pageok
pageok
pageok
Careful with Those Studies:

Scott Horton in The New Republic suggests there are "unsettling issues" about "whether a selective attitude is taken in prosecution--that is, whether the Justice Department is treating Spitzer in a manner consistent with other (notably Republican) figures caught in a similarly compromised position."

I don't know enough about the subject to comment on the big picture question, but I did want to point out one item. Horton writes,

PIN has emerged as one of the most highly politicized branches of a highly politicized Justice Department. According to a study done by two university professors, under President Bush PIN has initiated 5.6 cases involving Democrats for every one case involving a Republican. This statistical data strongly suggest that PIN has a habit of aggressively pushing cases on the basis of partisan political criteria.

But Horton doesn't point out that for the most analogous class of officeholder -- "State-Wide and Federal Elected Officials" -- the study reports (emphasis added) that "there is not a significant difference between the number of federal investigations of State-wide and federal elected officials and what would be expected given the representation of Democratic and Republican office-holders in the population." The breakdown of investigations there was 36 Democrats and 30 Republicans, which is almost identical to what would be expected given the ratio of Democrats to Republicans in those offices (33 to 27, according to the study).

I haven't examined the rest of the study carefully, and can't speak to the significance of its findings about local officials, where there is a large Democrat/Republican disparity; there may well be some bias there, though it's hard to tell without further investigation. But it seems noteworthy that this statistical study's empirical findings as to the officeholder category that's most analogous to Spitzer's do not support the inference for which Horton is using the study.

BGates:
There can be 3 possible results from this study:
Democrats are prosecuted more often, indicating aggressive partisanship;
Republicans are prosecuted more often, indicating Republicans are more prone to committing crimes; or
Members of both parties are prosecuted equally often, indicating aggressive partisanship to cover up the fact that Republicans are more prone to committing crimes.
3.13.2008 2:51pm
Hoosier:
My hunch is that corruption is more frequent in big city governments than at the state level. (I grew up in Cook County, IL, so that may skew my perspective).

Democrats control more big urban governments than do Republicans. So if you take away state-wide office holders, you're left with a disproportionate number of Democrats.

Well, it's a theory.
3.13.2008 3:02pm
Orielbean (mail):
Anyone else notice that the Utah Taser guy settled with the police?
3.13.2008 3:05pm
Anderson (mail):
I like Horton's blog, and have been quoting his around today, but he is not always the most cautious fellow with his facts.
3.13.2008 3:18pm
frankcross (mail):
Yes, this is quite spurious. The ratio is due to prosecution of relatively low level council members. And this may not even be a biased ratio, given the possibility that there are many more local Democrat officials. The absolute numbers are meaningless without a ratio statistic, which I didn't find in the data
3.13.2008 3:28pm
Jiffy:
Hoosier

<blockquote>
Democrats control more big urban governments than do Republicans. So if you take away state-wide office holders, you're left with a disproportionate number of Democrats.
</blockquote>

Frank Cross

<blockquote>
The absolute numbers are meaningless without a ratio statistic, which I didn't find in the data
</blockquote>


This factor is accounted for in their study of prosecutions of local politicians. It shows that there was more frequent prosecution of Democrats relative to Republicans than would be expected based on the relative number of Democratic and Republican office-holders. As Eugene says, they didn't find a significant difference in prosecutions of state-wide and federal office holders.
3.13.2008 4:32pm
George Weiss (mail) (www):
as for the big picture..let see here

absolute prosecutorial discretion+absolute prosecutorial immunity=absolute disaster
3.13.2008 4:52pm
Richard Aubrey (mail):
That this is supposed to be disturbing presumes that the reality is that repubs and dems are equally dirty. That is presumed, not demonstrated.
3.13.2008 5:02pm
Bob from Ohio (mail):
Where is the 1993 to 2001 study? So we can compare.
3.13.2008 5:04pm
frankcross (mail):
Jiffy, I didn't notice the "expected frequency." That does suggest something, though I wonder what their source was. You would need to see that. Also, they have no control variables.
3.13.2008 5:04pm
Warmongering Lunatic:
Nobody's going to bribe the only Republican alderman in Chicago, because he doesn't have any influence. And if somebody bribes a councilman in Nowhereseville, Wyoming, it's less likely to reach Justice Department notice than in Chicago.

