Do Coffee Shops Discriminate Against Women?

We’ll make this brief so we can jump right into the heated comments: an economist and her research assistants studied eight different Boston coffee shops and found that, on average, women had to wait about 20 seconds longer to be served. She controlled for drink types and the discrepancy remained. What’s more, “The delays facing women were larger when the coffee shop staff was all-male and almost vanished when the servers were all-female.”

There’s some speculation that the delays are caused by flirting, but the wait time increased the busier the shops became, which runs counter to any flirtation theory (assuming it’s more difficult to flirt acceptably when there’s a long line of people waiting to order).

Some unanswered questions: is Boston more chauvinistic or sexist than, say, Berkeley? What about a large city in the South? Do women take longer to order, even when they’re ordering the same drinks? We’ve noticed that at Starbucks, we tend to order coffee in, like, 3 seconds, while our communal Consumerist girlfriend takes upwards of half a minute while she weighs her options, gives precise instructions, or makes small talk. But of course that doesn’t explain why an all-female staff nearly removes the wait time.

In general, the study seems to imply, you’re likely to get the best service if you’re an older, handsome, white (or at least non-black) male:

There is also evidence that blacks wait longer than whites, the young wait longer than the old, and the ugly wait longer than the beautiful. But these effects are statistically not as persuasive.

The question now is, if current coffee shops discriminate, does that open an opportunity for competitors to come in and steal business by offering better service? One problem is that the discrimination is trivial enough that consumers might be willing to overlook it—if they notice it in the first place.

…a rival coffee shop would have to be very close indeed to justify a trip aimed at avoiding a 20-second wait. Even coffee retailing isn’t that competitive.

“Waiting for Good Joe” [Slate]
(Photo: Getty)


Edit Your Comment

  1. louisb3 says:

    How big is twenty seconds relative to the average total wait?

  2. headon says:

    You know we keep giving women rights and they always want more. We keep this up and one day one of them is gonna try to run for president

  3. upokyin says:

    communal Consumerist girlfriend?

  4. ptr2void says:

    So these discriminating shops are staffed by gay male golddiggers then? ;)

  5. DrGirlfriend says:

    How did they even notice this as enough of a trend to prompt a study? Not just the extra 20 seconds, but also that it only happens to women? It just seems so minimal as to be almost imperceptible.

    Maybe my barista is taking extra special care with my drink because I’m so darn cute.

  6. forever_knight says:

    any thought given to the fact that heterosexual men want to keep glancing at these women? hence, the wait time??

  7. deadhouseplants says:

    I read this article until I got to the 20 seconds part, and realized it’s just another feminist movement pulled from the ass of Jane Fonda. I’ve said this one, and I’ll say this again, equality does not equate to superiority. I will never take an American feminist seriously until she lives 5 years in the Middle East or Southeast Asia, then tell us how bad it truly is in America.

  8. MercuryPDX says:

    One problem is that the discrimination is trivial enough that consumers might be willing to overlook it-if they notice it in the first place.
    If it’s THAT imperceptible, then I think the study author is seriously reaching to make her point.

    Unless they put some secret code on the cup at Starbucks to denote “Female customer”, I’m calling BS on this. I’ve never seen this type of “discrimination” at any of the Portland Area coffee shops I’ve been to.

  9. Saboth says:

    I’ve actually made a little hobby of seeing what people order. My observation? Here is a man ordering (generally):

    “Give me a #1, light creamer, thanks.”. Pays with a $5 and leaves.

    “Hmmm well I’d like a Cafe Mocha, light on the cream, but not too light. I’d like some blue sprinkles, and 1 oz of whip cream, also 1.5 packs of sugar, but make sure it is not Nutrasweet. Also, do you offer low fat muffins? You do? How many grams of trans fat are in them? Ohhhh that’s too much. How about bagles? Good, I want it lightly toasted with 2.5 oz of cream cheese….” 2 minutes later, the order is completed. “Now, who do I make this check out to?” Sits down…comes back. “I said light creamer…can I get another one?”

  10. Omi says:

    It’s probably some subconscious thing that the employees don’t even notice themselves doing, surely this is not intentional. I mean I don’t think that every male staff member is thinking “Oh a woman’s come up to buy coffee, well I’m gonna spite her and take an extra, almost unnoticeable few seconds to serve her.” At least I would hope that’s not the case.

