Did you buy laundry detergent on their most recent trip to the store? Also comments on scientific publication and yet another suggestion to do a study that allows within-person comparisons

Please answer the above question before reading on . . .

I’m curious after reading Leif Nelson’s report that, based on research with Minah Jung, approximately 42% of the people they surveyed said they bought laundry detergent on their most recent trip to the store. I’m stunned that the number is so high. 42%??? That’s almost half the time. If we bought laundry detergent half the time we went to the store, our apartment would be stacked so full with the stuff, we wouldn’t be able to enter the door.

I think we buy laundry detergent . . . ummm, how often? There are 40 of those little laundry packets in the box, we do laundry once a day, sometimes twice, let’s say 10 times a week, so this means we buy detergent about once every 4 weeks. We go to the store, hmmm, about once a day, let’s say 5 times a week to put our guess on the conservative side. So, 20 trips to the store for each purchase of detergent, that’s 5% of the time.

Compared to us, lots of people must (a) go to the store very rarely, (b) buy really small containers of laundry detergent, (c) do the laundry all the time. I suppose that, all the time they lose by doing the laundry all the time, they save by not having to go the store every day. Some sort of cosmic balance.

P.S. Regarding their research, Nelson writes:

Minah and I could cheerfully use the same data to write one of two papers. The first could use a pervasive judgmental bias (18 out of 18 products show the effect!) to highlight the limitations of human thinking. A second paper could use the correlation (.95!) to highlight the efficiency of human thinking. Fortunately, this is a blog post, so I get to comfortably write about both.

I guess Nelson is joking here, but just in case he’s not: of course you should convey both results in your paper! One good thing about working in political science is that it does seem that our field has a good tolerance for complexity and ambiguity. It’s ok for us to write papers that convey mixed messages, if that is what we find. If it’s really the case that psychology papers need to have a single-minded focus, that’s too bad.

P.P.S. But before writing this paper I think the authors should do a within-person experiment (that is, asking the same people the same questions multiple times) with an aim toward decomposing their false-consensus finding (that recent detergent buyers overestimate the frequency with which people buy detergent, and recent detergent non-buyers underestimate that frequency) into two parts: the general frequency (we rarely buy detergent at the store so we, quite naturally, think of other people’s shopping habits as being in some general sense like ours) and the most recent behavior (it would seem more strikingly “irrational” if we make judgments about others’ purchases based on what we just did yesterday).

31 thoughts on “Did you buy laundry detergent on their most recent trip to the store? Also comments on scientific publication and yet another suggestion to do a study that allows within-person comparisons

  1. “Did you buy laundry detergent on their most recent trip to the store?”

    Is this a trick question about taxes or inflation?

    Also, I searched “42% Americans” to see what else came up. Perhaps it is the same 42% that approve of Obama?
    http://www.washingtontimes.com/news/2014/apr/2/obama-approval-rating-42-percent-americans-cant-de/

    Or this has something to do with creationism?
    http://cnsnews.com/news/article/michael-w-chapman/poll-42-americans-believe-god-created-humans-10000-years-ago

    Another thing is that detergent is used by some as an alternative currency, so perhaps that it pushing up the numbers:
    http://blog.ted.com/2013/07/25/10-alternative-currencies-from-bitcoin-to-berkshares-to-sweat-to-laundry-detergent/

    • I can vouch for the ‘laundry detergent is money’ hypothesis. My mother and her friends had quite a trade in cups of laundry detergent. They didn’t quite get a futures market going, but I can recal childhood errands to either take or collect a cup of laundry detergent from Mrs Grosspillerwurt or Aunt Caramelnutcake (names are falsified to protect the guilty)

  2. “Did you buy laundry detergent on their most recent trip” — on WHOSE most recent trip? I never buy laundry detergent on other people’s trips to the store.

    • Actually, I think the small print says they asked people about intention to buy on their own next trip, but assumed that this was equivalent to their last trip …

    • The point of this is that “did YOU buy laundry on THEIR most recent trip” is grammatically strange.

      probably what was meant was “did THEY buy laundry on THEIR last trip”, or “did YOU buy laundry on YOUR last trip”. Both of which refer to the same entity in the subject of the sentence and the object in the prepositional phrase.

    • He’s in New York, he travels by bike. He probably passes a lot of stores. New York residences are not known for their size, he can’t go to Costco and buy several months worth of canned goods.

      (I buy laundry detergent from Costco, and probably get it once every 3-4 months.)

  3. Andrew’s living in NYC & higher purchasing power makes a difference I guess; stores might be closer.

    Back in grad school, it was fairly common to do a Walmart run once every 2-3 weeks from what I remember.

  4. Could this be a geographic bias? Urban dwellers visit stores far more often, I would guess (due to less storage at home, walking vs. driving, convenience).

    When I was growing up in Utah, we would go to the store about once a week. Now, in Boston, it’s closer to 3-4 times per week. More if they didn’t have grocery delivery around these parts.

  5. As it says in Nelson’s blog, this is an accessibility/availability heuristic bias and fits within the area of cognitive aspects of survey methodology (a topic that NIH and others got going in the 1980s and is published fairly often in outlets like Public Opinion Quarterly and Applied Cognitive Psychology). So, there are lots of interesting memory effects about this question. Another aspect that is interesting is how parts of the question help to define other aspects. I am guessing when most people read this they defined “store” as being some location where they can buy detergent. If it were “pens” or “beer” or “fresh veggies”, it might mean Andrew, Leif, and Minah include other types of stores when consider what their last visit to a store was.

