Old-school Chicago Cheating

In Freakonomics, we talk a lot about catching cheaters using data. Based on a newspaper clipping that Stephen Stigler (a well-known professor of statistics at the University of Chicago, author of a wonderful book on the history of statistics, and son of the great Chicago economist George Stigler), there is a long history of using data to catch bad behavior in Chicago.

The newspaper clipping he sent me was from August 25, 1966 — even before I was born. It concerned a tainted exam for applicants to be operating engineers in the Cook County Sanitary District (whatever that is).

On the test, the newspaper reports that of 47 applicants, 15 scored between 90 and 95, 7 scored exactly 84, and 8 scored exactly 69. Superintendnt Vinton said the patterns “just didn’t look right.” A consultant was hired. He provided a 43 page report.

Some evidence in the consultant’s report:

1) Rumors of a cash transfer of $4,500.

2) When the consultant gave the test to Philadelphia sewage workers, no one scored
higher than 70.

3) A surprise retest was called, with the exam to be graded by the consultant. The top scorers sat arms crossed and refused to fill out the test. One man who had scored 95 retested with a 50 the second time around. Ten other suspected frauds
fell from 81.7 to about 50. Seven honest candidates scored about the same
on the retest.

The key breakthrough:

Each of the answer sheets had typing identifying them, done by Mrs. Peggy Regan before the exam and then she rubber stamped each of them. When they got out the microscope, it turned out the answer sheets of the suspected cheaters were cut from a different ream of paper, had typing from a different typewriter, and the
rubber stamp had been applied much more forcefully to the cheaters’ exam sheets.

The answer: the burly Ronald E. Huston had been paid to substitute answer sheets for the cheaters and had himself done the typing and rubber stamping on the falsified sheets.

And these cheaters didn’t go lightly. Superintendent Vinton’s automobile was subsequently firebombed!


Anonymous

Dude, that story's totally intense.

Anonymous

http://www.salon.com/politics/war_room/2005/09/29/bennett/index.html
Bill Bennett sounds like he's read _Freakonomics_ and put his own spin on it.

John M

To switch gears for a minute, I just read a Wall Street Journal Op-Ed suggesting that the number of criminals has actually been increasing as the crime rate has been falling - I don't think the Op-Ed writer has it quite right, but thought it would be interesting to get your take on his methodology. The link to the article is below:

http://online.wsj.com/article/0,,SB112795305361255317,00.html?mod=rss_opinion_main

Cheers.

Tor

I'm shocked. There's a test to be a sewer worker? I wonder if there's any correlation between scores on the test and proficiency in the job.
Peace,
Tor

John Fembup

"I don't think the Op-Ed writer has it quite right"

Then please provide an alternative explanation.

How do you reconcile the reduction to the crime rate and the fact that there is almost 3X's the number of people under corrective supervision today vs. 1980? Has the crime rate actually increased even though more people are under corrective supervision? Is it therefore your theory that the rest of us somehow commit more crime to make up for the bad guys that have been carted away?

Comments?

John M

The correct link might help...

The Hallmark of the Underclass
By CHARLES MURRAY
September 29, 2005; Page A18

http://online.wsj.com/article/0,,SB112795305361255317,00.html?mod=opinion%5Fmain%5Fcommentaries

Monkeydarts

from the department of DUH...
The crime rate is going down because those who would commit crime are more likely to be incarcerated than they were in the past. More perps in prison, fewer criminals free to commit more crimes. Rather than a conundrum it seems pretty self-evident.

Jeffrey McManus

Sanitary districts are the local governments responsible for making the sewers work. It's easy to see why people would go to great lengths to work there.

John M

More criminals in prison leads to less criminals in society leads to less crime seems simple enough, and if I didn't understand Mr. Murray's point I would indeed be attending the school of DUH. I am not interested, however, simply in the logic behind his line of reasoning, but in how his theory relates to Prof. Levitt's work regarding a declining crime rate through the effects of abortion (post Roe v. Wade). Is the decrease in crime rate due in part to a higher rate of incarceration and also in part to a declining number of would be criminals (who were never born due to Roe V Wade)? Or is one factor overriding? I am merely suggesting that it might be interesting to get the 'Freakonomics' take on Mr. Murray's position since it seems to (unfortunately) ignore the abortion effect. No?