    I would like to see the same test done to see if maybe a staff of all females takes longer to serve male customers, if a black staff takes longer to serve white customers, etc.

  11. sonichghog says:

    Maybee women complaine more, so they take extra time to get the drinks ready.

    Its not that they are treating the women badly, they are just giving the men a lower quility product.

  12. Jasmo says:

    If you even casually read this study (link from the slate post) you will find that the “enumerators” recorded information by methods such as “… ranked each customer’s appearance on a scale of 1 to 10. ” If this sort of subjective judgment, along with the numerous other factors they list, is the basis for coming up with an average of 20 seconds difference, one wonders if the margin of error far exceeds this 20 seconds.

  13. GreatCaesarsGhost says:

    Average wait time was 99.3 seconds.
    Sample set was 277 orders.

    Assuming that women order more complicated “fancy orders” than men (the opposite may be true), you really should not consider that data.

    For “non-fancy” orders, the wait time was 47.5 sec for men and 56.3 for women. Over such a small data set and with enumerators that are hoping to record bias, that doesn’t seem so bad.

  14. mandarin says:

    So this study has 1 conclusion:
    A lot of women in Boston use coffee shops.

    Don’t people in Boston have anything more important to study on these days? Sheesh. If you think males in Boston are sexist when it comes to serving women in coffee shops, just announce it in the open and stop making useless studies…

  15. DeeJayQueue says:

    I’d have to say that it’s not the coffee shop that’s sexist, it’s the people working at the coffee shop that are sexist. I bet the same results would repeat themselves in just about any retail situation.

    And, does a longer wait = worse service? Maybe she’s getting extra attention paid to her drink.

  16. harumph says:

    @deadhouseplants: to be fair, neither women or men from the u.s. can affect the laws or customs of foreign countries. are we to stop trying to better our lot in the u.s.(on any issue not just feminism) because it sucks worse somewhere else? i just find that argument a little silly.

  17. embean says:

    I work for Starbucks. I would say 20 seconds is very significant, but cannot be true. Females have much longer drink orders, but guys are more chatty, probably because they feel awkward not knowing how to order. In general, there are way too many customers for us to give a crap about something like this. When you have a 5000 person line, you don’t think about much.

  18. 12monkeys says:

    WOW i would serve her quick!! Hot asian babe

  19. AnnC says:

    The moral of the story is that people subconsciously treat other people differently. There are numerous studies that show this (drivers wait longer to honk at nice cars at a green light then at junkier cars; pedestrians are more likely to follow a person in a suit jaywalking then some one dressed casually; white interviewers are more likely to sit closer to white interviewees than black interviewees).

  20. bonzombiekitty says:

    I’d be willing to bet than any additional time it takes a man to serve a woman coffee is due to (un)conscious examination of the girl. I.e. “is she pretty?”, “is she my type?”, ” would I be willing to have sex with her?”, “would she be willing to have sex with me?”

  21. savvy999 says:

    I’ll take the Occam’s Razor approach… maybe it’s not some insidious sexist/flirtacious conspiracy… how about the simple explanation that maybe male workers in Boston are just generally more incompetent than female workers?

  22. ParadigmABQ says:

    @headon: “…one day one of them is going to try to run for president.”

    …oh wait.

  23. BrianH says:

    We’ve got kids dying in Iraq, global warming, a plummeting dollar, and some jackasses are worried about waiting 20 seconds longer? Beam me up, Scotty…. no signs of intelligent life down here.

  24. LucyInTheSky says:

    i am a total feminist but this seems awfully trivial. how do you control a study like this anyway?

  25. nffcnnr says:

    i read somewhere that it takes approx. 18 – 22 seconds to quietly hock up a quality half-ounce loogie. Maybe that explains the extra wait.

  26. deadhouseplants says:

    Okay, so I read the research paper, including the statistic at the end and I came with this conclusions. If you’re a black female over the age of 40 and ordering a fancy drink, don’t even bother, cause you’ll be there for years.

    Imagine though, you’re working at a coffee shop, and a customer pulls out a stopwatch, how would you respond to that. All I know is this, if this is what is counted as a graduate research paper at Middlebury University, I want to transfer there. I mean come on, my graduate paper is on the housing pricing infrastructure of Metro Portland, Oregon, in comparison to economic classes. Wow, if only I had known that all I had to do is go to any place of business and try and find some form of prejudice to get my Masters.

