Surveys and sociology
Now, I'm not a sociologist, but … oh, wait.
Damn.
I guess I am. Or I wasted a lot of time and money improving my job prospects by getting that M.A. in Chicago . Oh, wait.
Damn.
OK … let's start over.
Sociology begins with the observation that, when you ask a lot of people questions, it all eventually starts sounding the same. It then moves to the marvelously efficient step of skipping straight to the same answers without the bother of asking a lot of people questions. This is of particular importance to those who prefer number crunching and data mining to the comparatively tedious task of getting to know people. (Insert here the joke about those who are modeling data instead of dating models.) Especially when my absolutely free survey came to the same conclusion about today's burning issues that a significantly more expensive Ipsos-Reid poll came to: people want the government to spend more money on them.
The free survey mentioned above was a prelude to this week's interview with MPP Cheri DiNovo (NDP, Parkdale-High Park ). Being the helpful sort that I am, I thought perhaps I could provide my MPP with some helpful words from the vox populi. And the hoi polloi. And maybe even the pico de pollo.
The first such words were “Do I know you?” Only 64% of adults over 30 recognized photos of Cheri snipped from the campaign literature that had blanketed the neighborhood only days before. Among the under-30's, the percentage plummeted to 50%. Name recognition did include some minor variations, such as “Carla DiNovo” and “the NDP leader” and “Sheila Watson – no, wait, Cheryl DiNovi – oh, why did they both have to be blonde?” (How Cheri could be confused with NDP leader Howard Hampton or the entirely brunette Liberal candidate Sheila Watson was outside the scope of the survey. And, perhaps, of sociology itself.) When asked to finish the sentence “Cheri DiNovo is the MPP who …,” almost half of the respondents replied with blank looks or frank admissions of complete ignorance. 20% replied with simple descriptions like “… is blonde” or “was a United Church minister.” Fewer still identified her party affiliation, and only 12% chose to connect her to a particular policy stance (8% fer, 4% agin'). Makes you wonder how many hits www.cheridinovo.ca is getting….
Issues
The next comments were about the issues. Mostly because that was my next question to those whom I could chase down and survey without getting too winded to speak (see “Methodology” below): “We don't know who will be elected on October 10th, but whoever that is, what should their top 3 priorities be?” This sort of question makes sociologists salivate, because they can use actual math to weight the issues based on whether they are priority 1, 2, or 3. It also allows them to create broad categories into which they can shoehorn the actual results they receive. For example, the top-ranked issue, “Health Care,” includes responses ranging from “Health” to “Use some of those [budget] surpluses for healthcare!” Second-place finisher “Education” includes responses ranging from “The Education System” to “Not to mess up education more than they already have.” These two received as many points as all the other single issues combined.
The interesting surprise was that the third top scorer, behind health care and education, was personal advice to the incumbent. More than any other issue, people wanted their MPP to be careful about things like honesty, ethics, interaction with the public, representation of people rather than party, and following through on promises. And they wanted integrity with teeth, as one senior reported: “Do the right thing … if a politician is dishonest, put the bugger in jail!” (He went on to provide examples. This led to a lively off-the-record discussion of the issue of white-collar crime.)
Next came “Environment,” which includes responses ranging from “helping the community; you know, the environment” to “Green energy” to “ Preserve High Park .”
“Safety / Security” includes responses ranging from “Policing” to “Dealing with the shootings at Jane and Finch.” (I didn't have the heart to mention that Jane/Finch is in the York-West district, not Parkdale-High Park .)
“Taxes” were next. Just “Taxes.” They outranked the more colourful issue of “Urban Funding,” which included responses ranging from “Taxes for the city” to “Don't bankrupt Toronto , you morons.”
“N/A” includes responses ranging from people simply trailing off after one or two outbursts to “Oh, I don't know” to “I've had problems. With every government except federal. And the last trouble I had federally was 1974. And let me tell you…”
“Child Care” includes responses ranging from “I'm a mother, so I'd like to see more things for kids” to “Have you seen what they charge for day care? In Montreal it's like seven bucks a day – it's subsidized – while here in Toronto they want $70 dollars a day and even if you only want to have your child there
2-3 days a week you gotta pay for five days and you gotta provide your own juice and toys and such. They told us if we were earning under $80,000 a year we could probably qualify for a subsidy of some kind. I mean, what's the top one percent of Canadians making these days? $89,000? So they're the only ones who can afford day care? You gotta be in the top one percent of Canadian incomes to live in this neighborhood?”
Finally, and with a little prodding, people talked the arts. More on that next week.
Methodology, or What's Wrong With This Survey?
Sociologists love randomness. In fact, much of sociology's efficiency-making maneuvers depend on it. It's the soothing ointment for the burning guilt of thoughts like “But I didn't really ask everybody, did I?” because it assumes that the larger and more random the sample, the more likely that it effectively captures the bigger picture.
The trouble with surveys is that people are involved. People being the complex creatures they are, results are notoriously distorted by survey quantum, which is the fact that you cannot know people's position on an issue and the direction of a conversation at the same time.
Thus, in search of “what the people think,” you keep having to collapse the waveform into what people will say when someone tries to survey them. And your random survey of community opinions becomes limited to your stroll up Bloor toward Runnymede at 5:00 on a Friday afternoon. It is further limited by the exclusion of people carrying screaming children, leaky grocery bags, or hot pizza boxes (or some combination thereof). And people wearing iPods. And the obviously left-handed. And the largest category, namely people who don't want to talk to people doing surveys and can run faster than a 260-pound survey taker.
Then, once you finally buttonhole the victim who couldn't stonewall or outrun you, you discover the second effect of survey quantum, which is the fact that people cannot know what they think and what they will say to you at the same time. This is the leading cause of “I don't know” answers in surveys, as well as the answers people give and then immediately regret. Furthermore, it also results in people telling you what they think you want to hear … or what they want to be heard saying. It's truly amazing how many wave opinions become particle survey responses….
All of which tends to support Bill “The Blizzard” Hingest's argument to sociologist Mark Studdock in C.S. Lewis's brilliant That Hideous Strength: “There are no sciences like Sociology. … I happen to believe that you cannot study men; you can only get to know them, which is quite a different thing.”