Collection Bias for Conservative and Liberal Materials

Tomeboy has an interesting article at his website that analyzes collection development of liberal and conservative materials in libraries. Using WorldCat and other sources to compare library collections he comes to some interesting conclusions.

Comments

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Tomeboy's article

I'm hoping Michelle Malkin or Amy Ridenour pick up on his article. I've blogged about it today. If librarians truly want community support they need to do a better job of meeting the needs and interests of the mainstream reader.

suspicious methods & weak operational definiti

"223 to 1, according to the latest Library Journal, (10/04) is the ratio of librarians that have contributed to John Kerry’s campaign as opposed to Dubya. Again 223 to 1. Hardly a ringing endorsement for a profession hell bent on diversity."

So, if we're committed to diversity, more of us should join a political party that isn't? Huh? I almost stopped reading there, but tried a little more...

"This is not intended to be a scientific study... My sample was random."

I'm suspicious of anyone claiming they took a "random" sample, especially when they have a declared bias. Was every 10th book taken, or were perhaps only certain titles that may help skew the results selected? Probably not, but unless a truly random sample is taken, that's a possibility.

Lastly, and this is the biggest flaw with this data, the number of weeks on a bestseller list or the highest rank achieved may be lowly correlated with the demand for a particular title.

Book A could debut at #1, sell 10 million copies, then drop to #151 the next week, while Book B could hang at #150 for two years while selling 1 million copies. Which title deserves a higher holdings/score rank?

Gross sales would be a better number, but even that doesn't always match the demand for a book. Some books people are more likely to buy than borrow from libraries or friends to read. Or supposing that liberals use libraries more, there's nothing wrong with these holdings numbers; librarians are merely meeting the needs of their users.

And how many of these titles are still on the USA Today list, thus making the calculation of their ranks incomplete?

All these nitpicks aside, there probably is some truth in this guy's conclusions. I just wish the research logic were a little tighter.

Re:Tomeboy's article

Malkin? THE Michelle Malkin, legendary for her tantrums and incoherence? Norma, please tell us this is actually an example of your sly (and, usually, appreciated) sense of humor!

Re:suspicious methods & weak operational defin

Thanks for reading John.

Couple of points.

Random, for the purpose of my collection sample was just that, random. It’s your call on whether you believe this or not.

Lastly, and this is the biggest flaw with this data, the number of weeks on a bestseller list or the highest rank achieved may be lowly correlated with the demand for a particular title. Book A could debut at #1, sell 10 million copies, then drop to #151 the next week, while Book B could hang at #150 for two years while selling 1 million copies.

Demand in this context is directly related to sales e.g. the Top 150 list. Your example of a book dropping from #1 to #150 is unrealistic. If you have any examples of this happening please post. And yes, Book B could hang on for two years hence my inclusion of the “Holding to Weeks on the Top 150� table.

Re:Tomeboy's article

"tantrums and incoherence"? Not the Malkin I read.

Re:Tomeboy's article

I'm sly and coy, but eclectic. I enjoy her writing style because she drags us in with tons of information. Also I like that Steyn guy, who make Coulter and Malkin look like simpering kittens. Malkin is also a member of a minority ethnic group, providing us with much needed diversity in the media which is generally pretty WASPish.

Re:the mainstream reader.

Are you equating conservative and mainstream here?

Some Other Ideas

I'm sure that you're on to something, but I think you're methodology needs some work and you'll do better by revising your methodology a bit. Things to consider:1. Political books are much more popular now that years ago, something else to consider is the year published.2. Choose similar titles and compare them. E.g. Bush At War vs. The Savage Nation BAD choice to single out. Savage and Woodward? Easily disputed. Maybe Savage Vs. Moore is a better comparison.3. Try running some other numbers as well, raw holdings, circulation #'s, there must be some other interesting stuff to pull out.4. Increase your titles, or change them , limit them to books you'd consider "similar"5. Can you find information on political leanings of library users?6. Can you find any correlation between the circulation of a book and being on the best seller list?7. Do libraries have titles at the same time they're on the best sellers list? Does the # of titles held at a library affect circulation?People will take issue with your designations here, not sure that's even avoidable.You didn't ask, but my 2 cents:I'll agree with John, the "Hardly a ringing endorsement for a profession hell bent on diversity" isn't necessary or helpful. What you're doing is an interesting idea and embedded attacks isn't really opening any closed minds. If you're just trying to get into the echo chamber with another attack piece that's fine, but if you're really trying to do an objective look at subjectivity in the profession, it's a worthy pursuit, and a publishable one."Ideally the relationship between the two, holdings to sales, should remain roughly the same regardless of content."Expand on that, why do you believe that? I take that as a big part of your comparison and I'm not sure I agree with you. Why is the best seller list an accurate comparison? I'm afraid here that a library doesn't get a title until it's off the best seller list, or libraries just don't circulate the same books as on the list in the same numbers. I don't know, I just think it's important to clarify that area.