  27. spinachdip says:
  28. AnnC says:

    @savvy999: But for some reason, the male workers are more incompetent to female customers.

  29. BeFrugalNotCheap says:

    Well as a self proclaimed smart ass I’ll hold back on some knee jerk comments on this story. But I will say that after visiting several coffee shops in my short life I have never noticed anything like this. That’s not to say it does’nt actually happen. As I think about it more these results seem almost suspect. When I worked at a rather upscale chocolatier I tended to treat everyone the same just to avoid being thought of either as a racist or sexist. Even the funny little guy with the nervous tic that walked in mumbling about some samples got normal treatment. Since we are all cool with each other here I will say that a beautiful woman did get a little more time. Uh, but I swear it was unintentional……

  30. johnva says:

    The study only included 295 transactions, and it wasn’t very carefully designed. The data set was so small that it almost certainly doesn’t tell the whole picture, especially when they were trying to control for all these factors (making the effective data set even smaller when comparing apples to apples). They basically just had people manually time the transactions, which is probably an error-prone process (perhaps video would be more precise).

    More likely than discrimination being the cause of this is that they aren’t controlling for some other factor that tends to be different between men’s transactions and women’s transactions. For example, it’s possible that men and women communicate their orders differently and women tend to take slightly longer for the same drink order (especially when dealing with a male employee). I really doubt that most young Starbucks employees are intentionally discriminating against women or blacks.

  31. JohnMc says:

    I think the researchers missed a metric in their analysis. In the post to the piece the cat was let out of the bag when they said that the wait was the longest for all male staff but zero for all female staff. Operative conclusion —

    Men are slackers.

    Don’t need a PhD to figure that one out.

  32. Hanke says:

    Wait a sec, women waited longer for their drinks when served by men? Hold the phone, NASA needs to know that GUYS FLIRT WITH WOMEN!

  33. thewriteguy says:

    You could look at the results this way: Maybe the male workers subconsciously like to make sure a woman’s order is perfect — giving a little extra luv for the lady. But for other fellow males, the guys just slap it together fast. Female workers might be doing the same thing toward a female customer.

    Still, I think this study is bunk. I know a couple of friends who work at Starbucks and the only thing I hear is that they want to get customers’ orders fulfilled quickly.

  34. BeFrugalNotCheap says:


    I’m with you on that. Men are slackers. I’m no self hating heterosexual male here and both sexes have their respective quirks….But women tend to be more economical with their time and always seem to plan things out in advance. It could just be that female coffee shop workers are better aware of the trends seen throughout the day. For example the rush after people get out of work, saturday afternoons, etc. So to be ready they get everything organized and ready to go.

  35. 12monkeys says:

    Try to get a drink at a nightclub,see how long guys have to wait unless they are with a hot chick at the bar.

  36. Shadowfire says:

    @thewriteguy: Because women are more prone to complain if the drink is wrong? ;)

  37. ogman says:

    How much money was tossed away on this ridiculous study? I find it hard to believe that an economist could justify the waste of time and research funds.

  38. Sonnymooks says:


    I’m trying to figure that out myself.

    I’d go as far as to say, if you had 5 different women doing this study or reseach, you may even wind up with 5 different conclusions, especially with so many subjective elements being used.

  39. thewriteguy says:

    What if the female customer is dressed conservatively, like in office attire? What if she’s dressed casually? What if she has big boobs? What if she’s fat? Or skinny? Or walks in dressed as a clown?

    There are just too many possible variables, which could skew the results.

  40. rjhiggins says:

    @deadhouseplants: Wow, this is just mind-blowingly stupid. Following your argument: Black people are being murdered by the thousands in Darfur. Therefore, black people who are discriminated against in the United States have nothing to complain about.

  41. cflury says:

    Look, men want to flirt with/look at the women so they are going to take a few extra seconds pouring that late than if you were a smelly fat guy. I would like to see looks factored into this poll as well as sexual preference of the workers, do gay guys take longer to serve good looking guys? I think all these things need to be taken into account before we can start saying that men just don’t want women to have a freaking cup o joe.

