PDA

View Full Version : Meditating group reduces crime in Washington - experiment


litewave
15th July 2007, 07:54 AM
In 1993 a group of scientists led by John Hagelin conducted an experiment in which several thousand people in Washington D.C. would meditate together twice a day for almost 2 months. This was correlated with a highly significant reduction in crime in D.C.

More info:

Effects of Group Practice of the Transcendental Meditation Program on Preventing Violent Crime in Washington, DC:
Results of the National Demonstration Project,
June–July 1993 (http://www.mum.edu/m_effect/dc_md.html)

John Hagelin (http://en.wikipedia.org/wiki/John_Hagelin)

David Lynch has recently made a movie about it, here's a trailer:

The Square Root of One Percent
(http://www.youtube.com/watch?v=IIN8r9rgtUM)

-Fran-
15th July 2007, 08:11 AM
In 1993 a group of scientists led by John Hagelin conducted an experiment in which several thousand people in Washington D.C. would meditate together twice a day for almost 2 months. This was correlated with a highly significant reduction in crime in D.C.


Were all these meditating people criminals? ;)

prewitt81
15th July 2007, 08:25 AM
In 1993 a group of scientists led by John Hagelin conducted an experiment in which several thousand people in Washington D.C. would meditate together twice a day for almost 2 months. This was correlated with a highly significant reduction in crime in D.C.

Correlation does not imply causation. (http://www.stat.tamu.edu/stat30x/notes/node42.html)

Causation is not implied by correlation. (http://www.infoworks.ride.uri.edu/1999/techbrief/techbrief_8.htm)

Correlation implies causation. . . NOT (http://en.wikipedia.org/wiki/Correlation_does_not_imply_causation)

fls
15th July 2007, 08:39 AM
Robert Park mentioned this in his book "Voodoo Science: The Road from Foolishness to Fraud".

The crime rate did not drop, but rather rose during this period. The claim was that it did not rise as much as they predicted that it would have due to the temperature (there is a previously established correlation between temperature and crime rate). They were also very selective in which particular crime statistics they looked at for comparison - essentially discarding those that went up and selecting only those that went down (or went up less than they predicted) for their report.

Linda

Horatius
15th July 2007, 08:44 AM
Correlation does not imply causation. (http://www.stat.tamu.edu/stat30x/notes/node42.html)





...and none of that matters if there's nothing to correlate the cause to (http://en.wikipedia.org/wiki/TM-Sidhi_program#Study_on_the_Maharishi_Effect_in_Was hington.2C_D.C.).....



Overall, there was an 18-percent reduction in violent crime, he told the press. When a reporter asked, an 18-percent reduction compared to what, Hagelin answered, compared to the level of violent crime had the TM meditators not meditated.


. BUT ARE MURDERS IN WASHINGTON BEING COMMITTED MORE HUMANELY? (http://bobpark.physics.umd.edu/WN93/wn072393.html)
Physicist John Hagelin promised to reduce violent crime by having one thousand TM experts meditate coherently (WN 25 Jun 93). Oops! Preliminary results suggest they got the polarity wrong; homicide hit a record high level. But, at a press conference yesterday, Hagelin offered a lesson in data analysis: although murder is up from a year ago, there has been a decrease of 2.3% in "brutal crime." A clean shot between the eyes maybe. Could the drop in brutal crime be due to the increase in D.C. police surveillance and a new summer jobs program? Those programs, Professor Hagelin explained, have been more than offset by the heat wave--so the field must be working. Patched through to the press conference via satellite link from Holland, Zen master Maharishi Yogi, the intellectual force behind the experiment, explained the success of the experiment: "No one wants to be shot--it's painful." Amen.

Dr Adequate
15th July 2007, 08:46 AM
Maxwell Rainforth writes (http://istpp.org/crime_prevention/voodoo_rebuttal.html):

"It is true the murder rate did not drop during the course ... For six weeks ending the month before the experiment, from mid-March through April, homicides in Washington averaged ten per week. Beginning one week after the course and for twelve weeks thereafter, homicides also averaged ten per week. During the eight weeks of the experiment, in June and July, the average was again ten per week — except for one horrific 36-hour period in which ten people died. Apart from this brief episode, which was a statistical outlier, the level of homicides during June and July of 1993 was not significantly higher than the remainder of the year."

You got that? According to one of the authors of the study, the murder rate went up during the experiment --- but if you arbitrarily discount the 36 hour period when the murder rate was highest, then it was not "significantly" higher than the rest of the year.

Color me impressed by the awesome power of TM.

Mojo
15th July 2007, 08:50 AM
Look on the bright side: this work won them the 1994 Ig Nobel peace prize.

Moochie
15th July 2007, 11:00 AM
Looks like Lynch's An Inconvenient Meditation.

