Well, That's Something

Many of us--OK, the few of us who cherish this sort of thing--were horrified to learn that the Census Bureau was ending publication of that huge and magnificent source of information, the Statistical Abstract of the United States.

We can be a bit relieved at the news that ProQuest has announced it will be taking over production of the report beginning with the 2013 edition. No word (that I could find) on how much it will cost, although the press release notes that the online and digital versions will be available for pre-sale starting next month.

It's hard to see it as a totally awesome development, with everyone's tight library budgets, if we now have to purchase something that used to be included in the cost of our taxes, but I guess it's better to have that as an option than it is to have this resource disappear entirely.

So thanks, ProQuest. I'm glad someone's doing it.

I initially saw this on LISNews.

That's a Lot of Ones

It's 11-11-11. That seems momentous somehow. Babies born today are going to have an easy time filling out their date of birth on forms. I feel kind of bad for the ones who just barely missed it, though, and were born yesterday, or will be born tomorrow.

Sorry kids!

It's also Veterans Day, of course. Veterans everywhere, thank you for your service.

And now, I should get back to my writing.

Statistics. Gotta Have 'Em.

Speaking of the Statistical Abstract of the United States, which I often do at length and with passion, you should check out this report from Library Journal, which explains the sad situation about how it has been proffered up for discontinuation in a recent budget proposal.

We are going to miss those numbers later if we don't have them, you know.

The article has links to a Facebook group you can join and an online petition you can sign, if you're into that kind of thing.

.

Wildlife and Data

Two things make my heart beat faster this evening:

The Library of Congress Blog has an update on the hawk-in-the-reading-room situation (with, sadly, fewer terrible puns).

And More Grumbine Science refers us to an awesome project that involves transcribing climate data. It's called Data Rescue at Home, and it involves looking at a picture of a page of handwritten numbers, and copying them in type.

It aims to collect lots of historical temperature data into a format that can be easily worked with. The pages have all been scanned, but the scribbled numbers are too hard for the optical character recognition programs to handle, so they want a human eyeball to examine the pictures and indicate what the numbers are.

It's basically like doing a bunch of CAPTCHAs all at once...but for a good cause! You'll be advancing the frontiers of science and such.

You can choose to work on either "German Radiosonde Data from the Second World War" or "Tulagi (Solomon Islands), station data." All the temperatures I've seen so far have been quite warm, which could be good or bad considering it's the dead of winter right now where I am.

Am I comforted, or filled with jealous rage, to think of 80-degree weather while it freezes outside?

It's actually kind of challenging. Some of those people had really bad handwriting.

"Is that a 1 or a squished-up 9?"

I can't do science, but I can participate in science.

.

Other Peoples' Statistics

I was sympathetic to the news in this article, San Francisco Library Doesn’t Know How Many Books it Has.

Because you know, it's surprisingly hard to get a count for that. The story says that the San Francisco Public Library can provide a number for how many titles they hold, but not how many specific items, and I myself have wrestled with this little distinction.

We don't have nearly as many books as the SFPL where I work, and it's still hard to get a good number. I was especially sympathetic when I learned that SFPL is using Millennium, which is the system we use for our OPAC. Ah, Millennium.

Hearken to these quoted words from a library spokesperson, who reports that Millennium is a "clunky system that works great for circulation but it's hard to pull out metrics that are newsworthy."

Yeah, that sounds about right. Except I can't speak to its efficacy for circulation, since I don't work on that aspect of things myself. Clunky, though, I can attest.

Now I can count items, because our Millennium gives us a Bibliographic record (a description of a specific title, edition, etc.) and an Item record (saying that we have a specific individual copy of said title, edition, etc.).

So I can run a list to pull all of the Item records that have our library's code in them (to differentiate them from all the items that might be in other libraries within the university system), and get a number that way.

But sometimes you might want to include also the number of bound volumes of old journals that you have, and there's no item for that, and sometimes you might want to count, or not count, electronic versions of titles, and there's an item for that but it's not easy to separate them out, and sometimes we forget to include our library code in the correct format (I'm sorry, but it happens), and sometimes depending on the definition you might want to include or not include titles available in other parts of the library system, and it all gets very confusing.

So I'm not sure exactly what issues the SFPL faces in coming up with a specific number of items, but I am sympathetic.

