Sunday, October 22, 2006
VOIP WREAKING HAVOC WITH NANP, MAKING POLLSTERS GO CRAZY
Back before I started working full-time, I was something of a poll watcher, especially during presidential election years. I don't have a lot of time to weed through polls and election research anymore, but I do try to keep up with election trends in general. While I'm exposed to a lot of news every day through work, I rarely have time to actually read entire articles or look much beyond headlines and summaries.
If you're interested in polls and surveys, there are a few websites out there that do all the heavy lifting for you. PollingReport.com features a vast number of polls on many different subjects and issues. There is some subscriber content, but a lot of what you see on the front page is accessible for free.
Electoral-Vote.com was set up prior to the 2004 election and watched the presidential race closely, updating state polls daily to give a snapshot of what polls predicted the Electoral College outcome would be. One of my self-proclaimed expertises in college was the history and workings of the Electoral College, so I found the site to be fascinating. It is currently watching the midterm election which may see a power shift in both the Senate and the House.
Today's prediction has the Senate tied at 50 Republicans and 50 Democrats -- meaning the Republicans would retain their majority since Vice President Dick Cheney would cast any tie-breaking votes. However, the site is counting both Rep. Bernie Sanders and Sen. Joseph Lieberman as Democrats. Both men are on track to win their elections (in Vermont and Connecticut, respectively), but both are also independents. They'll likely caucus with the Democrats, but a 50-50 split like Electoral-Vote.com predicts would actually be a 50-48-2 split.
Regardless of who wins control of the Senate, there will likely be two independents who win their elections.
And the prediction for the House is quite a bit more dire for the Republicans: 227 Democrats, 207 Republicans and -- currently -- one tie. Those results are based on either current polling or "best guesses" tied to the last election. For instance, the first district of Kansas has not been polled, even though Jerry Moran does have an opponent this time around -- John Doll. Since Moran has won every election since 1996 by huge margins (91% last election, though he was unopposed), it's reasonable to assume that he'll win again.
Polls are polls, to be certain, and Electoral-Vote.com is more of an information resource rather than a way to firmly predict election results. The guy who puts it together does tilt to the left politically, and in 2004 at least, he seemed to pick the polls that he used based on which one made his preferred candidate (Kerry) look better -- ie: taking Zogby over Mason-Dixon. It looks like this year he's averaging polls, which is a good thing, though still not incredibly accurate.
His final prediction for Bush-Kerry in 2004 was Kerry 311, Bush 227. Of course, polls were sort of discredited after that election (especially exit polls), with Bush winning 286 to 252. So take his overall predictions with a grain of salt. They'll change from day to day, and are interesting to watch if for no other reason than to see the immense fluctuations that polls can go through.
Today on Electoral-Vote.com, the webmaster posted an interesting write-up about the North American Numbering System, voice over IP (VoIP) and how new forms of communication are affecting polling results. I'll post it here, but for much more information about polling, visit his site at Electoral-Vote.com.
Today's sermon is about the evils of the NANP. For those readers who don't happen to be telecommunications engineers, some explanation is in order; the connection to polling and politics will become clear shortly. The United States, its territories, Canada, and 16 nations in the Carribean use an integrated system in which all telephone numbers are 10 digits in the form AAA-NXX-XXXX, in which AAA is the area code, NXX is the exchange, and XXXX is the subscriber line. It is called the NANP - North American Numbering Plan. Back before the AT&T monopoly was broken up in 1984, most states had at most two or three area codes.
To a first approximation, the telephone system consists of exchanges (buildings full of switching equipment), each identified by an area code and exchange. Thus 914-949-xxxx is an exchange in White Plains, NY, to which 10,000 nearby telephone are connected, numbered 914-949-0000 through 914-949-9999.
With the need for ever more telephone numbers for faxes, computers, credit card verification, etc., plus the introduction of cell phones and multiple providers, the situation has gotten out of hand. Area codes and exchanges have proliferated wildly. Nevertheless, one principle has been maintained: area codes don't cross state lines.
All these developments have consequences for (telephone) polling. Suppose a polling firm is commissioned to do a poll in the close Senate race in Missouri. They know they have to poll area codes 314, 417, 573, 636, 660, and 816, although not equally since they are not equally populated. Complicated, but still doable.
The introduction of Internet telephony (VoIP) services, such Skype and Vonage, wreaked havoc with this scheme. VoIP customers can usually choose any area code they want. For example, a man in Omaha might choose Florida area code 561 so his mother in Florida could call him as a free local call. It also means that a pollster randomly calling 561 numbers might get someone who doesn't live in Florida. Since most people still have area codes that correctly designate which state they live in, for Senate polls, the problem is still manageable.
However, for House polls the problem is substantial. The layout of the area codes and exchanges do not align with congressional districts at all. While 914-949-xxxx numbers all lie entirely within NY-18, other exchanges straddle congressional district boundaries, especially when the CD has been gerrymandered into a pretzel. As a consequence, a pollster assigned to poll for some House race may have to call multiple area codes and exchanges, some of whose numbers lie within the district and some of whose numbers lie outside the district. Reverse lookup of the number about to be called is not always possible because many people have unlisted numbers.
As a consequence, some of the people polled may, in fact, not live in the district in question and some people who do live there may be missed. Of course the first question could something like be "Are you a registered voter in congressional district IN-07?" However, most voters probably don't know their CD number and some may be put off by such a question and hang up. Starting with "Hi, I'm doing a poll from the XYZ company. What's your zipcode?" is definitely a nonstarter.
This issue came up last week with a poll of IN-07 in which Eric Dickerson (R) was slightly ahead of incumbent Julia Carson (D-IN). Many people suspect that the result was due to the pollster inadvertently calling people just outside this district, which is shaped like an immune-system cell, with little hooks sticking out all over its periphery, carefully avoiding Republican territory. There is no obvious solution to calling the wrong people, so House polls have to be taken with a grain of salt.- by Andrew Tanenbaum, Electoral-Vote.com