« How worried should we be about the deficit? | Main | Assorted links »
The "paradox of choice" is not robust
I missed this one while traveling, so I am grateful to the loyal MR reader who pointed it out to me:
... the psychological effect may not actually exist at all. It is hard to find much evidence that retailers are ferociously simplifying their offerings in an effort to boost sales. Starbucks boasts about its “87,000 drink combinations”; supermarkets are packed with options. This suggests that “choice demotivates” is not a universal human truth, but an effect that emerges under special circumstances.
Benjamin Scheibehenne, a psychologist at the University of Basel, was thinking along these lines when he decided (with Peter Todd and, later, Rainer Greifeneder) to design a range of experiments to figure out when choice demotivates, and when it does not.
But a curious thing happened almost immediately. They began by trying to replicate some classic experiments – such as the jam study, and a similar one with luxury chocolates. They couldn’t find any sign of the “choice is bad” effect. Neither the original Lepper-Iyengar experiments nor the new study appears to be at fault: the results are just different and we don’t know why.
After designing 10 different experiments in which participants were asked to make a choice, and finding very little evidence that variety caused any problems, Scheibehenne and his colleagues tried to assemble all the studies, published and unpublished, of the effect.
The average of all these studies suggests that offering lots of extra choices seems to make no important difference either way. There seem to be circumstances where choice is counterproductive but, despite looking hard for them, we don’t yet know much about what they are. Overall, says Scheibehenne: “If you did one of these studies tomorrow, the most probable result would be no effect.”
That's by Tim Harford. In my view, the so-called paradox of choice is one of the most overrated and incorrectly cited results in the social sciences. The full account is here.
Posted by Tyler Cowen on November 24, 2009 at 10:15 AM in Economics, Education | Permalink
Comments
It's possible that people have gotten used to more choice in the past decade.
Posted by: Nancy Lebovitz at Nov 24, 2009 10:32:50 AM
Fascinating. I wonder if the jam experiment was affected by the 'posh supermarket' context. In some contexts a reduced choice is an expression of quality, it is associated with artisanal virtues, especially when it comes to luxury products. If the supermarket was posh enough, the passing trade might have approached the stall with that mindset. I can well imagine that effect with, say, wine. If a couple of varieties were offered by a new winery it would imply quality and I might be tempted to try if it tasted OK, but if they had a range of 24 it would look like another big brand mass-produced drink and I would be more sceptical, preferring to stick with the ones I already know and love.
Posted by: John Meredith at Nov 24, 2009 10:44:52 AM
This is the comment I made to Tim Harford. I like John Meredith's more sophisticated take above.
"This may be new in the literature, but it is ancient in practice. The key is that in almost all purchase situations, the time and attention which the buyer wants to devote to her, his or their decision is limited. Therefore most routine purchases are repeat purchases; and non-routine purchasing follows a standard pattern. In non-repeat purchases buyers usually want to first quickly scan a large range of choices. A wide range of choice therefore attracts more prospects (one reason for the success of hyper-markets). Buyers then zero in on a narrow range which they think worth the investment of time in serious consideration. Most of the time the narrow range is three to six options wide. This allows the buyer a manageable number of detailed comparisons of different aspects. (The number of comparisons rises sharply,of course, with each extra option added.) For a familiar example, think of how you buy a car.
If they cannot zero in to a narrow range, most buyers will put off the non-routine purchase if they can. The cost in scarce time of making the purchase decision becomes prohibitive.
I recall, when employed as an economist in market research, explaining this, lamely, to consumer marketing people 50 years ago. It also applies, with emphasis, when the buyers' time has an explicit price tag; as in corporate piurchasing (including hiring).