Research? Bias?

Librarians should know what these words mean. The author of this article does not.

Re:Some Other Ideas

"Expand on that, why do you believe that? I take that as a big part of your comparison and I'm not sure I agree with you. Why is the best seller list an accurate comparison? I'm afraid here that a library doesn't get a title until it's off the best seller list, or libraries just don't circulate the same books as on the list in the same numbers. I don't know, I just think it's important to clarify that area."

My experience with public libraries is that Bestsellers are an automatic purchase. No doubt some libraries do not operate that way but there had to be some reason why those libraries on WorldCat who bought the liberal end books did not buy conservative ones.

I've heard statements in the past that said conservatives are more likely to buy their own copy then go to the library. If the arguement
is then that libraries simply don't get requests for those kind of books why are the libraries simply giving up on half the population?

Re:Some Other Ideas

Blake – Just have a few minutes to reply here. More later.

. Political books are much more popular now that years ago, something else to consider is the year published.

Two questions. What do you base this on? And more importantly, how would this skew a sample taken from the same time?

Choose similar titles and compare them. E.g. Bush At War vs. The Savage Nation BAD choice to single out. Savage and Woodward? Easily disputed. Maybe Savage Vs. Moore is a better comparison.

I have chosen similar titles, those with a conservative bias v. those with a liberal bias. Comparisons from this part are made on similar sales and rankings. Taking your advise here would only leave this “informal study� open to more questions.

Try running some other numbers as well, raw holdings, circulation #'s, there must be some other interesting stuff to pull out.

That’s not my point here. I’m looking to see the disparity between public demand (sales) and library holdings.

Increase your titles, or change them , limit them to books you'd consider "similar".

See #1

Can you find information on political leanings of library users?

We are public institutions (pub libraries that is). Discerning our leanings would be to look at the latest election results. One may also argue that library users may be skewed to show more liberal patrons because their libraries cater more to their interests?

. Can you find any correlation between the circulation of a book and being on the best seller list?

Aside from personal experience? Again, I’m looking for public demand v holdings.

Do libraries have titles at the same time they're on the best sellers list? Does the # of titles held at a library affect circulation?Don’t know. I don’t see the relevance here Blake.

As for my diversity statement, I apologize for any offense taken. My question. What kind of response could we expect with if that ratio was in favor of Bush??

Here’s the rub Blake. There should be a correlation between sales and holdings. More popular books are more popular because they are purchased. This includes library acquisitions. There is a disparity here between the holdings of popular conservative books and liberal books. I’m not sure tweaking methodologies here is going to change much. I may be wrong and certainly encourage others to look into this. But here’s the most important question Blake “Is there any truth to liberal collection bias�?

Re:Research? Bias?

I could be wrong but I bet Tomeboy was working with a limited budget and limited time. Funding to conduct a scientific study would be helpful. Maybe ALA would foot the bill for this?

Re:Some Other Ideas

"There has been an increased appetite for political books, and publishers are responding by getting more of these titles into stores. So far this year, 50 political books have made USA TODAY's top 150 Best-Selling Books list. That's about three times as many as in all of 2000, the year of the last presidential election." Source. That's just the latest, we've run other stuff. Politcal books are selling more now than a few years ago.I don't know how or if would this skew a sample taken from the same time, it's just something to consider if this is pursued further. Comparing a book from 2001 to one from 2004 isn't going to be a good comparison."What kind of response could we expect with if that ratio was in favor of Bush??"No idea, don't care, doesn't matter."Is there any truth to liberal collection bias"? There's truth there, no doubt.Like I said, it's an interesting topic and one worth pursuing with good research that expands on the ideas you have brought up. I don't raise them to try and discredit you, but rather to raise further questions that might be worth pursuing.

Re:the mainstream reader.

If we have polls showing about a 50-50 split between our candidates then someone isn't being served in the collection development, and I suspect it is a large group in the middle--maybe people who might want to read two or three points of view?

People's requests only need to be turned down a few times to have them stop making suggestions. That leaves the field wide open for the librarians to conclude that they know the community's needs best.

Similar titles good idea

I confess to simple skimming of your article, but it does have intriguing ideas worthy of future research - other librarians, ALA, whoever.I have to agree with Blake that in addition to the liberal vs conservative bias, a better study would compare lib books on a topic with conservative books on the same topic, i.e.:Iraq WarRace ProblemsEducationHealth of Culture, etc.This would eliminate the effect of books not being widely held due to topic.I have to agree with you that ideally, under the modern model of librarianship, there should be a correlation between holdings and sales.Earlier models of librarianship put us in the position of cultural arbiters selecting the "best books", which I suspect led to an inverse relationship between sales and holdings.I still believe that librarians have a role in picking interesting, informing, or entertaining books that may not have the publicity to become best sellers. Otherwise we should all go home and let McNaughton lease bestsellers to individuals! A Netflix for books?A tip of the hat to bibliofuture for promoting your journal entry to the LIS story that it is. If people really dispute you, let them run their own studies.