  6. In addition to the possibilities that people “(a) go to the store very rarely, (b) buy really small containers of laundry detergent, (c) do the laundry all the time,” how about (d): people don’t tell the truth in surveys? I’d vote for that one, and I’d also agree that 42% seems bizarrely high. This seems like a simple one to test within about half an hour, by observing what fraction of carts at a grocery store have laundry detergent in them. (Assuming that ergodicity holds for supermarket shopping!)

    • Sampling carts is not the same as sampling people. Imagine 36% of people go to the store once per month and buys detergent every time. 64% go there ten days per month but buy detergent only once per month. If you survey people 42.4% will answer they bought detergent on their last visit. If you look at customer purchases on a given day only 14.8% of the carts will include detergent.

  7. Did they say how they measured the ground truth? In the blog, they claim they know “the correct answer”. Are they assuming that everyone can recall what they just bought at the grocery store? (btw, your “most recent” trip could be weeks ago.) That is a poor assumption. One of the papers I referenced in Numbersense is a study done by marketing professors in which they had mystery shoppers observe and record what people put into their shopping baskets. That’s how you measure ground truth. They found, by the way, that shoppers don’t recall what prices they are paying for the objects they placed in the baskets when they were intercepted 30 seconds before.

  8. I’m pretty sure NY residents are unusual in their store-going habits relative to most other Americans. The usual behavior in middle-class US households is to go infrequently to a biggish store like a Target or WalMart to stock up on things like laundry detergent, paper products, aluminum foil, toothbrushes, shampoo, and other similar consumables, and to go to a separate grocery store for food probably around 1 time per week. The Target/WalMart type stores are typically a couple stories, the area of a whole city block, and you go in a car, buy a large basketful, stick it all in your trunk, and then put it in your garage. I’ve found prices to be maybe 20% less at these stores compared to grocery stores, so it does make sense to make the extra trip. The store is often 1 to 5 miles from your house, so car trips are the only thing that make sense (except poorer families that do this by bus).

    That’s the way it’s worked for people in CA and IA, the only two states where I’ve spent any significant time, but it’s also been similar for family that I’ve visited in say WA or OR for example.

    Manhattan island, on the other hand, has small stores all over the place, and owning a car is often prohibitively expensive. People often walk or take the subway. They might pass 4 or 5 stores that they frequent on their way to or from work. It would cost essentially nothing in terms of extra time to stop by the store on the way home from work and pick up laundry detergent and salad greens or whatever.

    Living in the Pasadena area, I probably frequent Target about once every 3 to 6 weeks. For grocery items I typically go once a week, sometimes twice to a grocery store. I split my groceries between a discount place with lots of cheap produce, and a typical chain whose prices on canned goods and cereal etc is better. And I could see that I might buy laundry detergent on every visit or every other visit to Target because I don’t go that often.

    So, given all that analysis, I suspect there are two major contributing issues to the disconnect that Andrew had:

    1) People go to places where they might buy laundry detergent a lot less frequently than Andrew, who lives in NY. Typically this might be a store they go to mainly to get household consumables.

    2) When asked about laundry detergent, the question primes the responder to think about the store where they get laundry detergent, which typically “specializes” in household consumables, so they aren’t thinking about their last trip to a grocery store to get a deli sandwich for lunch or whatever.

    3) Given the infrequency with which people might go to the store, their recollection is hazy of whether they actually got laundry detergent last time they went.

  9. The question asked is ambiguous. Note in their first paragraph they have both “store” and “supermarket” listed. Supermarket is a subset of store. My last two trips to the store involved a Love’s truck stop to buy gas (but they do sell laundry detergent) and a trip to a Jewel supermarket — but to visit the ATM, not to shop, although I could have walked 30 feet and bought laundry detergent.

    Since their purpose is to highlight a false consensus bias, asking an ambiguous question is probably appropriate. If you were doing research on laundry detergent you’d be a fool to ask the question this way.

    The actual numbers? For stockup trips (i.e. the largest classification of trip, where you are getting a lot of groceries) the percentage of time laundry detergent is bought is about 11.5%. (Kaiser, that’s based on actual purchase data.)

  10. For comparison, I do 1 load a week, shop 1 or 2 times a week, and at 40 loads, buy detergent 1 or 2 times a year. Double that if you include other cleaning products.

  11. Another issue is at least here in the Northwest, detergent goes on sales during cycles, so if you sample close to a cycle then you would expect more savvy shoppers to have bought it recently, if you sampled “off-cycle” then you will get much less. Detegent is a infrequently purchased but significant cost product so you should expect more cognition and awareness fo its price cycles than let’s say eggs or milk.

    It’s like asking “Did you buy Fruitcake the last time you went to the grocery store?” You are going to get a much higher positive percentage in December than you would in July.

  12. > Did you buy laundry detergent on their most recent trip to the store?

    No.

    > I’m stunned that the number is so high. 42%???

    Me too. Maybe they sell 0.5 oz bottles of it in Jung’s study area?

    > So, 20 trips to the store for each purchase of detergent, that’s 5% of the time.

    Yeah, that seems about right.

    • Well, sure, it’s like asking to settle one of those debates: “Think of the color of the skin of the last person who sexually harassed you on a street.”

      42% of American women might answer black, even though blacks are not 42% of the population.

  13. This is what supermarket scanner data is for: answering questions like this objectively rather than asking people to make dumb estimates. Thirty years ago, the firm I worked for had 10,000 households in four towns having all their supermarket and drugstore purchases reported. It’s a lot better than calling people on the phone and asking them.

Leave a Reply

Your email address will not be published. Required fields are marked *