After another delicious and friendly breakfast at Dutch Manor Inn (I feel sorry for all of you delegated staying at the Westin!), I headed down to catch the long morning session on evidence and library impacts.

Got to hear the tail end of Fiona Bradley’s talk on the Sustainable Development Goals.  Then David Streatfield and Sharon Markless, a team of Global Libraries impact assessment consultants, gave an overview of their work.  Unfortunately, appears that Global Libraries did not build in any randomized trials for evaluating their ICT buildout to whole library systems.  Maybe Global Libraries was not really sure what to be measuring.  or maybe worried that conclusion might be: “public access to Internet in every library in Chile enabled lots more people to play online poker and surf you-know-what.”  Probably though, since digital/Internet access is framed in rights-talk, then cost-benefit really did not matter, and there was no compelling reason to waste resources in measurement.  That is, there probably was not going to be any likely measurement, after a pilot, that would have led Gates Foundation to say: “Let’s NOT help libraries have Internet access.”

Jessica Dorr of the Bill & Melinda Gates Foundation then spoke on the Global Libraries data portal Global Libraries Atlas.  This is a very worthy effort and the site is fantastic.  But I am of two minds about data visualization work like this.  Very expensive, I imagine, to build the interactive website.  And since no measurement of benefits (either) of usage of such a website, not really clear that money is well-spent.  How do we know this is better than just posting the an Excel file and letting researchers write R scripts to visualize?  Is this data visualization really going to help in advocacy?  Is advocacy really more successful because the advocate has a great infographic?  Does the Gates Foundation itself make funding decisions based on infographics?  And while we are on that subject, I wish glatlas.org would just have a button to download the whole dataset.  Seems like that should be gold standard for transparency and open data.

Finally Karin de Jager spoke about ISO standard for measuring library impact.  She gave a nice overview of the issues, and it sounds like a very good resource.

The questions and discussion that followed rehashed some basic debates about what should library advocates be looking for. To an economist, I hate to say this, the debate sounded like something from 50 years ago.  Yes, data does tell the story, and the best impact evaluation studies in the social sciences do indeed let the statistics talk.  Good impact evaluations do not need anecdotes.  Indeed, since we know just how powerful story-anecdotes are, we should be extra careful about using them, because they easily lead us to bias.

Part of me prefers to be explicit: If you want a warm-glow story to make people feel good about donating, or convince people to allocate tax dollars to the purpose, just say so, and don’t call that impact evaluation.  But don’t say that a set of non-randomly selected stories (chosen by you to resonate with donors and advocates) is impact evaluation.  I support libraries and reading because I like libraries and reading.  Libraries enable lots of people to read a lot more than they would otherwise (though surprising how little we know about that basic effect, but all the small-scale studies we have done in Burkina Faso find large increases in reading).  Does reading more books then do something else, other than increase vocabulary?  There is little social science research to answer that question.

The “Beyond Access” and “libraries for development” thrust is trying to build an evidence-base and rhetoric that can shout an enthusiastic YES to that question.  It would be great if more research funding comes along so that the claims can be evaluated and replicated.  But social scientists cannot even agree on measuring the effects of deworming pills…. and extra reading is quite different from a pill.

Got a tweet from Sarah Jaffe, Senior Manager of Research at Worldreader, so we had a nice chat over coffee.  They are expanding rapidly, soon to be putting Kindles (not sure exactly how many- sounds like 10 or 20) in every Kenyan public library (there are about 50).  Would be neat to explore whether availability of e-readers turns out to be complementary to paper book sales.

After that I heard a great presentation by Bookdash.  Wish we could do something like that in Burkina Faso, but it sounds like something for a rich country… costs about $10,000 for the event itself, where 30 illustrators, editors and authors come together for an all day session to make 10 children’s books.  I am very partisan to that approach: make them quickly and get them printed and available.  In Burkina Faso, the problem is we have absolutely no institutional buyers or sponsors.

Then walked around downtown Cape Town with Hui-Yun Sung, a library science researcher from National Chung Hsing University in Taiwan.  Found the tourist market and had fun bargaining a bit.  Lots of the usual stuff: beaded elephants and masks.

Had dinner later at Bocca near my hotel. Very good Neapolitan pizza!