If I do a new version of Give Us a Dollar and We’ll Give You Back Four, it will be quite a big different from the current one (here are links to the PDF and the hardbound versions)…and the current one will probably remain available for a while. I’d take advantage of some of the work in Graphing Public Library Benefits.
The changes can be summarized as Simplify, Amplify, Clarify and Compare.
I now believe that I included too many different metrics and too many divisions for key metrics in the current version–”too many” in that they may obscure the overall picture of America’s public libraries, but also in that the sheer number of tables and length of the book may intimidate some potential readers/users. I also believe that, while theoretically desirable, basing divisions purely on reality may not work out as well as I’d like.
Here’s what I have in mind for a new version, subject to revision:
- Spending brackets: Reduce from the current 10 to, probably, five–in part because it’s possible to make charts with five lines that can be read in black-and-white (using different line dot-and-dash combinations), while I don’t think that’s true for 10. The brackets would probably be based on the median per capita spending and would be something like this: A. <1/3 of median. B. 1/3 to 2/3 of median. C. 2/3 to 1 1/3 of median. D. 1 1/3 to twice median. E. More than twice median.
- Size (LSA) brackets: Reduce from the current 18 to, probably, nine, with one bracket each for libraries serving fewer than 1,000 people and those serving at least 100,000, and seven others based on actual distribution (looking at roughly 1,000 libraries per section).
- Other metrics: Include circulation per capita (reducing current nine brackets to maybe six), reference per capita (reducing from ten brackets to maybe six), patron visits per capita (reducing from nine to maybe six), program attendance per capita (reducing from eight to maybe six), PC use per capita (reducing from eight to maybe six) and visitors per hour (reducing from nine to maybe six). Omitted from detailed metrics: hours open (but see below), total PCs, PCs per thousand patrons and circulation per hour.
- I’d still have the benefit ratio, probably calculated very similarly, used as appropriate.
The overall net effect is that a given library would be comparable to around 200 other libraries for spending. or around 166 for other metrics. And that most graphs would involve around 1,000 libraries (but I’d probably remove the top 10% from some graphs.)
The new version would be amplified from the current in several ways:
- I would not exclude libraries with very low funding, those with very high funding, and those with less than 0.25 FTE librarian. I would still exclude territorial libraries, closed libraries and libraries with no reported operating expenditures.
- The new version would include graphs as well as tables, as appropriate.
- Rather than peculiar “combined tables” showing quartiles for given metrics at different expenditure levels, there would be single tables, one for each metric–and I’d use the extra space to add 10%ile and 90%ile to the current Q1 (25%ile), median (50%ile) and Q3 (75%ile) figures. That would offer a much better picture of what’s out there, while still ignoring extreme cases.
- I would include correlations as appropriate (as I do in GPLB).
The current version is, how you say, light on textual commentary. Once you get past page 21, it’s basically nothing but tables.
Which, as a pure tool, may make sense–but is a little overwhelming.
The new version will include some commentary, pointing up noteworthy items in the tables and graphs, providing at least a little textual clarity.
The current version looks at one year. While I do suggest that it’s likely that more money would yield better and more numbers, I don’t have any hard evidence for that.
The new version would compare 2010 and 2011 figures (and would include only libraries present in both years). It would also attempt to show correlations between changes in spending per capita and various other metrics. I would probably include changes in total open hours here.
Oh, and one other change–if this happens at all and if it makes sense:
I’d split the state-by-state sections out into a separate book, and those sections would include some comparisons to overall figures that aren’t there now. That would make the separate book an interesting overview of differences in metrics across the nation.
Best guess as to length (the current book is 262 pages; Graphing Public Library Benefits is 222 pages): Somewhere around 150-200 pages, ideally closer to the first, for the main book; probably around 200 pages, maybe more, for the “Viewing the States” book.
Price would be $9.99 for PDF, whatever it works out to for paperback (probably around $15.50 if it’s 150 pages, around $16.50 if it’s 200 pages), $40 for site-license or state-license (for the state-by-state) ebook version without usage restrictions.
No, I still don’t know whether it makes any sense to try a Kickstarter or IndieGoGo campaign to prefund this book, possibly with a stretch goal of making the PDF version free. I also still don’t know whether I’d do this. Since the new figures should show up in July, I’m coming close to a decision.
If this helps you think about these issues, you can still respond to the survey.