So, if we control for population of locality, does the same disproportion happen?
3.13.2008 5:19pm
Freddy Hill:
Partisanship may be an explanation, I suppose, but it seems to me that we should apply Occam's razor to the entire thing and discard simpler theories before considering more elaborate ones.

The fact is that Spitzer is viewed as an a-hole that made powerful enemies. Why isn't this sufficient explanation for what happened to him?
3.13.2008 5:27pm
PatHMV (mail) (www):
There are really significant flaws in the "political profiling" study, which have been long since pointed out the last time the authors were trotting out this "study." I spent a fair amount of time pointing out the flaws in several posts, all linked to at StubbornFacts.us. The authors have addressed some of the criticisms; I'm hoping to finish a post shortly addressing their revised analysis and improved description of their (flawed) methodology.
3.13.2008 5:39pm
Bruce Hayden (mail) (www):
You missed the obvious alternative:
There can be 3 possible results from this study:
Democrats are prosecuted more often, indicating aggressive partisanship;
Republicans are prosecuted more often, indicating Republicans are more prone to committing crimes; or
Members of both parties are prosecuted equally often, indicating aggressive partisanship to cover up the fact that Republicans are more prone to committing crimes.
Which is the possibility that Democrats are dirtier at the local level.
3.13.2008 6:18pm
wfjag:
There's also a distinct possibility of researcher bias in the study. The The Kansas City Star, Wed., October 24, 2007 Wednesday at B, Pg. 4, in an article headlined "Shields' brother says she was victim of federal 'political profiling'" reported:


"Katheryn Shields said the U.S. attorney's office was playing politics when she was charged with mortgage fraud a day before she announced her candidacy for Kansas City mayor.

Now her brother says she was just one of hundreds of Democrats who became targets of 'political profiling' by the U.S. Department of Justice.

Testifying before two House Judiciary Committee subcommittees, Donald Shields, a professor emeritus at the University of Missouri-St. Louis, said his review of 820 investigations from 2001 to 2007 showed that 80 percent targeted Democrats, while only 14 percent focused on Republicans and 6 percent aimed at Independents."

**

"Also, he noted a conflict of interest: 'Katheryn Shields has raised claims of selective and vindictive prosecution in her own case, and submitted her brother's study as part of her claim. The claim was rejected by the judge in the case.'

The former Jackson County executive's mortgage-fraud trial began this week.

"She's one of the ones that drives the case home," Donald Shields said in an interview. 'They started investigating her four, five years ago.'"


So Prof. Shield's sis in one of the local Demo. officials that the big, bad Bush Justice Department prosecuted (in her case, for fraud).

Anyone know the result of the trial?
3.13.2008 6:49pm
frankcross (mail):
I think she was acquitted.
3.13.2008 7:45pm
Duncan Frissell (mail):
Is it possible that Democrats are more criminally inclined than Republicans? Certainly, most murderers in America are Democrats (or would vote Democrat if they voted) since there is a great overlap between the Democrat demographic and the murderer demographic. Purely accidental, I'm sure.
3.13.2008 9:43pm
MnZ:
Or could it be that Red states actually prosecute their own corrupt officials before the Feds get a chance while Blue states don't?

I used to live in Ohio. In 2005, Governor Bob Taft was charged in municipal court with taking $5,000 worth of minor gifts from lobbyists.

I now live in the Chicago area. Here, the idea that a politician would be prosecuted in local court for $5,000 worth of gifts is laughable.
3.13.2008 11:12pm
Bob from Ohio (mail):
Still waiting for the study of the Clinton DOJ.
3.14.2008 12:04am
Student:
Trying to think of a Democrat politician at the national level who's gone to jail lately.....I can think of several Republicans. Anybody?
3.14.2008 1:02am
wfjag:
You're right. She and her husband were acquitted. Now they are seeking $200K+ from DOJ for their defense costs.
www.kansascity.com/115/story/492195.html
3.14.2008 12:44pm
NickM (mail) (www):
Serious networked criminality (buying a council, not just a councilmember - think Carson, CA) generally requires a machine political operation, which is normally found in a one-party city.

One-party GOP cities (big or small) just are not that common.

Nick
3.14.2008 3:52pm