Also, I bet it's a lot. Just go with that, SFPL. "A whole lot."

I saw this on a roundup at This Ain't Livin'.

.

Treating Numbers

I'm going to have to check out this Evidence-Based Medicine-related site: the Number Needed to Treat, or NNT.

It has some good explanatory material about the statistics involved in calculating whether a treatment is likely to be beneficial or harmful, and quick summaries of various treatments, presented in the form of information on how many patients who use it are helped or harmed, using both percentages and numbers like "one in eight."

There's also a nifty, traffic-light-style code indicating whether the evidence suggest that an intervention is likely to be beneficial, neutral/unknown or harmful.

It seems to be limited by, well, a limited number of interventions covered at the moment, at least as judged by my unsuccessful searches for some of the example conditions we use in our searching-skills classes, but it looks like a cool resource for EBM concepts generally, and for those topics it does cover specifically.

My thanks to GruntDoc, who recommended it.

.

Yet Another Nifty Date

For those who enjoy entertainingly numbered dates, I cannot help but notice that today is 10-20-2010. Or 20-10-2010 if you use the European system.

Either way, it's pretty fantastic, right? We should all start baking celebratory numeral cakes, or whatever we're supposed to do to recognize entertaining dates.

A nice pie, maybe?

.

Fear Not Numbers

Apparently girls can do math. Now you tell me, after I've gone and gotten these numerically challenged degrees in English and library science.

I salute WomensHealth.gov for highlighting this study--actually a review of 242 other studies. As we tell students in our EBM-related classes all the time, reviews are good! Saves you having to read all those 242 studies yourself.

Because we may be able to do math, but we have busy lives and are often short on time.

Now, girls in school, get some practice crunching numbers. If I were 12 right now, I would be all over that.

It will serve you well once the Association of Research Libraries comes asking for statistics on your collection, you know. And you may think it couldn't happen to you, but don't be so sure. The ARL has a long reach.

.

Sometimes You Want the Numbers

A good point on FWD/Forward about how touch screens can be problematic for people who can't see them well.

If you want to pay with a debit card at the grocery store, but you can't see the numbers on the touch screen, what are you supposed to do? Tell someone your PIN so they can type it in for you, and then also potentially use it to take your money? Hmm.

I mean, I suppose in the grand scheme of things the likelihood of someone memorizing your PIN and then also happening to get hold of your card is small, but still, I certainly don't go around telling people mine. And if I did have to tell someone for some reason, I'd feel compelled to change it as soon as possible.

Which should not be considered a reasonable solution for people who can't see touch screens, since having to keep changing your PIN, and remembering a new one every time you go shopping, is few peoples' idea of convenient.

.

It's All Numbers, Numbers!

Ugh. I'm working on those soul-crushingly hideous number questions, so dreadful that merely to contemplate them would make Santa Claus himself vomit with rage: ARL statistics.

Shudder. You know, those questions always seem so innocent and straightforward, and then you start trying to count things and you realize that apparently we have no means of accurately keeping track of anything!

How many journals do we currently receive?

Do we count all the stuff in those giant packages that we linked on the web site because, well, who knows when someone's going to need to consult Waterbirds for a waterbird-related health question, so we stick it on the e-journals page, but we haven't gotten around to putting it in the OPAC yet, because we don't know, but we can guess, and while our guess must perforce be a non-never date given the impossibility of absolutely ruling out the arising of some waterbird-related health issue, we still don't think it's going to happen anytime soon. I'm going to catalog them, eventually, but it's just not been at the top of my list.

So I ask you, do we consider these 'currently received'?

I have to warn you, I will be highly dissatisfied with any answer, because I am sick of this and all related questions.

.

Careful with the Numbers

Useful little post on The Health Care Blog about the importance of understanding statistics, and how few people actually do.

I have the classic How to Lie With Statistics in my office somewhere, although I haven't gotten around to reading it, and I've never taken a statistics class, but I know I should. It's infamously possible to make all kinds of arguments seem plausible using cleverly spun statistics, and I should know how to do it!

How else will I further my plans to achieve world domination via misdirection?

.

09-09-09

I will probably fail to post precisely at 09:09:09, since I don't have a second timer handy on this draft...but there you are. I did my best to appropriately commemorate this awesome occasion, as I planned.

.