An Unplanned Response to Comments about my book, Left Turn

If I had followed my plan for guest blogging, my second post would have discussed (i) my resignation from the UCLA admissions committee, (ii) when I resigned, why I suspected that UCLA—in violation of Prop. 209—was using race as a factor in admissions, and (iii) how an L.A. Times article distorted the truth about UCLA, race, and admissions.  I’ll discuss those topics later, maybe this afternoon.

Except here is one detail I want to give now.   Before resigning, I attended the Martin Luther King Day celebration at my daughter’s elementary school.  At the celebration, Mrs. Baker, my daughter’s teacher explained: “Just as we shouldn’t judge a book by it’s cover, we shouldn’t judge people by the color of their skin.”

My post yesterday, as well as my book in general, seems to have touched a few nerves.  I’m now ready to respond to some of the comments about that post.  Here goes.

Ted says:

Uh. Does anyone expect a book titled Left Turn: How Liberal Media Bias Distorts the American Mind is going to contain legitimate and honest research? I mean, it might, but I don’t think any one would expect it to do so, and more often than not, they would be correct.

If you were serious about the issue, and wanted to actually prove a point via research and analysis, you title the book Media Influence: How Media Bias Affects Consumers’ Beliefs and Actions. That’s something I might read if I wanted to learn something, but judging from the comments so far, that’s not the book that Prof. Groseclose wrote.

Mrs. Baker’s comments, I think, were wise not just in regard to race, but also literally in regard to how one should judge a book.

As for judging the book by the comments of VC readers, I’m certain that very few, if any, of them have actually read the book.  I provide some evidence below.

Dilan Esper says:

I’d be curious where Groseclose gets his funding from and how much it was.

As a general rule, conservative books that have polemical titles with a short catchphrase followed by a colon and which are put out by publishing houses with a long history of publishing conservative titles are generally well-funded and have a function of serving the interests of movement conservativism, not a truth-seeking function. I’d especially be interested in knowing whether, if his “study” had found that there was actually right-wing media bias or that left-wing bias had no effect or was overstated, his financiers would have gone ahead and supported the book.

Generally, if you don’t want to look like a hack, this isn’t the way you go about publishing a study. A peer-reviewed academic journal is a better way to go.

Again, Mrs. Baker, I believe, gives some wise advice.

I hope readers will do a quick search on Amazon.  St. Martin’s Press is not a publishing house like the ones the commenter describes above.  I’m pretty sure that no employee at St. Martin’s has a PQ below 40.

While researching and writing the book, all of the research money and salary that I received came only from the universities that employed me, Stanford and University of California.

The Introduction of the book, as well as many other parts, explain how the book extends a peer-reviewed article by me and Jeff Milyo, an economics professor at the University of Missouri.

Chapter 4 explains how PQ scores are based on a separate peer-reviewed article by Steve Levitt, Jim Snyder, and me.

Lots and lots of the other findings are based upon peer-reviewed research by others.

The Introduction explains my belief that the left tend to be more vicious than the right.  The main reason, as I explain, is that they worship “the god of Equality.”  The above commenter, I believe, illustrates the point beautifully.  If he or she had read the Introduction—and therefore understood how well he or she illustrates the point—then I don’t believe he or she would have written the above comment.

Powerlineblog reprinted the Introduction.  I encourage readers to read it and determine for themselves if the commenter really does illustrate my point so beautifully.  I hope readers will also notice how his or her tactic—to claim, without any evidence, that an author is rigging his conclusions to appease a conservative benefactor—can backfire.  That is, please note (i) how Eric Alterman did that very thing, (ii) how angry it made one of my liberal colleagues, and (iii) how it inspired the colleague to write a long email to Alterman explaining how inappropriate his actions were.

(Also, see this video, where Steve Doocy, of Fox & Friends, interviews me about the incident.)

Randy says:

Additionally, I would note that either the Times or Fox News has rarely, if ever, criticized a Republican President, unless it was to complaint he wasn’t conservative enough.

