In Online Information, Ebooks, and Moral Panic, Rick Anderson asks:
What percentage of poor children has access to a decent public library collection, as compared to the percentage that has access to the Internet?
It is a very important question and gets at the kind of information librarians need as we make decisions about how to provide the best possible access to information to the widest possible set of people.
Some time ago, I joked on twitter “Never ask a librarian a hypothetical question … you are likely to get a real answer … with citations.” I guess I should amend that to “Never ask a hypothetical question on a library blog … you’re likely to get a real answer …with data.”
So yes, let’s see if we can answer that question. First, we need to operationalize our concepts in ways that render them measurable. Easier said than done, it turns out; but let’s give it a go.
If the point of the question is to determine whether an ebook or other online information is more accessible to low-income patrons than print, then I think it is fair to operationalize “access to the Internet” as “access to the Internet at home”. In my opinion that is a rather conservative measure of access to the Internet, since many people (especially kids) in low-income families share computers, making it more difficult to read long-form information than for those who own their own computer or e-reader. But let’s go with “access to the Internet at home” as our measure of access to e-books and other online information.
According to a June 2013 report prepared by the National Telecommunications and Information Administration, the Economics and Statistics Administration, and the US Department of Commerce: “Just under half (46 percent) of households with incomes of less than $25,000 access the Internet at home.”
Operationalizing “access to a decent public library collection” is even more difficult. In theory, everyone has access to public libraries, right? But meaningful access really means a library that can be accessed via a reasonable commute – preferably walking or safe, affordable public transportation. Good luck finding a measure of that. On the other hand, it seems safe to assume that visiting a public library indicates access to a public library. In fact, using “visited a public library” as a measure of access is likely to grossly underestimate access — surely we can agree that more people have access to a public library than actually visits one, right? But let’s play it safe and go with actually visiting a public library as our measure of “access to a public library”.
Since I couldn’t find any articles or reports containing statistics about public library access or use, I decided to use the SDA: Survey Documentation and Analysis site from UC Berkeley to do some quick and dirty analysis of the General Social Survey data. A quick cross-tab tells us that in 2008 65% of respondents in households with household incomes under $25,000/year visited a public library at least once in the past year. In 2012, 64% of such respondents visited a public library at least once. I used the VISLIB variable (asked in 2008 & 2012 only) and the INCOME variable, recoding them both into dichotomous variables for simplicity’s sake. Please feel free to do your own analysis.
“What percentage of poor children has access to a decent public library collection?”
At least 65%.
“…as compared to the percentage that has access to the Internet?”
No more than 46%.
Online information is less accessible than printed information in a public library to the 54% of poor kids who do not have home internet access. Sure, those kids can access online information at their public library, but unlike the printed books, they can’t check that online information out and take it home with them. Any assertion that online information is radically more accessible than print ignores the reality that online information is radically inaccessible to over half of our poorest citizens. Note also that only 57% of African-American households and 58% of Hispanic households have home internet access (pg. 26, Exploring the Digital Divide), so ebooks and other online information is pretty inaccessible to over 40% of African-Americans & Hispanics too.
Another question Anderson poses is “If you really want to get a book into the hands of a maximum-possible number of poor kids, is encoding it into a physical document (rather than, say, making it available via cellphone) really the best way to do it?”
Considering that even among low-income individuals who own a cell phone, only 55% of them use their phones to go online (Cell Internet Use 2013, from Pew), making information available via cellphone may not get that info into more poor kids’ hands than having it available in a public library (where 65% of low-income kids will visit in a year).
Another bit of data from the Pew Internet report on e-book reading is suggestive of whether print books or ebooks are more accessible to poor kids. According to the Pew data, 19% of people with household incomes below $30,000/year read at least 1 ebook in 2012, compared to 44% at the highest income levels. E-book reading is growing across all income categories, but at a much slower rate among low-income individuals than among higher-income readers.
The difference in reading between income levels shrinks considerably when you look at reading on all platforms (print and electronic), with 66% of those in the lowest income bracket reading at least 1 book, compared to 84% in the highest income bracket. Moreover, there is little difference between the mean number of books read by low-income readers (14), and the mean number read by high-income readers (15). Actual behavior would seem to indicate that being in a low-income household has a stronger negative impact on ebook reading than on reading across all platforms.
I long for the day when technology really does fulfill the promise of enabling equal access to all kinds of information for all kinds of people. But the data I can find sure seem to indicate that we simply are not there yet, and I believe that “solutions” that ignore that reality are likely to increase rather than mitigate current inequalities.
Final note: I would love to be wrong about this. If anyone has different data and/or does a more careful analysis of the GSS data and proves me wrong, bring it.