« I am dared to believe the dollar is undervalued | Main | China fact of the day »
Who waits longest for coffee?
Tim Harford reports:
Caitlin Knowles Myers...with her students as research assistants, staked out eight coffee shops in the Boston area and watched how long it took men and women to be served. Her conclusion: men get their coffee 20 seconds earlier than women. (There is also evidence that black people wait longer than white people, the young wait longer than the old, and the ugly wait longer than the beautiful. But these effects are statistically not as persuasive.)
This does not seem to reduce to greater complexity of drink for the female customers, read these clarifying remarks from the researcher. One question I have is when the order counts as having started; in my family I am sure that the women spend more time ordering. The simplest explanation, however, is that the staff feel more implicit psychological pressure to meet the needs of the male customers. I've also found that indecisive behavior at the counter leads to slower service, I have one particular (male) friend in mind. I am myself virtually always a decisive orderer.
Along not totally unrelated lines, here are new but not surprising results on which waitresses receive the biggest tips.
Posted by Tyler Cowen on November 15, 2007 at 06:40 AM in Food and Drink | Permalink
Comments
"I've also found that indecisive behavior at the counter leads to slower service, I have one particular (male) friend in mind."
Perhaps that's because of his irrational fear that he might work in the coffee shop someday...
Posted by: Ironman at Nov 15, 2007 7:48:34 AM
This was discussed on the Starbucks gossip blog. One comment was:
I don't think that the men get treated "better." The women wait longer for drinks because we can be more conversational with them, which is better treatment then the men who just want to be in and out of Starbucks.
So by that account, they're not getting worse service at all.
Posted by: Theophile Escargot at Nov 15, 2007 8:06:13 AM
Tyler, the serving time is counted from the point the order finishes (see Myers' comment that you link to), so that shouldn't be a concern.
There's also been a suggestion that the longer wait time could be due to people taking more care making the coffee, which would again suggest better rather than worse service.
On the other hand, it could just be worse service.
Posted by: conchis at Nov 15, 2007 8:19:30 AM
Did the world suddenly run out of important issues to research?
Posted by: Russ R at Nov 15, 2007 8:44:50 AM
Pray tell us what the important issues are Russ.
Posted by: jason voorhees at Nov 15, 2007 8:58:20 AM
Jacob Grier, former barista and now at Cato, offered his explanation here: http://www.jacobgrier.com/blog/archives/831.html
Posted by: Paul Sherman at Nov 15, 2007 9:09:01 AM
They should repeat this study in a bar, and watch the results reverse (that is if it measured from the point the customer reaches the bar.)
Posted by: Trent McBride at Nov 15, 2007 9:30:44 AM
This does not seem to reduce to greater complexity of drink for the female customers, read these clarifying remarks from the researcher.
Not to be a grammatical pedant, but this seems to crop up regularly: that comma should be either a period or a semicolon.
Posted by: Stuart Buck at Nov 15, 2007 10:04:50 AM
This issue was discussed on fark the other day.
Posted by: BlogReader at Nov 15, 2007 10:27:31 AM
This bit in Myers's reply bugs me: "It may also be employees wanting to chat with women more then men."
Weren't she and her students sitting there watching? She should know from experience whether this is the case or not.
Posted by: Dolohov at Nov 15, 2007 11:18:18 AM
Yeah Russ, what's more important than coffee and sex?
Speaking of gender and service, did you know that fertile lapdancers earn more (and much more than menstrating ones)?
Posted by: David Zetland at Nov 15, 2007 11:23:23 AM
Did the study control for gender differences in regularity of drink ordered? I get the same drink every day at my coffee shop, so the barista starts making it as soon as she spots me walking in. If more men than women had a regular drink, this would explain some of the difference.
Posted by: Brandon Adams at Nov 15, 2007 11:52:53 AM
interesting. In every Starbucks that I frequent (5 between home and office) the cashier and person making the drink are different, the shops are very busy, and the orders are queued so that that are made sequentially. Short of reading the name on the cup I doubt that the drink maker has a clue about the gender, race, or any other characteristic of the customer.
Posted by: mike at Nov 15, 2007 12:14:35 PM
1. I'll have to ask my daughter about this (she took a college hiatus and served as a barrista / shift manager at a Starbucks).
2. Without asking her, though, I suspect she was more chatty with the female customers. Chatty young blonde barristas may have their chattiness misunderstood for flirting by male customers, and my daughter clearly found this a bit tiresome. Sullen, precaffeinated men may also not particularly want to chitchat.
This would imply that good service for a middle aged man is getting him his coffee, stat. Good service for a middle aged woman may be a few seconds of conversation showing some interest in her life.