Chapter 16 examines two facts, both equally true, about George W. Bush’s tax-cut proposal.  One fact was critical of Bush’s policy.  The other was complimentary.  As p. 185 notes, Fox News’ Special Report noted the critical fact 59 times and noted the complimentary fact 28 times.  As p. 186 notes, the Washington Times noted the critical fact 44 times and noted the complimentary fact 18 times.  I’m willing to bet that the above commenter, like the other commenters, did not read the book.

Marcus says:
I also found it odd that there were a number of 60–0 votes on behalf of Democrats in the Senate, but do not recall the Democrats having 60 seats in the Senate. Obviously, the professor thinks of independents as closeted Democrats.

On p. 12, just after I first report the Republican and Democratic vote totals on a roll call vote, I list footnote 2. It says, “For the vote totals on each issue, I count Vermont senator Bernard Sanders and Connecticut senator Joe Lieberman as Democrats. Both are considered officially as Independents; however, both caucus with the Senate Democrats, and are treated as Democrats for the purposes of committee assignments. On p. 16 I note that Joe Lieberman’s PQ as a Democrat was 74.7 On the same page I also note that his PQ, once he became an independent had hardly changed. Namely, it dropped only to 74.0.

Marcus says:
It very well may have been easier for him to just decide to call them Democrats. However, I don’t believe it speaks well of his work if shorthand trumps accuracy. What else did he cut corners on?

Well, I was very upfront about what I did. (See previous answer.) I was trying to minimize words that people would have to read (partly because I had a word limit in my contract with St. Martin’s). I promise I wasn’t trying to fool anybody.

Drew says:
The problem is that it appears to assume any deviation from the baseline is considered “bias.”. This brings to mind Colbert’s point — “reality has a left-wing bias.”
Presumably, on a given policy question (“was welfare reform a success?”) or factual dispute (“is Obama a secret Muslim?”) there is a correct answer. Reported truthfully, Some of those questions will produce a right-wing answer, some of those questions a left wing answer. So to me the best way to judge these things isn’t left/right, but true/false.

In Part II of the book (Chapters 6, 7, 8), I go into great detail to address this criticism. Here again, I’m certain that the commenter has not read the book.

(I also address the criticism on a blog post, entitled “Media Bias: A Response to Stephen Colbert.”)

Martinned says:

That said, this has to be the weirdest survey I’ve ever taken. What’s up with asking my opinion about certain bills twice? And why all that attention for abortion?

As I explain on p. 39, I let a liberal interest group, the Americans for Democratic Action, choose the questions for me. Some of the bills they considered were considered in the House and the Senate. For that reason they are listed twice.

Simon P. says:
I’m disappointed that Eugene has seen fit to promote this guy’s work. You don’t even need to look into the methodology to smell something rotten here — “How Liberal Media Bias Distorts the American Mind”? The very way the title is phrased suggests a flattened perspective that assumes that there’s no “right answer” to political or policy questions, just equally-valid opinions on the right and the left that ought, in the normal course, be equally represented in our “media.”

Like all social scientists who understand the terms, I believe that there are right and wrong answers to positive questions, but not normative questions. The latter, at least partially, always depend on a value judgment.
Once again, Part 2 of the book addresses the above comment. My method examines sets of true statements. However, some of the true statements were ones that liberals would want you to learn. Others were true statements that conservatives would want you to learn. See especially the points above regarding Chapter 16 and the Bush tax cuts.
Finally, the commenter notes, “The very way the title is phrased suggests a flattened perspective.”

Once again, I think Mrs. Baker has some wise advice. To learn the truth about what I do, one needs to read the text of the book, not just the title.

Boom! says:
Let me guess, the ‘left’ is defined by the statements/positions of Congressional Democrats, ignoring the 1/4 of the nation that falls to the left of that? It’s not hard to find ‘slant’, when you redefine the center-left as the left.

