Part I
The expected
All the questions youâd expect to ask, but not in the way youâd expect to ask them
As the saying goes, even the worst piece of communication only needs two things changed: all the words, and all the images. As trite as that sounds, there is some truth in it.
Weâll explore how words like tax, offset, climate change, and global warming can evoke responses in ways you wouldnât expect, and how imagery â either synthetic or constructed â affects our decisions in covert ways. Are these principles at the root of variously attributed lament: âHalf of all advertising works, I just wish I knew which halfâ?
Weâll look at how a tiny hammer in a box that could kill a cat is on a par with asking a question. How the strength of your desire is nothing to do with what you say and everything to do with the fact that you said it â and that can kill you. And why your highfalutin âflights of fancyâ are triggered by the âwhyâ: LĂ©vi-Strauss is involved.
Later in this section issues of timing and incentive keep us on familiar ground, but are expressed in unfamiliar ways. (Youâll want to keep the snack food out of reach if youâre on a health kick when you read about bunched messaging and cognitive load; the reasons will become clear.) Incentives and incentivising behaviour focus on cash and fun. Cash is an obvious and lazy place to go (the BOGOF is as much a bane of the marketerâs life as is taxation for a legislator), but cash incentives for resource-reducing behaviour can be inexpensive if the correct question is asked: itâs a cheap thrill.
As expected, welcome to the Expected.
1 Just ask
âIf you want anything just ask for it, old sport.â
F. Scott Fitzgerald, The Great Gatsby (1925)
In 1935, in an attempt to illustrate the bizarreness of quantum mechanics, Erwin Schrödinger devised a thought experiment to show that the act of observation can change the observed. He imagined a cat sealed in a box with a phial of poison, a radioactive particle, a Geiger counter, and a tiny hammer. The radioactive particle roamed the box. If it was detected by the Geiger counter the tiny hammer would break the phial of poison, which would kill the cat.
Thatâs not the bizarre bit.
Shortly after the experiment had begun, the cat would exist as both potentially alive and potentially dead: thatâs the bizarre bit. This simultaneous state would resolve only when the lid was opened and the cat observed. Itâs only this act of observation that resolves the state of the cat.1
The lesson Schrödinger sought to highlight is thereâs no way to engage with the world without affecting it.
Heâs right.
Schrödingerâs car
It was long thought that strength of desire was a good measure of the likelihood of action â and to find the strength of desire, you just ask. Say Fishbein and Ajzen: âIf one wants to know whether or not an individual will perform a given behaviour, the simplest and probably most efficient thing one can do is to ask the individual whether he [or she] intends to perform that behaviour.â2
Vicki Morwitz from the Stern School, New York, and Gavan Fitzsimons of the Wharton School, Pennsylvania, discovered something quite different when they asked 40,000 people about the strength of their desire to purchase a new car: Simply asking about the strength of desire disproportionately increased action.3
That wasnât supposed to happen.
The actual rates of car purchase among the group in the following six months shot up to over 35% above average.
Simply by measuring the state, they changed it.
This effectâs been dubbed the Mere Measurement effect (even though thereâs nothing mere about its effect). It tells us that our response to an initial-intent question changes our subsequent evaluations by activating an existing attitude â and that attitude remains accessible and pronounced for some time. However, if there were no existing inclination one would not appear. Much like a countryside path â if you have a groove, youâll get a rut. So of those 40,000 with existing inclination to purchase, their attitude was amplified. This amplification of intent is not solely about purchase: Greenwald, Carnot, Beach, and Young saw the probability of voting increase by 25% for people who had been asked about their voting intention for the following day.4
These are big numbers.
And they have been observed elsewhere, affecting computer purchases, volunteering, exercising, and recycling â and even affecting harmful behaviours like food choice5 and drug taking.6 It may even be responsible for killing people: well-meaning government and charity drug support programmes often question heavy and serious drug addicts about their previous patterns of drug use in order to allocate future resources efficiently, thereby invoking the Mere Measurement effect and so increasing their drug-use rates.
The workings from this data are contested.7 However, in a similar study of New Jersey teenagers on a suicide prevention programme, those that were told about the worryingly large number of teens who take their own lives became more likely to see suicide as an answer to their own problems.8 Similarly, an eating disorder help-group for young women involved many women describing their harmful eating. Participants had more disorder symptoms after attending the groups.9 (See also chapter 12, âAsk using the right authorityâ.)
Death and disorder are extreme examples, but the problematic effect of the Mere Measurement effect doesnât end there â it questions the foundations of market research: âThe American Marketing Association (AMA) code of ethics clearly segregates the conduct of marketing research from any form of sales or opinion-influencing activity.â10 I think we can safely say that âopinion-influencing activityâ is the definition of the Mere Measurement effect.
