Posted 1:06 PM by Robert

Asking The Wrong Question, Again?

Blogging has been light recently, mostly because the Man Without Qualities is in the process of buying an additional house not far from the current abode, which abode is also subject to ongoing construction. All of which takes up a lot of time and personal energy.

But, in addition, I have to some extent been waiting for additional wisdom from the Mystery Pollster about cell phone usage and polling - especially regarding "five factor households." Perhaps I'm slow on the uptake, but I don't think discussing - as the Mystery Pollster has done - whether households with land lines and cell phone often or seldom or almost never answer their land line sheds much light. Of course many people with land lines and cell phones often answer their land lines - but they often only answer calls to their land lines if the caller ID on that land line identifies the caller as known and desirable. Indeed, just one minute ago I declined to answer a call on the land line not two feet from this key pad because my call ID revealed the caller to be "Out of Area," but 10 minutes ago I accepted a call on the same line from my mortgage broker. Such people - including the Man Without Qualities - are functionally equivalent to "cellular-only" households as far as pollsters are concerned. No matter how many times the pollster calls that land line, the target will not answer. I therefore see no good reason to fold the "five factor households" problem into the general issue of poll "response rate" while breaking out "cellular only" households for special attention.

With respect to "cellular-only" households, The Mystery Pollster suggests (quite reasonably, in my view):

My point was that the available demographic estimates of the mobile-only population suggest a Democratic orientation, but the same demographic pattern also suggests they are historically low turnout voters. Even if we assume comparable turnout and a plausible pro-Kerry margin, their relatively small size still implies a very small effect. Of course, this inference is a matter of opinion. Yours may differ.

The analogous question for "five-factor households" is then: Is there anything that suggests that five-factor households have a partisan orientation?

There is at least a plausible (but by no means definitive) argument that "five factor households" tend to be upper income, and therefore possibly Republican-leaning in most areas (places such as West Los Angeles and the Upper West Side of New York would be exceptions). Of course, a "five factor household" needs enough money to afford both cell phone and land line service - so that already give the group a bit of a nudge up the economic scale. Perhaps more significantly, prior to the dawn of the national "do not call" registry, many people screened their land line calls with caller ID to avoid telemarketers. Such households were probably disproportionately relatively upper-income - because that's the demographic telemarketers mostly target. For example, it seems reasonable that more telemarketing calls went into prosperous La Canada-Flintridge on a per capita basis than into more modest Pico Rivera, located a few miles away. The establishment of the "do-not-call" registry means that a good reason for wanting (or needing) to screen one's calls was reduced. But that registry only went into effect recently, after many households had already established their screening procedures - procedures that they continue to maintain. Moreover, charitable (I do answer when the LA Opera and Philharmonic calls) and political organizations (including political pollsters and, much more often and annoyingly, consumer preference survey takers who seem to call only when my 5-year-old is near the phone to answer it!) and commercial outfits that have somehow obtained permission to call, still call in unacceptable numbers and at bad times - thereby prompting the desire to screen incoming calls with caller ID. A lot of these residual callers (including the consumer preference surveyors, but not the political pollsters) seem, not irrationally, to target upper-income households - although, again, this is anecdotal.

I, personally, know perhaps one true "cellular only" household with American citizenship (there appear to be more of them among the large number of illegal immigrants I know who buy prepaid service - are these included in Arianna's Democratic rescue-squad?). But I know dozens of "five factor households" who have citizenship and will vote.

How will they vote? Who knows? I haven't asked them all, but of those I have asked, I haven't found a single one who has answered a call from a political pollster this year.

Of course, that's not science.