  42. dugn says:

    This has got to be the dumbest study-of-the-day meant to outrage and incense this month. Not only do I agree with previous posts that this sampling to too statistically insignificant, that it doesn’t follow sufficient scientific principles and skimps on other causally related details, what about less scientific variables (WRITEGUY mentioned attire, attractiveness, etc.). Add to that, perhaps more time was taken to get the women’s orders ‘just right’. Perhaps more time was taken to offer superior customer service. Maybe it was the part of the country where women are respected and catered to (the South) or elsewhere. This is bogus science at its best meant (and I repeat) to incense and drive talk radio banter.

    Of course the 20 seconds is nearly insignificant. I spend more than 20 seconds interrupting the barrister to make doubly sure they’re using whole milk.

  43. SOhp101 says:

    Overall I’m not surprised this is true but there is a possibility that it is not malicious discrimination.

    It’s a bad study overall, though. Sounds like something that an undergrad would do, not a grad student.

  44. deserthiker says:

    Why should we assume that because the service is slower it is worse? Perhaps the servers were making sure that the drink that women ordered were prepared properly and that women got 20 seconds MORE SERVICE. It seems to me that the men were just shuttled through the line and the women actually got waited on.

    Now if women really want to complain about an inequality, they should complain that when they turn 18 they are not given to going to sign up for selective service like men are. Now that’s discrimination!

  45. metoometoo says:

    @deadhouseplants: Middlebury is not a university, and this was not a graduate research paper. It was a class project for an undergraduate seminar, and “American economist Caitlin Knowles Myers” is an assistant professor at Middlebury College. I’m guessing that the main purpose of the project was to teach undergrads how to conduct a field study, not to make a serious argument about discrimination. I have no idea why Slate picked up this story… None of the class projects I did at Middlebury ever made it into Slate…
    @ogman: And I doubt that any money at all was devoted to this study, since Middlebury is a small liberal arts college, not a research university.

  46. deserthiker says:

    @deserthiker: Edit: they are not given THE HONOR OF going to sign up for selective service.

  47. GreatCaesarsGhost says:

    Exactly. Perhaps the title should be, “Do coffee shops discriminate against men?” It’s important to note that the time metric here is from when the order is placed to when it is filled. It’s not as if women are being ignored and not waited on.

  48. krom says:

    Of course, to illustrate this story, the picture shows a woman drinking a froo-froo frappuccino (is that caramel?). Which of course take a lot longer to make than most SB drinks. Just sayin’.

  49. BigNutty says:

    Was this study set up in real coffee shops like Starbucks without drawing attention to themselves? How do you actually measure such a study.

    How many participants were involved? Were the research assistants male or female? Were they drinking coffee at the time of the research, thereby filling themselves up with caffeine?

  50. vanilla-fro says:

    @thewriteguy: Maybe they are more worried that a female, especially one dressed as the girl in the photos, is more prone to complain if her drink isn’t right where most guys would drink it anyway and get on with their day.

    Anyway, this “study” hardly seems scientific at all.

  51. johnva says:

    @metoometoo: That makes a lot more sense (although it seems they didn’t do a very good job of teaching them how to conduct field research, given the way the study was conducted). I agree with you – the real mystery is how on earth did they convince a Slate writer that this was a worthwhile study to report on as if it was conducted by real scientists? Or did the writer just randomly come across it and write about it on their own?

  52. mgyqmb says:

    I’m sorry, but when I am helping a nice, beautiful woman (or when I did…phone support now), I take my time to make sure everything gets done right. Not too long so that she gets mad, but long enough so that I don’t ruin her experience or make her mad. And do it well enough to score points.

    Why do I need to score points? I’ve got a girlfriend, will never see this girl again, and genuinely am not really interested in nooner trysts. Because I am a guy. I need to work on these skills to survive.

  53. Buran says:

    @upokyin: I hope they use protection if they share her.

  54. PaulMorel says:

    Economist desperately tries to grab headlines with pointless study!!! News at 11!!!

  55. Buran says:

    @BigNutty: My guess is an average-looking person sitting in a corner sipping a coffee with a newspaper and/or laptop on the table along with a notebook wouldn’t draw attention, especially if they were out of the easy line of sight of anyone walking toward the counter and not where they’d draw the attention of the staff. It’s easy to look like a college type in a place frequented by college types, and no one would think anything of the notebook or even the stopwatch or other timing device that was used to record wait times.

    To really do it right the researchers would send both men and women to try to even out any accidental bias brought on by the observers themselves or any bias resulting from the staff knowing that male or female patrons were sitting at tables.