When this model does not apply is when the buyer considers her or his time spent on wondering what to buy costs very little (e.g., for the world's poorest, it is often worth spending a lot of time working out how to spend their little money), or indeed has a positive value (think of how many women and some men enjoy shopping for clothes). I am not familiar with a wide range of the experimental frameworks for the experiments quoted, but at least some experiments in this field ask that people set aside time for the experiment. If they do so, the buyers concerned will not regard time within this little segment of their lives as scarce. Therefore they will dally over the purchase, and look at a wide range of possibilities."
Posted by: David Heigham at Nov 24, 2009 10:55:17 AM
Then why can't I choose between the 10 gyms in town!
Posted by: john personna at Nov 24, 2009 11:05:17 AM
Isn't 'too much choice' a big deal for the left-ish book writers like Robert Frank and others? And maybe even for the libertarian paternalists?
If I remember right, a big part of their critiques of today's weak capitalism is based on the that people can't process well when they have many choices, and that it is better for choices to be limited.
More generally, it would be great to have a list of the top 50 consumer-oriented economics books and a list of their priors so that we could watch as they gain strength or are undermined.
Posted by: tom at Nov 24, 2009 11:11:40 AM
I'm reminded of two effects.
First, I distinctly remember the reactions of my friends from Communist and former Communist countries when they enter an American supermarket for the first time. Paralysis. If they were left to their own devices, they would have to spend hours looking over the value proposition of one brand vs. another, try to determine just how much value "premium" represents (it must be something, since they bothered to so mark the package) and a whole plethora of other such bits of confusion. In this case, choice has a very real cost.
Americans can experience this same choice paralysis in the wine store (I call it the "wall of wine" effect). There are a number of odd effects at play here (I used to work in the wine industry and saw them first-hand). At the value wine level, you see a few brands, few flavors, and price becomes the fast criterion. At the premium wine level, though, you've got thousands of little brands, and the consumer wants this. In premium wine, the consumer wants to find a great-tasting wine that his or her friends have never heard of or know how to find. The correct answer is there, in front of you, but you only have wine labels to guide you. Unless you're a serious aficionado, price becomes the big selling point, and it's still hit-or-miss at that.
Eventually, one learns to simply ignore most of the information broadcast at you--you untie the Gordian knot of advertising by slicing through it. Once you do that, choice becomes a wonderfully liberating thing, allowing you to learn, experiment, and come up with new, innovative solutions that are powered by choice. Until then, however, choice breeds paralysis, confusion, and bad wine for dinner.
Posted by: IVV at Nov 24, 2009 11:17:13 AM
the paradox of choice is not alone. Keith Chen has a forthcoming paper that finds that most papers on cognitive dissonance all share a statistical flaw. List and Levitt reanalyze the data from the Hawthorne effect, and find that the Hawthorne effect doesn't exist in the Hawthorne data. They also have some data that finds little evidence for Claude Steele's stereotype threat. I'm wondering how many of these old oft cited studies are indeed robust, especially when even cognitive dissonance seems not to hold up.
Posted by: Ben Ho at Nov 24, 2009 11:23:14 AM
Does it not all begin with the (1) the nature of the motivation - e.g., need or special gratification and (2) the factors initiating the motivation - e.g., impulse vs objective-directed search?
Posted by: R. Richard Schweitzer at Nov 24, 2009 11:23:15 AM
If the extra choices allow me to find exactly what I want, without too much compromise, then the effect of choice will be to motivate me to buy.
If none of the extra choices give me exactly what I want, but instead force me to spend lots of time trying to decide between nearly-equal unpleasant trade-offs, with the end-result that I am making the same compromises that I would have had to make anyway if there were fewer choices, then choice demotivates me.
Posted by: Anon at Nov 24, 2009 11:27:20 AM
Freedom of choice is OK, but often it is even better if it I get freedom + some recommendations, cf. Amazon.com.
Posted by: me at Nov 24, 2009 11:34:03 AM
I think it matters how the choice is presented. Is there a simple-to-navigate menu with the options listed? Or, are certain choices bundled together so that we can't easily get exactly what we want or we can't find it because navigating the menu is so difficult?
