Search This Blog

Monday, July 01, 2013

Goring sacred cows

I like the E-Rate.  I think it's been very helpful in bringing broadband to the classroom.  I think further growth of the E-Rate would be a good thing.  [Can't you just feel the "but..." coming?]  But today, I want to complain about two stats that I have seen repeatedly over the last few weeks.

First, the chestnut that the success of the E-Rate is demonstrated by the percentage of schools that are online.  If you read anything by a politician supporting the E-Rate, you'll read that before the E-Rate, 14% of schools were online, and now 95% are online.  I won't quibble with the actual numbers; I don't know where they come from, but they're probably pretty close to right.  What I do quibble with is cause-and-effect.  Would the percentage be lower if the E-Rate hadn't existed?  Let's take a look at how Internet use has increased for those not benefiting from the E-Rate.

Here's a nice graph of Internet use among American adults. (You know I prefer to get my facts from pictures, not text.  And yet I almost never put graphs on this blog.  Just lazy, I guess.)  The use of Internet among adults has grown from around 20% in 1996 to about 80% today.

What about other organizations?  Less than 1% of components of federal agencies do not have Internet access (see the bottom of page 32 of this OPM report).  So the feds are better connected than schools.

Unfortunately, I couldn't find any data on municipal governments' Internet connections.

It's also hard to get stats on private companies, but it looks like in 1995, about 15% of companies had Internet access, but that climbed to over 60% by 1997.  I can't find any figures for 2012, perhaps because the assumption is that every company is connected.

Why don't the politicians mention libraries?  The Census Bureau says over 99% of public libraries offer public Internet access.  Surely that is a better stat; 99% beats 95%.  Oh wait, only 40% of libraries participate in the E-Rate. (Hang on, the Census Bureau says there are twice as many libraries as the ALA info I used, so maybe only 22% of libraries apply for E-Rate.)  So what percentage of libraries without E-Rate funding still offer public Internet access?  No matter how I slice it, I come up with at least 98% of libraries with no E-Rate funding still offering public Internet.  E-Rate can't take credit for that.

So I'm hard-pressed to show that E-Rate funding caused the increase of Internet access in schools.

I think the compelling message is not how many schools have Internet access, but the speeds that those buildings connect at.  So I'm happy to see that the new proposals focus on connection speeds.  Now if we could fix the Item 21 Attachments and stop hiding them, we would have a very complete picture of the bandwidth available to U.S. schools and libraries.

Without the E-Rate, I think close to 95% of schools would still be connected, but that connection speeds would be lower.  Alas, that's harder to prove, and more difficult to cram into a simple sentence, so we'll keep hearing "14% to 95%."


Second, several sources have said that 80% of  schools don't have sufficient broadband connections.  I have 2 problems with this.  First, the stat is based on a survey, and it's in our nature to want just a little more than we have.  Ask any American, "Do you have enough _____?" with anything in the blank (except taxes), and most people will say, "No."  How much is enough?  More than I have now.  If we took The DeLorean back to 1996, snagged a school employee and brought her to her school in 2013, and asked her if the bandwidth was good enough, we'd probably have to sit through her saying "Wow!" half a dozen times before she collected herself enough to say, "Oh, yes!"  If we asked the 2013 version of that employee the same question, chances are she'd say, "No."

I also doubt that most schools know how they are using bandwidth now and where the bottlenecks are.  I used to think Verizon's data network was causing my phone to load web pages so slowly, but then I connected it to a WLAN, and discovered the problem was the device, not the bandwidth.  In my experience, school network admins are getting much better at monitoring, and have a better handle on where the bottlenecks are, but still, most of the admins I talk to are too busy putting out fires to really determine why Ms. Smith's computer loads web pages so slowly.


I know we have to come up with something to support our drive to increase the size of the E-Rate, but I wish we had something better than these two factoids.

For those of you who were worried, I got a new phone, and I'm very happy with the speed.  I'm sure that's a result of a faster processor, not a faster network.  And I'm sure that in a year or two, I'll find it unbearably slow.

1 comment:

  1. Two updates:

    First, I was just reading an ALA notice about what a big E-Rate week we have coming up (http://www.districtdispatch.org/2013/07/senate-e-rate-hearing-to-feature-maine-state-librarian/), and it had the following factoid:
    "70 percent of libraries now offer internet connections speeds greater than 1.5Mbps...."

    Why is the ALA's number so much lower than the Census Bureau (http://www.census.gov/compendia/statab/2012/tables/12s1155.pdf), which says that 88% of libraries offer public Internet at speeds of at least 1.5 Mbps.

    Second, two months after I got my new phone, I now find the Web browsing speed too slow. 22 months from now, when I qualify for another discount on my phone, I'll be tearing my hair out at how slow this phone is.

    I am positing the Riordan Corollary to Nielsen's Law (http://www.nngroup.com/articles/law-of-bandwidth/). Every increase in bandwidth will keep the user satisfied for 2 months, after which it will seem too slow.

    ReplyDelete