Oops.
And it gets even more fascinating. After 40,000 people were exposed to the survey question in the Fitzsimons and Morwitz study, it became apparent that car owners increase their likelihood of purchasing from their current manufacturer, while those who donât own a car increase their likelihood of purchasing from any one of the frequently advertised brands.11
Frequently. Advertised. Brands.
By definition, a manufacturer of an electric car has a product thatâs both new to the market and a new category for many consumers. Theyâd do great business simply by advertising. Itâs almost irrelevant what they say, only that the brand is exposed. Or maybe they should do a simple survey?
The Mere Measurement effect was evoked from a âsimpleâ question that puts the receiver as the main actor in a scenario. Imaging other people in scenarios is not as effective in changing subsequent actions â this is why âyour new carâ is so powerful. But what happens if we ask a more nuanced or complex question?
Nuanced questions change outcomes too (as if you hadnât guessed), but it depends on how the question is structured: imagining positive, negative, and avoidance outcomes affects outcomes differently.12 Understanding this is vital from a sustainable point of view.
A positive statement more readily accesses an existing attitude than a negative statement, so your âmereâ question will get more of a response.13 âPositiveâ usually translates as aspirational â like ownership of a new car (âAre you going to buy a new car?â). But positive aspiration is a little tricky to elicit for products or services that require an obvious side- or down-shift â the products and services that are seen as negative.
A positive or neutral statement seems to work best when asked âwith the grainâ to keep mental calculations and translations uncomplicated. For instance, donât ask how many times do you put the bins out per month? if the âwith the grainâ answer is once every fortnight. Donât ask how many times per month do you have a haircut? when the âwith the grainâ question is how many weeks do you wait before you next visit? In short, donât put any unnecessary calculations in there.14
A negative statement, like can you see yourself not eating sugary foods?, seems to have two characteristics: our âimagined stateâ is poorly constructed because it requires a greater load on working memory than imagining positive or alternative scenes; that which we manage to construct in our âimagined stateâ frequently goes AWOL. Information about ânot doingâ something seems to be treated as casually as a trivial Post-It note.15 Behaviour is rarely changed. This is a perennial problem for the Sustainaratti. Indeed it is a perennial problem for the overall challenge of mitigating the effects of climate change: the upstream cost versus the downstream benefit often requires the evocation of a negative feeling now for the promise of a positive one in the future (and of course that benefit is for someone else, in some other place, at some point in the future, in an unclear way, which makes it a whole heap of awkward â see chapter 7, âAsk for a commitment (in the future)â).
An âavoidanceâ construction can help in this case: âWhat is the likelihood youâll avoid X?â It seems much easier for us to imagine ourselves performing the necessary actions to avoid something rather than the negative construct of not doing something.
But, however constructed, we tend to want the shortest route, so we ignore the negative âimagined stateâ and grab for positive, or, failing that, avoidance. In this sense our imagination is like a river trying to find the shortest path to lower ground, where the shortest path is a positive construct. The only exception is if the âshortest pathâ is a negative statement that chimes with an existing negative attitude (a socially undesirable behaviour â like murder), because it acts much like a positive statement in the sense that ânot doing itâ is good in a socially acceptable way.16 A pseudo-positive, perhaps?
Both avoidance and positive questions are more palatable and memorable. They do more than prime behaviour â they create an implementation imprint that endures after the question that affects our evaluation and subsequent behaviour.17
______________
âSimply askingâ is one of the most inexpensive ways of communicating. There really isnât a great deal of infrastructure involved â only clipboards or web pages. But by being both persistent and near costless to enact, it seems the Mere Measurement effect is anything but âmereâ.
You werenât aware? All you had to do was ask.
A slice of life
Marlene Schwartz is the Deputy Director of the Rudd Center for Food Policy and Obesity at Yale University. She took on a problem that the United Statesâ National School Lunch Program (sic) had been trying to solve since 1947: to ensure that American schoolchildren have access to a nutritionally balanced and affordable lunch.18
For 60 years theyâve been trying. And failed.
American schoolchildren were malnourished in 1947 because of a lack of calories. In 2007 â sixty years later â American schoolchildren were still malnourished despite having access to vast quantities of calories. Why? Because the consumption of nourishing foods like fruits and vegetables was considerably lower than needed. A school districtâs Health Advisory Committee had a hunch how they could increase the nutritional intake of pupils.
They thought they could solve this problem simply by asking a question.
They drafted in Marlene Schwartz to run the test. Two schools were selected, both with the same fruit and fruit juice options on the school dinner menu. She asked one school to have the dinner ladies say nothing, and the other school have the dinner ladies ask every pupil âWould you like fruit or fruit juice with your lunch?â A simple, innocuous que...