I spend at least nine pages (36-44)  discussing how I define “left” and “liberal.” This includes a long discussion of a heated argument with my leftwing friend, Hollywood producer Loucas George. (Yes, there really exists a producer with that name.) Perhaps most important to the above question, as I note on p. 39, liberal and left are defined by the positions of the Americans for Democratic Action.
I think the commenter is mistaken about how many Americans place themselves to the left of the Democratic party. As I note in footnote 8 of Chapter 10: “As I have observed, members of the far left are often guilty of seriously overestimating their numbers. Especially within elite intellectual circles, far-left individuals often believe that a majority of Americans share their view: that the Democratic Party is not left-wing enough. A recent Gallup poll, however, refutes this. It found that 46 percent of Americans think the Democratic Party is to liberal, and 42 percent think it is “about right.” Only 8 percent think the party is too conservative. In contrast, members of the far right would be more justified in holding the mirror view about the Republican Party. As Gallup found, 17 percent of Americans think the Republican party is too liberal. (See Jeffrey M. Jones, “More Americans See Democratic Party as ‘Too Liberal.’” June 30, 2009, available at

Elliott says:
Doesn’t Groseclose assume that the American public’s PQ is the same as the PQ of Members of Congress? I think that’s a pretty questionable assumption (see, e.g., recent polls re: taxing the rich).

In one sense, yes. But not exactly. I devote an entire chapter, “Chapter 5: Defining the Center,” toward explaining this, as well as defending my assumptions.

SecurityGeek says:
From the Weekly Standard review:
In calculating American centrism, Groseclose has made small adjustments to account for the small-state-favoring composition of the Senate and other factors.
Considering that Wyoming has two Senators representing a smaller population than the suburban California county where I live, I think “small adjustments” are not going to cut it.

On p. 49 I note: “I calculated a weighted average of Senate PQs, where the weights were proportional to the population of the state. (That is, for instance, California’s weight is approximately sixty times that of North Dakota.)

Mike says:
Professor Groseclose — What steps have you taken to ensure that your political views don’t interfere with or unduly affect your research and work?

I devote an entire chapter, “Chapter 3: But I’ve Been to Oklahoma,” to that question.

Finally, let me address two comments that I do not already address in the book.

frankcross says:
It may not matter a great deal to results, but I think it’s a bad violation of research protocol to put the legislator’s votes on the PQ questions. What possible reason could there be for this?

I wanted readers, as best as possible, to have the same information that legislators had when they cast their votes. For the typical roll call, the legislators have at least 15 minutes to vote. But after the 15 minutes expire, the Speaker or the Senate Pro Tempore usually allows several more minutes. And usually (in fact, I’m aware of only two violations of this), before ending the time for voting, the Speaker or President Pro Tempore asks, “Is there anyone in the chamber who has not voted or who wishes to change his or her vote?” Accordingly, at that moment—which is the moment legislators make their final decision, they can see the votes of all their fellow legislators. Thus, legislators have that information—the way the two parties have voted—when they make their final choice.

Another reason involved the summarizing of the bills on which the legislators voted. Often, if I listed only a few-sentence summary, a reader would naturally think, “Of course everyone would favor that bill.” The problem would be that the summary would not contain several riders that might cause legislators to vote against the bill. I listed party vote totals to show that at least some people would not feel that way—that there were reasons that reasonable people might oppose the measure, even if the summary might suggest otherwise. Of course, maybe a better way would be to summarize all the riders, as well as the main part of the bill. However, if I did that some of the summaries would be several pages. That, I believe, would be too taxing on the quiz taker – I wouldn’t have many people finish my quiz.

Marcus says:
Oh yeah, this should be a hoot. And I love the “That’s right, I nailed you libs good” look on the professor’s face on the banner book advertisement.

I promise that thought wasn’t going through my head when the picture was taken. Here again, I think Mrs. Baker has some wise advice.

Powered by WordPress. Designed by Woo Themes