CleanWater

The Cook County Sanitary District is now the Metropolitan Water Reclamation District of Greater Chicago - builders of Deep Tunnel. It was a hot bed of corruption for years - supposedly slightly cleaner these days. In the early 1970's, some of the commissioners led by Valentine Janicki, went to prison for accepting almost a millions dollars in bribes to award a sludge hauling contract. About 15 years ago, to celecrate its 100th anniversary, the built a fountain - it sprays a stream of water over the Chicago River just west of the Lake Michigan locks

Anonymous

So, according to you:

"It is true that, on average, crime involvement in the U.S. is higher among blacks than whites. Importantly, however, once you control for income, the likelihood of growing up in a female-headed household, having a teenage mother, and how urban the environment is, the importance of race disappears for all crimes except homicide. "

.....well - yeah, everything would be equal except for the minor detail of MURDER. Heck - why make an issue about the fact that - no matter what fudge factor you insert - blacks commit more murders than any other race in America? Gee -- you, if you discount the Jews that Hitler killed who were from Germany, the Salvic states, or any European country - he hardly killed anyone at al!! Brilliant logic. People HATE to look at the facts: blacks, who are about 15% of our population, commit disproportionately MORE than 15% of the crime in America. The reasons come later are irrelevant to the statistics. Blacks commit more crime - period. Here's some reasons people don't want to look at:

-The percentage of black males between the ages of 18 and 30 who are not looking for, nor interested in getting, work has increased from ~4% in 1970 to over 30% today!! THAT contributes to a higher crime rate - and it is DIRECTLY the choice of BLACK PEOPLE.

-The rate of illegitimate children born into single-parent families in the black population is about 60% - as is as much as 80% in certain cities: such as New Orleans. THAT contributes to an increased crime rate and, again: is the direct product of the choices of BLACK PEOPLE!

Read more...

49erDweet

In spite of what Anonymous 9:53 PM sez, there is more to consider than "choices" of blacks. The term itself envisions a level playing field of alternatives in lifestyles available to all persons which at some point one may simply plug into to ensure life-altering results.

I submit that for many reasons - most not of their own making - that chance is not available for many, many poor young persons, including a larger than average number of minority youth. Blacks probably being the group most deeply impared.

I would be interested to learn of statistics concerning the graduation rates of student athletes versus other academics of US universaties and colleges, say over the past 20 years, and as a secondary question the percentage of black student athletes graduating versus all others. I suspect the failure of black student athletes and the staffs of institutions of higher learning to create a larger cadre of well-educated, professional young blacks contributes significantly to this situation.

Read more...

Doc

Double Duh. More in prison as we arrest newbies and keep geezers incarcerated as part of a growing population. Lower crime rates at the margin due to the Baby Bust. Young people commit crimes at a much higher rate than old people. Look for crime stats to turn around in the face of the Echo Boomers turning into young adulthhod at the same volume as Baby Boomers.

Anonymous

Dude, that story's totally intense.

Anonymous

http://www.salon.com/politics/war_room/2005/09/29/bennett/index.html
Bill Bennett sounds like he's read _Freakonomics_ and put his own spin on it.

John M

To switch gears for a minute, I just read a Wall Street Journal Op-Ed suggesting that the number of criminals has actually been increasing as the crime rate has been falling - I don't think the Op-Ed writer has it quite right, but thought it would be interesting to get your take on his methodology. The link to the article is below:

http://online.wsj.com/article/0,,SB112795305361255317,00.html?mod=rss_opinion_main

Cheers.

Tor

I'm shocked. There's a test to be a sewer worker? I wonder if there's any correlation between scores on the test and proficiency in the job.
Peace,
Tor

John Fembup

"I don't think the Op-Ed writer has it quite right"

Then please provide an alternative explanation.

How do you reconcile the reduction to the crime rate and the fact that there is almost 3X's the number of people under corrective supervision today vs. 1980? Has the crime rate actually increased even though more people are under corrective supervision? Is it therefore your theory that the rest of us somehow commit more crime to make up for the bad guys that have been carted away?

Comments?

John M

The correct link might help...

The Hallmark of the Underclass
By CHARLES MURRAY
September 29, 2005; Page A18

http://online.wsj.com/article/0,,SB112795305361255317,00.html?mod=opinion%5Fmain%5Fcommentaries

Monkeydarts

from the department of DUH...
The crime rate is going down because those who would commit crime are more likely to be incarcerated than they were in the past. More perps in prison, fewer criminals free to commit more crimes. Rather than a conundrum it seems pretty self-evident.