Lies, damned lies, and statistics


A philosophy professor of mine had this awe for the printed word. “When I read an opinion or a report in print,” he said, “I tend to give it more credence than if I heard the same view expressed verbally.”

This admission came from one of the sharpest minds I’ve personally known – a Jesuit, who taught logic for 30 years, and whose idea of a relaxing evening was reclining in his easy chair, a glass of vintage port at hand, leafing through a volume of Kant or Kierkegaard.

The instinctive reaction of my Jesuit philosophy prof to printed material – I suspect – is no different from the way many of us respond to surveys and statistics.

Chances are we’re more than likely to question or challenge a contentious opinion voiced by an individual – whether verbally, in an online blog, or in print.

But the same statement reported as a survey “finding” – and “backed up” with a lot of charts, graphs, tables and numbers – carries an aura of validity.

Take the recent survey “finding” by Statistics Canada – based on a 2005 general social survey – that “heavy” Internet users spend less time in social face-to-face contact with other people.

Some popular media, including my favourite radio station, reported StatsCan’s assertions as facts.

My first reaction too, was to take those “findings” at face value. After all this is StatsCan – the paragon of respectability and its report is based upon a scientific countrywide survey.

It’s only on further reflection that some of the cracks became apparent. And the problem wasn’t so much in the numbers; it was in what we were clearly led to infer from them.

For the purposes of its study, StatsCan defined heavy users as “those who [spent] more than an hour on the Internet during the day.” These people, we were told, devote less time to socializing with their spouse or partner, as well as their children and friends.

The report clarified that the study considered only personal use of the Internet over a 24-hour period, and not use of the Web for other reasons, such as work or school.

Be that as it may, are we to assume that people who spend an hour or so of “personal” time on the Web, or seven hours or more a week (the so-called heavy users) are less social than those who spend the same or more time on other “passive” hobbies – like watching TV, for instance?

On that note, it’s interesting to compare the “Internet impact” study with another StatsCan report on the TV viewing habits of Canadians released this March. According to that survey Canadian men between the ages 18 -24 watch the “least” TV. And “least” is defined as 12.3 hours a week.

But wait a cotton pickin’ minute – that’s more time than the so-called heavy Internet users spend on the Web. And I don’t remember there being any suggestion in the TV study that staring at the boob tube for so many hours a week is hazardous to one’s social health.

Two other key “findings” of the StatsCan “Internet impact” survey are:

• Heavy Web users devote “significantly less time” than non-users to chores around the home;

• They spend less time relaxing, resting or thinking;

Again these findings raise more questions than they answer.

Take the issue of heavy Net users not making themselves useful at home. The implication seems to be that it’s their Web surfing habits that prevent them from pitching in.

But is that really so? The survey also says heavy Internet users are nearly eight years younger on average than non-users, many of them are students, and six out of every 10 heavy users are men. Put that all together and you learn that a great many of those surfing the Web in Canada are young male students.

Is it very likely that many of these hot blooded young bucks – if they were not cyber surfing or connecting with their peers online – would get a sudden urge to pick up that vacuum cleaner, or wash the dirty dishes. Me thinks not.

Or consider the point about heavy Net users spending less time sleeping, relaxing and thinking!

Is it that inconceivable that many heavy users do much of their thinking when they are online…that many of them find Web surfing intellectually stimulating? Is it beyond belief that surfing for some can be as relaxing as yoga? In fact, the StatsCan study itself supports this last view. Heavy users, it says, reported being “less stressed.” (How on earth do they manage it – deprived of sleep and relaxation?!)

Another finding: the unemployed constitute a higher proportion of heavy users. But that begs the question: are they unemployed because of their Web surfing habits, or is it the other way around? Perhaps they just don’t want to work, in which case where they fritter away the hours – whether Web surfing, watching TV, or hanging out with their other jobless pals – is hardly material. (Of course, there’s also the possibility that the Canada’s umemployed may be online more often – because that’s where they hope to find a job!)

The long and the short of it is that mulling over this most recent StatsCan study got me thinking about how alluringly deceptive surveys can be.

My parting shot about stats, studies and the whole shebang echoes what some bloke said about love at first sight: “You’ve got to take a second look.”

And when you do, don’t be surprised if you find yourself cussing under your breath and saying words Benjamin Disraeli uttered many years ago “lies, damned lies and statistics.”