  56. Buran says:

    @PaulMorel: Personally I see this as more of a psychology or anthropology study, and even the most pointless-seeming research can have surprising ties to other research. Why and how we treat others the way we do is a pretty interesting subject, actually, and it never occurred to me to think that I might be waiting 20 seconds longer for my order than the guy in front of me might be. I now find myself wondering what the cause is, and if the same root cause might turn up elsewhere.

    Of course, I’m geeky and I love science, so I find this interesting when someone else might find it silly.

  57. synergy says:

    @12monkeys: You beat me to it. To get an order filled at a bar you have to look like you have money and/or are or are with someone very good looking. Otherwise, forget it. You can make waving motions, attempt to yell over people talking or music, or in general plant yourself in their direct line of sight and STILL not get a drink order taken.

  58. Buran says:

    @krom: It’s a stock photo. Woman drinking coffee.

  59. rdm24 says:

    A couple of people missed some key points:

    The delay is statistically significant, but short. It’s a real difference. It’s a value judgment to say whether or not the difference matters–something science can never do.

    They controlled by drink type. Therefore, it doesn’t matter if a woman orders a frou-frou latte with all the trimmins’. A man ordering the same thing gets served faster.

    This study is fascinating, and I can only wonder why we get these results. There are lots of possible confounding factors, like the attire of the customer (business suit? t-shirt and jeans?), location of the business (downtown vs. neighborhood coffeeshop), type of business (indie shop vs. starbucks vs (in Boston) dunkin donuts), etc.

    I don’t think anyone believes that this discrimination is the result of overt policy.

  60. rdm24 says:

    @deserthiker: I’d like to point out that a large portion of the population (gays and lesbians) is protesting discrimination in their abiility to serve in the millitary. Some groups are willing to object to discrimination, even when it benefits them.

  61. rdm24 says:

    I hate seeing everyone protesting “bogus” science when they haven’t even read the study. You aren’t going to find all the info in the popular media, so don’t assume the study was flawed because something isn’t mentioned.

  62. johnva says:

    @rdm24: I read it. It’s linked right from the article. I still think it’s very poorly designed from a methodological standpoint, doesn’t properly control for all the possibly relevant factors, doesn’t use a large enough sample size to draw the conclusions that they do, and doesn’t provide any evidence that discrimination has anything to do with the delay, even if it’s real.

  63. metoometoo says:

    @johnva: Um, duh. It’s an undergrad class project! Not even a thesis, just a class project for an undergrad econ class. Not a sociology or psych or anthro class.

  64. johnva says:

    @metoometoo: That’s what I said – I agreed with you that that explained it. It’s not a real scientific study (though the Slate writer kind of presented it as if it was).

    I was replying to @rdm24, who said that we shouldn’t assume it is flawed just from the Slate article. It’s obviously not a real publication-quality scientific study if you bother to read it.

  65. Mary says:

    @savvy999: “how about the simple explanation that maybe male workers in Boston are just generally more incompetent than female workers?”

    That’s my assumption. I’m not sure what they mean by “being served” because my assumption was how long it took to actually get around to placing your order, the part of the transaction I hate most. But if they’re talking about the actual making of the drinks, and they say that when female workers were present the delay ended, then I would say the problem is in the workers, not the customers.

    They do say they took different drink order types into account, so saying women order more complicated drinks might not be the answer. Who knows, maybe there’s a delay because they’re making the drink better because they think women are better tippers?

  66. metoometoo says:

    @johnva: Whoops, sorry you’re right. I didn’t connect that you were the same person who made the earlier comment. I was really just responding in general to all the comments continuing to criticize it as though it were a real scientific study.

  67. krom says:

    @meiran: Could there be a belief among baristas that women customers are more particular and more likely to complain about their drink if it’s not just right? — therefore they take extra care to do their drinks right.

    Of course… this whole study seems to suggest that the person making the drinks is aware of the gender of the person making the coffee. In my experience at a typical SB, that’s not necessarily the case unless you’re the only one at the counter. One person takes orders and maybe grabs things from the case, another does warming and maybe blended drinks, and then at least one other person makes the espresso drinks. It’s unlikely that if a man and a woman both order a venti nonfat latte that the person making the drinks will know which one is for a woman.

  68. alk509 says:

    @BrianH: …said the guy reading The Consumerist…

  69. floydianslip6 says:

    This study must have been very well carried out to account for all possible variables. Yep. This was probably REALLY scientific like.