3. Perhaps I should interject my theory on fru-fru drinks here. It's not that women really prefer half-caf non-fat soy cappuccino with a shot of vanilla and men prefer a tall black. It's that women are more likely to feel valued by a special ordering process. [my theory is purchase dependent, and the effect likely reverses for sportscars or boats] So, if a woman just orders a tall black (which may be because their budget does not extend to the extravagance of a special drink), she may still appreciate being fussed over a bit, perhaps by being asked "room for cream"?
Posted by: ZBicyclist at Nov 15, 2007 4:27:36 PM
Contra ZBicyclist , I think men get served faster precisely because young baristas are flirting with them...
This is especially true if the guy is a regular; it's very flatering to have your order ready by the time you reach the counter.
Posted by: tadhgin at Nov 16, 2007 3:24:05 AM
If you read the full article, you'd see that
the wait time disparity disappeared with female
barristas waiting on the customers. So if women
customers wanted more of a chat, you'd think
the female barristas would realize that and give
it. Also, when the coffee place was crowded with
long lines, the wait time didn't disappear,
which throws a damper on the "they're just flirting/
chatting" argument, as why would you waste time
doing that when there is a line?
Posted by: Melissa at Nov 16, 2007 7:44:17 AM
I only have my time spent at Starbuck's and my girlfriend's experiences to draw on, so here we go:
1. Most baristas of the male gender are, to put it mildly, less than imposing physical specimens, so they would be more comfortable chatting up women (this observation is probably skewed by the location of my Starbuck's-next to the local Y, and would have more big, imposing guys coming in than the average).
2. Study authors can say what they want, but the things that some women order begger the imagination, and seem to take 2 minutes just to order.
3. As with everything else in life, women tend to want their coffee JUST RIGHT, and, like my girlfriend, have no problem bring it back up to be "fixed". This would make baristas loiter a bit to punish or take extra time to try to make the drink perfect. I have a woman friend who gets a cappuchino every day and complains that it's never as good as they make in the North End!
3. Women are notoriously bad tippers (unless they were once waitresses). 80% of the women I see at the coffee shop pull out the change for the exact amount. My girlfriend refuses to tip for coffee. I tip pretty well, and it paid off at my local Starbuck's. I usually get a cinnamon chip scone, which they run out of pretty early, so a couple of the baristas (young ladies) began to save one for me!
Posted by: Brutus at Nov 16, 2007 8:05:41 AM
This is hardly strong evidence point towards discrimination as the writer in FT tries to suggest. It is effectively a blank slate that allows the reader to project their preconceptions and biases on it.
Could this effect be discrimination? Sure, at least once you can define the notion for me in an coherent principled fashion. However, it could also be a huge number of other things as well.
It could represent the bias of the person doing data recording. It could represent a bias induced by women ordering more softly (either in the data or the filling of the coffee). It could reflect the fact that on average women in coffee shops are less uptight about time than men (just a small percentage of stay at home moms might be enough to establish this). It could reflect subtle clues the people ordering the coffee conveyed.
Or my favorite this could simply reflect a bias inherited from the well known disproportionate representation of white men in executive positions who either have the freedom to get coffee at less busy times or pay more for it.
Or what if it's just that men were more pushy and more likely to get mad at baristas. Would that count as discrimination?
----
Besides in situations like this discrimination is really more of a term that people use to project their world view into the evidence it's not an accurate or particularly meaningful notion here. Of course there are very clear cut cases of discrimination but when we are talking about complex social interactions like this it is just a mess.
I mean suppose it was the other way around and men took 20 more seconds but it turned out this was because the baristas flirted with them. Would this be discrimination? What if this instance was the result of male baristas flirting with women? If the women don't like being flirted with does this change the answer? Given that men and women do act differently in social situations what sort of different behaviors would justify reacting faster to one gender than the other? Does responding to apparent anger count as discrimination if the other gender is socially encouraged not to be angry? Ultimately there is the very real danger here than any principled distinction you want to make will collapse down to the mere question of who ends up overall better off and to decide if their is discrimination we have to do overall happiness surveys.
This isn't meant to suggest there aren't clear cut cases where the term obviously makes sense but in more subtle questions like this we should be asking what changes could reduce unnecessary suffering rather than fighting it out over a questionable and vague notion of fairness.
Posted by: TruePath at Nov 16, 2007 8:21:23 AM
Posted by: 翻译公司 at Feb 25, 2008 9:53:30 AM
Posted by: 鑽石 at Apr 2, 2008 11:26:20 PM
Posted by: Alii at Apr 3, 2008 9:16:30 PM