M.

litewave
15th July 2007, 01:59 PM
We should keep in mind that no one expected that reducing such a complex phenomenon as crime rate would be as straight-forward as turning down volume on a radio. It was expected that there would still be fluctuations but the overall crime rate would be reduced, presumably as a result of non-local connections, similar to those found in quantum physics.

So, they took crime data from the two-month period of meditation and found that the overall crime incidence during this period was lower than predicted for this period by the historical pattern. And importantly, they also found that the difference from the historical pattern was statistically significant, that is it was very improbable that the reduction could have happened by chance (by usual fluctuations within the historical pattern).

In order to reasonably assess the data and the significance of the found difference one might need to do statistical analysis, or at least take a look at a bigger picture. But Robert Park, in his Skeptical Inquirer article, didn't seem to do that. He just mentioned a prominent fluctuation and concluded that the crime rate rose. Moreover, the fluctuation he mentioned was in murders, while murders, as was pointed out by Maxwell Rainforth, make up only 3 percent of overall violent crime. The overall violent crime includes assaults, murders and rapes, and this was found to be significantly reduced during the meditation period. And if Rainforth has the data right, Park also exaggerated the murder fluctuation by claiming that "the murder rate for those two months reached a level unmatched before or since." Rainforth says there was a 36-hour period of murders that caused the murder count for that week to jump, but otherwise the average weekly murder count during the meditation period was the same as in the rest of the year (10 murders per week).

Horatius
15th July 2007, 02:12 PM
You got that? According to one of the authors of the study, the murder rate went up during the experiment --- but if you arbitrarily discount the 36 hour period when the murder rate was highest, then it was not "significantly" higher than the rest of the year.

Color me impressed by the awesome power of TM.

Rainforth says there was a 36-hour period of murders that caused the murder count for that week to jump, but otherwise the average weekly murder count during the meditation period was the same as in the rest of the year (10 murders per week).



No extra comment really needed, is there?

JoeEllison
15th July 2007, 02:24 PM
So, in other words, John Hagelin is a liar and a fraud.

litewave
15th July 2007, 02:28 PM
No extra comment really needed, is there?

I wanted to contrast it with Park's claim. If Rainforth has the data right then Park's claim is an obviously misleading exaggeration.

Horatius
15th July 2007, 03:06 PM
I wanted to contrast it with Park's claim. If Rainforth has the data right then Park's claim is an obviously misleading exaggeration.



If.

fuelair
15th July 2007, 03:09 PM
So, in other words, John Hagelin is a liar and a fraud.
and a wooo ruin of a once real scientist. Got it in one!!

fuelair
15th July 2007, 03:10 PM
So, in other words, John Hagelin is a liar and a fraud.
and a wooo ruin of a once real scientist. Got it in one!!

JoeEllison
15th July 2007, 03:18 PM
and a wooo ruin of a once real scientist. Got it in one!!
and a wooo ruin of a once real scientist. Got it in one!!

You got it in two!

JoeTheJuggler
15th July 2007, 10:11 PM
We should keep in mind that no one expected that reducing such a complex phenomenon as crime rate would be as straight-forward as turning down volume on a radio. It was expected that there would still be fluctuations but the overall crime rate would be reduced, presumably as a result of non-local connections, similar to those found in quantum physics.

Oh well, if you say "quantum" it must be scientific.

Dr Adequate
16th July 2007, 03:25 AM
I wanted to contrast it with Park's claim. If Rainforth has the data right then Park's claim is an obviously misleading exaggeration. No. The data is right, and so is Park. What is wrong to the power of wrong squared is Rainforth's interpretation of the data, which involves ignoring 36 hours in which ten people were murdered. His technique of statistical analysis is to discard precisely the data which makes the biggest contribution to proving him wrong. I emailed him about this a few years back, and got no sense out of him on this issue.

Do you not yourself smell a rat? Rainforth discounts the period with the highest murder rate, precisely because it has the highest murder rate. Using the same technique, I could prove what a long dry spell we've been having down my way lately, so long as I'm allowed to ignore all the days when it really pissed it down.

litewave
16th July 2007, 10:02 AM
No. The data is right, and so is Park. What is wrong to the power of wrong squared is Rainforth's interpretation of the data, which involves ignoring 36 hours in which ten people were murdered. His technique of statistical analysis is to discard precisely the data which makes the biggest contribution to proving him wrong. I emailed him about this a few years back, and got no sense out of him on this issue.

Do you not yourself smell a rat? Rainforth discounts the period with the highest murder rate, precisely because it has the highest murder rate. Using the same technique, I could prove what a long dry spell we've been having down my way lately, so long as I'm allowed to ignore all the days when it really pissed it down.
I don't think he discarded the murder fluctuation. As I understand it, this fluctuation simply got lost in the change of the level of overall violent crime, of which murders make up only 3 percent.