In which countries do social norms spread more rapidly?

There is a new piece of interest in Technology Review, here is one excerpt:

Psychologists have always assumed that patterns of behavior change more quickly in countries that emphasize collectivism. Once an idea has taken hold, the pressure to conform means it spreads rapidly. “It has previously been argued that social support mechanisms in collectivistic societies make it more likely that a person will stop smoking,” say Lang and co.

And conversely, in countries that emphasize individualism, patterns of behavior must change more slowly because there is less social pressure to conform.

The puzzle is that the data on smoking shows exactly the reverse. Sweden was much slower to adopt smoking and much slower to stop.

Now Lang and co think they know why. They’ve created a mathematical model that includes the effects of social pressure allowing them to simulate the way behavior spreads through societies with different levels of individualism.

The model reveals why Sweden stopped smoking more slowly. “Our model suggests that … social inertia will inhibit decisions to stop smoking more strongly in collectivistic societies than in individualistic societies,” say Lang and co.

The original research, by Lang, Abrams, and De Sterck is here.  Their results do not rest on Sweden alone, but for the record I consider the Swedes to be relatively individualistic by most metrics, most of all when it comes to atomization.


Define atomization in this context?

The definitive book on this topic is Diffusion of Innovations by Everett M. Rogers, now in its fifth edition.

In general, diffusion is an S-shaped logarithmic curve that is of the form y'(x) = A y(x) - By(x)^2, which makes sense if you think about it, and can be solved exactly in closed form (see M. Braun for solution).

If I had to look up a solution for that equation by schoolteachers would turn in their graves.

A lot of good it did them.

Just want to quickly point out that the empirical basis for the theory in this paper is literally just a n=25 country correlation between individualism scores and peak year of smoking. Other than that, they just show that smoking prevalence follows a n shape in Sweden and the US which in no way provides any support to claims about the cause of the differences between countries.

There is no individual level data used in the paper (i.e. do more collective people pick up smoking faster than individualists?). Overall, a very weak evidence base for the claim that they have "revealed why Sweden stopped smoking more slowly".

To clarify there's more models that they present but it's still fundamentally an n=25 analysis. Their theory rests on an individual level mechanism but this is never tested.

As Cowen, I'd be extremely skeptical of any measurement that rates Sweden as relatively collectivist.

"Psychologists have always assumed that patterns of behavior change more quickly in countries that emphasize collectivism."

I would assume it could go either way. Wouldn't conformity bias slow change until a tipping point is reached, then speed adaption by the rest of the population.

This Norm guy got game?

I dunno about this one, because there are many things that may not be controlled or defined well, or are assumed too easily.

For example:

1 Look at the premise: "Conversely, the social freedoms in individualistic societies allow anybody to make a decision to stop smoking more easily." Have you ever tried to stop smoking. Do you think it is simply that social pressure makes you lose nicotine addiction, or reinforces it, or neither.

2. Does the study control for family effects: During WWII, US soldiers got free cigarettes, Swedish citizens were lucky if they got cabbage, much less cigarettes. When a soldier went home, was he more likely to smoke and later have his wife and adult children follow. What about the Swedish cohort. Why would they start. When they started, did husbands and wives start together, and grown children later.

3. Its nice to show a bell shaped curve, but tell me the age, and, more importantly, by sex, given how cigarette advertising and marketing to females changed over time, as did marketing by different media..

4. And, what about institutional restraints: Did Sweden prohibit television advertising like the US, or not. Is the US individualistic or collectivistic when it prohibits cigarette advertising and requires warnings. Could you be individualistic in one domain, and collectivistic in another domain, such as public health.

There are many straws you could grab to explain things, and I think grasping for collectivism v. individualism is a rather weak one to hold onto unless you control for an awful lot or prefer thinking of yourself as the Marlboro Man and the other person as the tribal follower.

In general, cultures conserve and reform different things, so that a cultures that seemed conservative on some measures (such as 19th Century Spain and politics) might put a lot of reformist energies into, say, rationalizing spelling, while Britain would be the reverse. This can lead to negative correlations.

Do you know anything about 19th century Spain or was this picked at random? After 1812 Spain was the second most unstable country in Europe other than France. Yea it managed to produce a lot of reactionaries like Donoso Cortes, but that's precisely because Spain was hardly conservative.

Another academic waste of perfectly good pixels that could have been devoted to dog and cat videos.

I guess this just seems sort of elementary. They initially thought that change spreads more quickly in collectivist societies, but the data showed the reverse, so now they think the reverse -- that change spreads less quickly in collectivist societies.

"...but for the record I consider the Swedes to be relatively individualistic by most metrics."

They're well within the "normal" range of individualism by the Viking metric.

In more individualistic cultures there are more open avenues for social distinction, to signal status or coolness. But like the general economy, entrepreneurs and innovators are a small minority. They set trends and set off
an arms race of follow ons. Once the fashion is widely adopted (mainstream) it loses it's distinctiveness and signal value, prompting search for new innovations. In a society with as much creativity and free expression as the US, this cycle is enhanced by highly developed cultural supply chains. America's wealth, and lowering transaction costs make the cycle even faster. Think about how quickly norms around pot and homosexuality have flipped. It's not shock that these are also now high status forms of social distinction that are popular at the highest value ends of the cultural supply chain.

I have not read the paper. But any study on smoking in Sweden has to take into account the fact that Swedish men use snuff (snus, tobaco inserter under the lipp) on a massive scale. Unlike smoking, snuff doesn't lower life expectancy.

Sweden has the lowest share of regular smokers in the OECD. Only 13% smoke, but 12% use snuff. Sweden is the only country where women are more likely to smoke than men. Women rarely use snuff, it's considered gruff.

Comments for this post are closed