Posted by: Craig at Nov 24, 2009 12:06:32 PM
It seems to me that it's possible that people have changed. Not sure how old the original studies were, but I find that people now are much more accustomed to having and exercising choice.
My dad used to have trouble figuring out the menu at fast food restaurants, because he was asked to make so many choices. Since then particularly fast food places are now all about customized choices, and even I when going to a new style fast food place can be bewildered by the array of choice.
Posted by: Justin at Nov 24, 2009 12:28:51 PM
Let's say you are sailing and you come to a harbor. You are faced with a choice between Dock A and Dock B. They are identical in appearance. How do you choose? It's a paradox.
Posted by: Andrew at Nov 24, 2009 12:30:57 PM
From self-observation, I think there's a LOT to the paradox of choice. To give you two examples of my consumer behavior...
My switch from PC to Mac might have had something to do with too much choice. Buying a new PC was always vexing. What brand? What processor speed? When I looked at the Mac site, the decision was far simpler because of the reduced choices, and thus posed less anxiety.
Or choosing a toothpaste. I had always chosen Colgate simply because that was the brand I grew up with, but faced with a gazillion choices under the Colgate brand I finally said to hell with it and bought another brand. My brand loyalty was shaken loose by product fragmentation. (Sure, these are just anecdotes, but self-observation has to be where we evaluate hypotheses.)
My feeling is that anxiety of choice will be greater when the decision is important to us, when we feel more information is available if we sought it, and when we feel our competence at making the choice is inadequate. These factors might not have been replicated adequately in the experiments.
Posted by: Wagster at Nov 24, 2009 12:37:29 PM
You are on to something Wagster. But what if you were denied the choice to buy the Mac? That's the paradox!
Posted by: Andrew at Nov 24, 2009 1:00:33 PM
On the other hand, I could say "I bought couscous from one brand three weeks ago, then from another brand two weeks ago, and I wasn't particularly anxious about the fact that there were multiple brands with different flavours etc. So evidently there is NOTHING to the paradox of choice." - but I don't because I will decide based on the experiences of people on the whole.
Posted by: Millian at Nov 24, 2009 1:00:38 PM
It has always appeared to me that the believers in "more options are bad" have been exactly those who want to limit the options available to others.
Posted by: Yancey Ward at Nov 24, 2009 1:01:00 PM
The "paradox of choice" always seemed like nonsense to me - one of those theories that, to me, couldn't pass the laugh test.
Posted by: Bob Montgomery at Nov 24, 2009 1:04:18 PM
I'm irritated when my choices are limited. But then again, that's just me. I know others who would rather just have someone else limit their choices in some arbitrary manner.
Posted by: Dustin at Nov 24, 2009 1:26:29 PM
There are similar studies being conducted at other universities involving the same thing but in a different format: product comparisons where there are a list of features. What happens is that if, say, you have three product, and under each you list a column of features: people fixate on the top of the list for making a comparison to the other two without looking further down on the list, and then pick that candidate and compare it to one more product. It seems we can only do comparisons between two products, and we cannot do comparisons far down on a list, choosing a shortcut of just looking at the top of the list.
So, product features may not be as important as how they are displayed, and creating a complexity of comparison leads to inevitable shortcuts in analysis.
Posted by: Bill at Nov 24, 2009 1:34:33 PM
Costco has done very well for itself offering its members/customers limited choices.
Posted by: Steve Sailer at Nov 24, 2009 2:48:07 PM
Elaborating on my take on what Richard Schweitzer says above -- it seems to me it may make a lot of difference whether one is actively seeking jam or not. If I need jam, then I probably appreciate choices. If I do not need it, then you might get me to make a special purchase if you have a sample of something I fancy, but a large number of samples requires me to spend too much time on a purchase I don't have any particular need or desire for.
