In “IUUI 4 followup” on June 10, 2013, I noted that the possibility of doing a book about everyday statistics, and a related book showing librarians step-by-step how to gain useful information from IMLS and NCES statistics without (a) becoming statisticians, (b) going crazy or (c) even having access to Access (see outline here) was still very much up in the air.

I closed the followup post with this:

My sensible side says there’s just not enough interest to make this worth doing.

My other side keeps wondering whether I could do a good enough job that it would get the word-of-mouth marketing that self-pub books really require (unless you’re ready to spend serious dough).

I think where things stand is that I might try writing the first two chapters and see whether they point to something I’d be proud of and believed would both be short enough to appeal to people and useful enough to satisfy them and me.

### I gave it a shot…

I did try writing the first two chapters of “Mostly Numbers,” a slightly revised title for the “general everyday statistics” part of the project.

And failed.

Which doesn’t mean that I think the idea’s useless. But it apparently won’t work for me, at least at this point. My difficulty in even writing draft chapters in an area I know well says that it isn’t meant to be. I found myself doing almost *anything* else rather than focusing on this.

Maybe it’s because it really isn’t a learning process in this case. Maybe it’s because, the more I looked at the issues with “misleading graphics,” the more tentative I became–there’s a huge gray area between intentionally misleading graphics (e.g., the crap NEA pulled years ago in trying to prove that Americans don’t read) and choosing techniques that emphasize a point without actually misleading.

Maybe it’s because I didn’t think I could do a good job of it in a small enough space to make it attractive–and really didn’t think I could market it well enough to get back subminimum wage for the effort (e.g., at least $3.50 an hour!).

So that one’s on the back burner, at least until various other projects are complete, which is likely to mean March 2014 at the earliest.

### Then there’s the library part…

I haven’t *quite* given up on the book specifically targeting academic and public librarians, or rather a shorter and simpler version of that book. Here’s sort of what this might look at. Let’s still call it “Mostly Numbers” with a subtitle “Coping with Library Statistics.”

- Introduction
- Why Everyday Statistics are Mostly Numbers
- Doing Statistics Right: Transparency and Ethics
- Fair Presentations and Coping with Outliers
- Everyday Statistics: The Terms You Need to Know
- The Other Terms You’ll Encounter
- The Tests You Can Probably Ignore
- The Tools I’m Using for This Book
- Using Excel to Expand Your Public Library Awareness
- Using Excel to Expand Your Academic Library Awareness

I’m not sure this one works either. Again, I might try writing a chapter or two. The last two chapters may be the most helpful/useful. I’m not sure.

Walt, I can relate to your ambivalence on tackling this project. I have been conducting a short series of training sessions on the very basics of statistics for librarians. While I receive praise for taking this on, it is not well attended (about 12-15 (out of about 50 professional-level staff) per session) and I’m not very hopeful that it will inspire more than a few to look at statistics more carefully.

However, I do like your approach of using the data sets already available to provide some reality-checking and context for our own library’s data. Perhaps a “traditional” ebook (PDF) is not the right medium for this…What about something more interactive, like a series of presentations accompanied by demonstrations or exercises using the data? I’m not sure what the business model would be – charge the same fee for access to a site? Downloadable software? Something more akin to an online, interactive textbook?

There are quite a few “statistics for dummies” or “statistics for non-math geeks” out there, primarily because statistical literacy is less about the math and more about the use of logic and reasoning. But there are fewer of these books aimed at librarians, particularly public librarians. The trick is to convince them that they will benefit from using this, and I haven’t figured out how to do that.

Keep us informed.

Karen, Thanks for the comment. Another librarian friend suggested a webinar. Unfortunately, I really don’t think I’m the right person with the right tools and right connections to do anything along those lines.

I’m guessing your training sessions provide more real-world info on statistics in libraries than I would. The best I can hope for is to show reluctant librarians a few tricks of the trade. (I’ve written a draft first chapter, which really doesn’t help me decide whether to do the project, and it includes a line like this: “If you already know how to build a VLOOKUP statement, you probably don’t need this book.”)

So I’m still up in the air. Worse yet: IMLS just released the 2011 dataset–and it’s now in Excel or CSV mode, eliminating one tricky conversion step. (For some bizarro reason, the Excel version is three times as large as the CSV version when converted to Excel or any previous .MDB version converted to Excel.) So: I could be spending my time working on actual research, or I could be trying to help people do their own. The latter is looking like a tough row for a retired introvert with no library-consortium connections to hoe.