Re:Some Other Ideas

I *think* that some of the bias might be explained by the fact that 8 of the "Liberal" books were in the top 5 bestselling books (4 at #1, 2 at #2, 1 @ #3, 1 @ #4), whereas only 3 of the "Conservative" books were in the top 5 bestsellers (1 at #1, and 2 at #5).

If libraries use bestseller lists as purchasing guides, I imagine the ranking on the list will impact what is purchased, the higher the rank, the more likely to be purchased (I don't imagine most libraries purchase all 150 titles on the USA Today list, but maybe all 5 titles on the New York Times list - http://www.nytimes.com/pages/books/bestseller/)

It would be interesting to see if controlling for bestselling rank (only look at the books ranked #10th or higher for example) gives a different picture.

(Yes, I could work it out myself, but I'm lazy!)

Interesting snapshot though, and definitely food for thought.

Re:Some Other Ideas

Geekster - There is little change when narrowing to the top 10. (I’m too lazy to type all titles here so I’m just going to give the totals)

The 8 liberal books in the top 10 have a composite score of 16 (124/8) on the “Holding to Rank Scale�. The top 6 conservative titles have a composite score of 10 (59/6), which is a 60% spread.

Thanks for reading.

Re:Research? Bias?

I think he was also working with limited comprehension and/or intellectual honesty.

Tomeboy equates criticism of the current administration with "liberal bias". A tactic he may have learned by watching FOX News. He demonstrates this with his comment about the ratio of librarians who donated to Kerry vs those who donated to Bush. He further demonstrates this with his "random" list of books. To Tomeboy, these books are obvious in their leanings which gives us some insight into how deeply he thinks about such things.

He uses his numbers to say exactly what he wanted them to say (a deeper look, I suspect, would show that they don't really say much at all). No consideration is given to what else they could mean. "No special reason" why he chose the year 2000 as a stopping point, Tomeboy claims, though it happens to be when Bush was (s)elected.

He seems to think that one book is as good as another the only pertinent difference being the perceived partisan leanings. No thought given to considerations of the quality of content. Librarians read book reviews, and some even read the books they buy for their collections (*Gasp!*). Some librarians even understand what real journalism is, which could be why Bob Woodward's book is found in more libraries than books by Hannity or O'Reilly (even despite the criticism the Woodward book has received for the quality of journalism it demonstrated, I might add). I'm sure the new Seymour Hersh book will provide you with more numbers to prop up your claims. But I guess you can chalk that up to those darn biased librarians.

There is no reason library holdings should correlate to best-sellers lists. In fact, I would consider a strong correlation possible evidence that librarians weren't practicing good collection development. Libraries constitute one kind of information provider in a larger marketplace of information, this needs to be considered in collection development and any investigation of so-called bias in collection development.

In closing, biases in collections certainly exist and whether that is a good or bad thing is debatable on a case by case basis. This article, however, offers no useful insight into the issue and is designed to support an ideology to which the author happens to subscribe.

jimbo

Re:Research? Bias?

Your comments seem as spiteful as his.
Examples:
A tactic he may have learned by watching FOX News.
Some librarians even understand what real journalism is
some insight into how deeply he thinks about such things

Also stated by jimbo, "This article, however, offers no useful insight into the issue"
So why don't you put some numbers together and show us what results you come up with. Clearly Tomeboy's study is not up to the scientific standards of a double blind drug study but I believe it raises some interesting questions. Instead of critisizing his methods why don't you do your own study and discuss your results. It would be hard I admit but Tomeboy got past that difficulty.

Re:Research? Bias?

Your comments seem as spiteful as his.

Not so much spite towards him personally as contempt for what he was trying to accomplish.

So why don't you put some numbers together and show us what results you come up with. Clearly Tomeboy's study is not up to the scientific standards of a double blind drug study but I believe it raises some interesting questions. Instead of critisizing his methods why don't you do your own study and discuss your results. It would be hard I admit but Tomeboy got past that difficulty.

Far from being up to the "scientific standards of a double blind drug study", Tomeboy's effort and purpose is the opposite of a scientific study. Rather than uncovering truth, his purpose was to obscure truth. Maybe I am assuming too much, like Tomeboy does in his article. Is Tomeboy a librarian? Is he experienced with scientific inquiry? Has he even taken a logic class? What I can do is totally irrelevant, what Tomeboy has done is what is in question here. Not only does his article provide no useful insight, it is detrimental to the overall discussion.

jimbo

Syndicate content