Just as a for instance, I simply cannot imagine going to an ice cream shop and leaving empty-handed because they had too many flavors to choose from. I might get exasperated at the difficulty of chosing, but if so, I will simply pick a flavor I suspect I will like at random...
Posted by: Sol at Nov 24, 2009 2:53:32 PM
Next you'll tell me the human caused global warming data isn't to be trusted .... oh wait ..
Posted by: Greg Ransom at Nov 24, 2009 3:11:35 PM
But a more fundamental objection to the “choice is bad” thesis is that the psychological effect may not actually exist at all. It is hard to find much evidence that retailers are ferociously simplifying their offerings in an effort to boost sales. Starbucks boasts about its “87,000 drink combinations”; supermarkets are packed with options. This suggests that “choice demotivates” is not a universal human truth, but an effect that emerges under special circumstances.The German based discount supermarket chains ALDI and LIDL are expanding globally with the no choice model. They offer different flavours of jam but only one brand. These companies don't expand as rapidly as they could though, because the octogenarian founders don't believe in debt.
No frills cell phone plans are increasingly popular in Europe.
What is meant by choice anyway? There can be choice between alternatives (flavours), additives and sizes. These can be perfectly orthogonal like in a Starbucks. And the objective function is our intuitive sense of taste. So those 87000 drink combinations don't tell us much.
Cell phone plans, pension plans, cell phones, PCs, all of these are packages of various components which result in all kinds of complicated trade-offs and there are multiple objective functions to maximize. Plus the choice results in a commitment for some time.
Apathy is just one aspect of the paradox of choice. The other is decreased satisfaction with your choice.
Posted by: dieter at Nov 24, 2009 3:40:55 PM
Trader Joe's is another store that has done well by simplifying choice.
Posted by: john personna at Nov 24, 2009 3:42:30 PM
The problem isn't being parsed enough and results will not be instructive until the question is asked properly.
Like a doctor said about when there are numerous choices for a procedure means there is no cure.
Posted by: Andrew at Nov 24, 2009 4:23:21 PM
I think it comes down to a couple things things. First of all, there are some things that we want a lot of choices in, things we are passionate about, and some things we don't really care about, where we might be annoyed or confused by all the choices. I like coffee and soda, so I like having a lot of varieties to choose from. I don't really care much about toothpaste or paper towels, so I grab whatever is cheapest or on sale. But somewhere, there is probably someone thrilled that they can buy Colgate Plus Ultra Peppermint Gel with flavor crystals, no whip, and chances are the benefit they get from it outweighs my annoyance in the toothpaste isle.
The other thing is that we have ways of filtering large choices, like the Starbucks one. If I know I want a hot beverage, that knocks out half the options. If I know I want a Venti, that kills another 66%. If I know I want coffee and not tea or chai or hot chocolate, and that I want skim milk and not whole or soy or 2%, that takes a bunch more out, and in the end I'm really only choosing between a handful of options.
Posted by: MadAnthony at Nov 24, 2009 4:35:15 PM
I think the explanation boils down to what you care about.
I love chocolate and would never think there can be too many varieties to choose from. In fact I actively seek new chocolate shops to see if they have varieties I haven't experienced yet.
On the other hand I couldn't care less about shampoo and the incomprehensibly vast array at my local Walgreen's terrifies me, so I end up leaving and getting whatever one brand is on special at Costco, preferably in bulk so I only have to shop for it once or twice a year.
Economists would talk about cost-benefit trade-off between mental transaction costs and the marginal difference between the optimal product and the average product in a category you are not emotionally invested in.
Posted by: Fazal Majid at Nov 24, 2009 4:42:44 PM
I'm shocked. Someone called up Barry Schwartz and tell him his TED talk is full of it.
Barry Schwartz's TED talk on the Paradox of Choice: http://www.youtube.com/watch?v=VO6XEQIsCoM
Posted by: Joe at Nov 24, 2009 5:04:37 PM
I choose using heuristics. My heuristics include: pistachio, black, semillon, Paris....
