I think there's always been a tension between morning people and evening people. Those waking early are productive and accomplished. Those staying up late are up to no good. 

A new study making the rounds looked at the self-reported energy patterns of 130 participants after staying awake for 24 hours. It breaks up this previous divide by showing that this association only accounts for 73 of 130 participants. There were significant portions whose energy levels (always high or always low) didn't follow a morning/evening pattern at all. 

While I can't comment on whether this particular experiment was well designed to understand sleeping patterns and energy levels, it seems that just dividing people into morning people and evening people is a classic case of "high significance and low model fit."  There's clearly a pattern, but it didn't really explain more than 56% of the observations. 

I think it's an interesting note about the science journalism going on (an example here and here) both seem to leave the idea that the most important idea is what to call these new groups of people (anyone who's run a segmentation knows this phenomenon) rather than what these people tell the world about how we sleep and the role of sleep in our lives. 

There's paper that piqued the interest of everyone on Twitter and Facebook about the drivers of divorce. The claims that caught most peoples eyes the most were around the fact that a big, cheap wedding seemed to make it less likely to divorce. 

It's interesting that they only used 1 source of data collection (i.e. 1 survey), which asks people to remember how many people attended a wedding and how much they spent on it. These might remain salient issues to some people long after the wedding, but you could argue that for the couples that divorced, they may actually remember the event differently over time than those that remained married. 

Also, just because some participants are married now, doesn't mean they'll remain married forever. Methodologically, it would make sense to approach this as a "missing data" problem and you'd probably need to assume that some of the people who are currently married would eventually get divorced. 

Which begs the question: is a successful marriage just one that lasts until one party dies? If not, then why are people so obsessed with the longevity of relationships?  Overall, I agree with the paper's conclusions that 

I'm still making my way through the book Nudge (hopefully in time to see Richard Thaler at the Marketing Research Event in Nov). But I was listening to backlogs of Science Friday about fascinating work in energy conservation. Apparently, regardless of the efficiency of lighting, people throughout history tend to spend about 0.72% of GDP on lighting (be it kerosene, CFB, or candles, I haven't read the paper and it is nested in some work about the economic impact of solid state lighting but it is available online). If true, this would have enormous implications for public policy and another example of "latent demand."

But I digress! Someone called in to say she had dishwasher that would always default to heated dry and you'd have to manually turn it off each time. Nudge talks a lot about default options but how much water and energy could be saved if your washer defaulted to a cold rinse for clothes? (Interestingly enough, that would not be captured by the current energy star rating system!)

It seems like the coffee mug experiment has been coming up everywhere these days. In fact, one of my classes attempted to reconstruct the experiment. Rather than a coffee mug, the professor placed a Bucky Badger glass up at the front of the classroom and had half the class fill out a questionnaire saying "would you sell the mug to someone for X" where X started at $1 and increased by $1 with each question. The other half of the class filled out a questionnaire that said "would you buy the mug for X." However at the end of the experiment the professor claimed "you're all MBA students, so you're too clever for this experiment" because the mean selling price and buying price were identical.

My previous experience with this experiment however actually gave the sellers coffee mugs to hold in their hands. This experiment got the expected finding: the sellers wanted to sell it for a much higher price than the buyers would buy it for.

This key difference occurred to me after class was over: touching the mug itself. It reminded me of the work of Joanne Peck (a UW professor of course) researching the effect of touch in marketing. If the sellers could have touched that water glass (without letting the buyer's touch it of course) would the results still have come out this same or are MBAs just too clever for these well-known experiments? Experimentally though, you could also take the issue with the questionnaire being too leading and asking a single shot "how much would you sell the glass for" might prevent the participants from overthinking their responses.

(Note: my first run in with the experiment was in undergrad, so perhaps we just weren't that clever yet...)

The most recent Quirks newsletter had an interesting piece about mitigating response bias from socially acceptable responses. While the item count method presented there can reduce the stigma of having to actually check a box associated with any particular risky behavior, I think part of the trouble with surveying about socially unacceptable behavior is also about identity. The respondent doesn't even want to admit to themselves that they perform the activity.

This was particularly important in my previous work on bill payments. People don't want to identify as a person who doesn't pay their bills. Few people want to admit to paying bills late or not at all, however people were much more willing to complain about having to pay late fees. Sometimes little tweaks can help navigate around socially acceptable responses.

I would be curious if the authors have conducted any research-on-research to see what the difference in estimated prevalance is between a direct response and the item count method.

This is just to kick off this blog. Thanks for checking it out. Check back for the latest news on marketing research methods and research in the business world.