babies born this late each year. Yours is just one of them.” This response would not only have been reassuring to the wife, but persuasive to her husband as well.

But just hold on. In hypothesis testing, the probability of 0.003 is extreme for the null hypothesis to be true. In our example, the null hypothesis is: “This baby is my husband’s.” In traditional hypothesis testing, you reject the null hypothesis if the P-value is smaller than the levels of significance. Mathematically, a P-value of 0.003 would reject the null hypothesis. So, we normally would reject the hypothesis that this baby is San Diego reader’s husband’s baby. So, if it was really the husband’s baby, how do we explain our wrong test result? Simple. At times, life falls outside the predictions of probability theory! In short, life is not all math.

Author :
Reported On :
Sector :
Shoulder :
IE, the business magazine from south was launched in 1968 and pioneered business journalism in south. Through the 45 years IE has been focusing on well-presented and well-researched articles. When giants in the industry stumbled to keep pace with the digital revolution, IE stayed affixed embracing technology.
Read more
Economist Communications Ltd is committed to ensuring that your privacy is protected.
Read more
You agree that your use of this Website and the purchase of the magazine will be governed by these terms and conditions.
Read more
S-15, Industrial Estate,
Chennai - 600 032.
PHONE: +91 44 22501236