Posted by: dearieme at Nov 24, 2009 6:31:17 PM
The thing that struck me especially about the jam choice was that the experiment was not set up the way I (and I presume, others?) shop. I usually have a clear idea about what I want to buy (a specific brand of cherry preserves), search for that, buy it if I can find it, and widen the search choice from there if I cannot. I may search for other brands, and then other flavors, but I walk out of the store with something approximating what I came in for in some way. The only way I might leave empty handed would be if the only choice was grape jelly. Since that seems to be the most popular choice, I am as grateful for the wide range of choices as others are. The experiment, as I recall, waylaid people with no strong prior commitment and asked them to choose from a limited or wide range of choices with some kind of enticement to buy. Unless my shopping is terribly unique, I don't see any reason to believe that the experiment reflects the way humans look for things to meet their needs. Rather, it tests the way that they take advantage of unforeseen opportunities.
I can think of at least one real-world problem and solution to unforeseen opportunity: I am occasionally but not frequently attracted to art I see at street or Renasissance fairs. I used to get contact information and thought I would look at their website and buy the one I liked best. I never bought anything that way, possibly for the reasons explained by Schwartz (fear of not optimizing?). Now, if I see something I like, I buy it at once if I can afford it. I buy more art now than before, and am happier for it.
Now if I could just get around to hanging some of it.
Posted by: Eric H at Nov 24, 2009 6:52:19 PM
Trader Joe's has plenty of choice. They have tons of different kinds of cheese, turkey breast, nuts, etc. Sure there aren't multiple brands of the same *exact* item, but why wouldn't it be just as daunting to select from 17 different "trail mix" combinations?
Posted by: Andy at Nov 24, 2009 7:19:16 PM
Interesting thread and thanks to the many commentators.
For me, "paradox of choice" definitely exists for "big" purchases or commitment purchases (i.e. gym membership). When I have to make a choice that is difficult to reverse and each option has pluses/minuses, I really don't enjoy having a lot of options. I notice this in myself in a very direct manner: I get irritated and immediately think "I need to figure this out later, I probably should ask some people, I should look at X and Y." Honestly, in many of these cases, if there was only one reasonable choice, I would happily choose it.
Interesting, this also exists for items that feel very personal to me (my laptop, cell phone). But then again, these choices often have "difficult to reverse" as a quality too.
Posted by: Mark at Nov 24, 2009 7:21:06 PM
I have no difficulty choosing from a hundred entrees, but I would never want to specify even a fraction of that for a single one or I would cook it myself from a fairly modest set of choices.
Posted by: Lord at Nov 24, 2009 7:32:01 PM
The only time in which large number of options to chose from are bad is when you have no idea about what the choices are and have a limited time or ability to review them.
For instance, if your an American and go to an Asian country and go to a restaurant with 500 dish choices and not a single one of them has a name similar to what you have had in the USA and no pictures, you would have a very hard time picking something you are likely to enjoy.
Posted by: astonerii at Nov 24, 2009 10:26:26 PM
I think we should be equally wary of saying more options are always better. I think it is a good argument ceteris paribus but we also have to consider do we want a lot of bad options or a single good one. Variety isn't always free.
Posted by: Simon at Nov 24, 2009 11:49:24 PM
To factor in exchange must change the dynamic. A person may spend more time choosing a jam if that jam also costs her a dollar of her own.
Posted by: Casey Pahl at Nov 25, 2009 12:41:02 AM
I dislike Subway because I have to choose a bread, cheese, toppings, mustard, etc. I would prefer just to order a turkey sandwich without making a bunch of micro-decisions. However, although I think there may be something to the paradox of choice, I resist the policy implications drawn by some that: (1) I feel the same way about 401(k) as I do about turkey sandwiches; or (2) I would be better off if the government would restrict choice in turkey sandwiches.
