What percentage of people wash their hands after going to the restroom?
This is a difficult question to answer accurately. Surveys are unreliable because people are not likely to be truthful about behavior that is considered socially unacceptable. Surveys under-report the true percentage.
Even if it is difficult, estimating the percentage of people who engage in a specific illegal or frowned-upon behavior is critical as it has important uses in policy-making and enforcement. Research studies have shown some improvement in accuracy by asking participants to estimate the percentage themselves. People are more likely to be open about what others are doing than they are about themselves. Obviously this works best for populations that are well-known by the participants, such as employees in a company.
Survey researchers have come up with a randomized response technique (RRT) that encourages participants to be more open. The experimental design is as follows:
- Before answering a question, the participant throws a single dice which the researcher cannot see.
- The participants answers ‘yes’ if a 1 comes up and ‘no’ if it is a 6, regardless of the actual answer to the question.
- For all other numbers, the participant answers the question truthfully.
Because a ‘yes’ response does not necessarily mean the person actually engaged in the undesirable behavior, people are more honest. Admittedly, the forced yes and no means at least of 1/6 of the answers are incorrect but, despite introducing this statistical noise, the overall results give better answers.
Introducing error to increase accuracy seems counterintuitive but it works in multiple domains. As an example, comparing RRT results to drug screening on hair samples shows that it is 30% better than traditional techniques.
Now go wash your hands. I bet you forgot.
That’s fascinating Jonathan. The intersection of statistics with individual psychology has always been interesting to me. I had a course back in college (that’s WAY back btw), that explored “systematic influences on how we evaluate the world” and ways where we regularly misestimate numbers through the heuristics we use. A simple example is “availability” – where we overestimate the occurrence of some classes of things because the memory is more “available” to our memory than other things; and the classic example of that is the risk of airline travel. By an order of magnitude, more people die from car accidents than airline accidents, but when asked, a lot of people think they are roughly the same. It’s that every time there is a plane crash, it gets covered in the news extensively so the image of airline crashes is more “available” to people than the thousands of car crashes where people die every day, but are rarely covered in the news. (Great book on the topic that’s still in circulation is “Judgment Under Uncertainty: Heuristics and Biases” by Kahnmann and Tversky)
http://www.amazon.com/Judgment-under-Uncertainty-Heuristics-Biases/dp/0521284147
I had fun with my final project for the class by surveying people buying lottery tickets and asked them which possible outcomes were more likely on sample lotto cards. Some cards had numbers like 01-02-03-04-05-06 or the filled-in circles made a straight line across the card and others had 23-94-34-75-04-49. Most people thought the “random looking” numbers were more likely than numbers with any patterns. (A few gave me a smile and said “I know exactly what you’re doing and they are exactly the same.” Ha!)
Anyway, I love the “survey dice model”. The authors have come up with a clever and creative way of working around people’s built in bias – with just a small “cost” of introducing some statistical error in the results, but overall more accurate numbers getting to the bottom of what’s really happening. Very cool.
And in case you wondered, I always wash my hands. 😉
Jonathan, interesting post. I’m not so sure about the approach, although numbers rarely lie. We live in a world of people behavior, which is rarely predictable. I’ve seen people do the most amazing things, the most puzzling things, and the most awful things. Work would be easier of we could predict it… All we can do is try to influence it — which is why marketers and PR people have so much work to do. 🙂 I hope you are well my friend. Oh, and this EMT’s hands are clean…
Bill
The topic of washing hands was covered by the Freakonomics folk. They studied whether doctors washed their hands, and what could be done to improve the frequency. Here’s an excerpt: http://www.youtube.com/watch?v=AEkOmn5hjFU
Cool post, Jonathan. Thank you!
It made me re-read about research that was done in the 60s by S. L. Warner. He described the idea as “randomized response”, allowing respondents to respond to sensitive issues (such as criminal behavior or sexuality) while maintaining confidentiality.
Here is an example of how it works: Ask a man whether he had stolen something this month. Before he answers ask him to flip a coin. Instruct him to answer “yes” if the coin comes up tails, and truthfully, if it comes up heads. Only he knows whether his answer reflects the toss of the coin or his true experience. Half the people—or half the questionnaire population—who have not stolen something get tails and the other half get heads when they flip the coin. Therefore, half of those who have not stolen somthing will answer “yes” even though they have not done it. So whatever proportion of the group said “no”, the true number who did not steal something is double that. For example, if 45% of the population surveyed said “no”, then the true fraction that did not steal something is 90%.
Cheers
Peter
I should add, that I quoted Wikipedia above… What would we do without them 😉
http://en.wikipedia.org/wiki/Randomized_response
Great post and terrific comments
Even when respondents look to be “honest” there is often a cognitive bias where they under or over emphasize a positive outcome more than a negative one—in essence being truthful in their minds, while still not reflecting the situation accurately.
I have actually seen this play-out very well when ethnograhic research, where respondents are observed “in-action” and this input is used in conjunction with survey responses. I have had the benefit of “observing” consumers make purchase decisions and unknowingly evaluate products in ways that are in conflict with what they “believe” they do or value—a fascinating exercise into understanding consumer behavior and cognitive blind spots.
Oddly, I had the opportunity to work with a leader in ethnographic research who did studies around such issues as handwashing, per above, where actual behavior differed markedly not just from reported behavior, but to the poiint above, often over/under-stated from what was believed to be reality.
Beyond research we can take this into the area of social science via the unconscious mind—the current book, “the social animal” by David Brooks speaks to this unconscious layer that drives much of what occurs, despite what we think is occurring (or in our “best-selves” we’d like to have occur)
Hand washing posts are all the rage this week as I wrote one as well. The trouble is people first need an operational definition as to what is proper hand washing before they can answer the question.
As I wrote in my post http://riskczar.com/2011/10/19/rant-about-mobile-phones-and-e-coli/ I usually witness/hear men run their fingers under water for 3 seconds (no soap) then leave the washroom; in their minds *that* is washing.
If a proper definition is provided (e.g., wet hands, lather soap for as long as it takes to sing ‘happy birthday to you song’ and rinse) and then you ask people if they wash your hands, I submit about 5% of people really wash their hands. It’s quite gross.