Posted by: FE at Nov 25, 2009 10:22:32 AM
It isn't a matter of choice, but a matter of decision, their number and difficulty.
Posted by: Lord at Nov 25, 2009 3:18:06 PM
My colleague Jayson Lusk and I have also run some of these paradox of choice experiments, which were published at...
Norwood, F. Bailey, Jayson L. Lusk, Bharath Arunachalam, and Shida Rastegari Henneberry. 2009. “An Empirical Investigation Into the Excessive-Choice Effect.” American Journal of Agricultural Economics. 91(3):810-825..
We also the "paradox of choice" experiments difficult to replicate. However, we also ran an experiment where we directly asked people whether they wanted a large (30 item) choice or a small (6 items) choice, where the 6 items would be randomly chosen from the set of 30...a majority of people chose the large variety, but a very large percentage (40%), I think did choose the small variety
The point is that some people have blatantly stated they prefer less to more....that is, if you think we ran a good experiment :)
Posted by: Bailey Norwood at Nov 25, 2009 7:16:03 PM
I concur with the PC/Mac story. If you try to buy a PC, even from a big brand like Dell, you are faced with millions of possible combinations. I work in IT so I should know what's best; yet even I was paralyzed.
In the end I bought an iMac. Just four to choose from. Simples.
Cars are a similar story. The number of possible combinations of options (air-con, CD changer, leather seats, etc.) runs into the tens of thousands. To avoid this problem, manufacturers group the options into class levels, e.g. Ford Focus S / SE / SEL / SES. This is a long-established trend in the motor industry, and it certainly suggests that real-world companies acknowledge the paradox of choice.
Posted by: Andrew at Nov 25, 2009 7:42:35 PM
There seem to be circumstances where choice is counterproductive but, despite looking hard for them, we don’t yet know much about what they are.
Web search is an obvious example of where 10 blue relevant links are far preferable to millions of choices.
Posted by: asdf at Nov 26, 2009 3:47:19 AM
From Stefano DellaVigna's "Psychology and Economics: Evidence from the Field," JEL June 2009, page 355:
Marianne Bertrand et al. (forthcoming) examine the impact of a small or large menu set in the context of a field experiment on the mailing of 50,000 loan offers in South Africa. The authors randomize, among other things, the format of the table illustrating the use of the loan. The small-table format lists only one loan size as an example, while the big-table format presents four different loan sizes. The finding is consistent with the choice avoidance results. The take-up in the small-table format is .6 percentage points larger compared to a baseline of 8 percentage points, an effect size equivalent to a reduction of the (monthly) interest rate by 2.3 percentage points.
Choi, Laibson, and Madrian (2006) also provide evidence from a field experiment that a smaller number of investment options increases participation in a 401(k) plan. Participation increases by 10 percentage points when nonparticipating employees receive a card that allows them, if mailed back, to enroll in a default plan (3 percent contribution in a balanced fund).
Posted by: John at Dec 10, 2009 9:05:04 PM
In my online product marketing, I've found that more products on a page equals LESS BUYERS.
But not all products are created equal. Lots of options for an impulse buy can be overwhelming to a consumer: "Who cares about 100 different kinds of gum, just show me the top 5."
But when purchasing a vehicle or a home, consumers like to peruse a bit more.
Posted by: David Lupica at Jun 10, 2010 4:10:25 PM
"Costco has done very well for itself offering its members/customers limited choices."
Unfortunately, I do not have the choice of shopping there. Sams seems to be okay, though. And Wallyworld. And Albertson's. Some things I prefer to get at the farmer's market or the co-op, like local eggs and produce. And then there's the local Trader Joe's knock-off. And there's the poor folks store that always stocks lots of club soda (we don't drink sweetened drinks, and we prefer our water to be bubbly but not expensive).
Each one of them offers a limited supply of stuff. I agree, it's great to have limited choices.
Posted by: Eric H at Jul 29, 2010 9:30:04 PM
