If you have read my blog, you know that I love digging through data to find new insights and I’m a believer that choice questions (such as those used in Discrete Choice Conjoint or MaxDiff) are the best way to engage respondents and unlock what they are thinking. Given that, a book called “Data, a Love Story” should be a natural fit for me because it is about the ultimate choice…choosing the right person to marry. Ultimately I decided against buying the book (wasn’t sure my wife would see it as purely a curiosity). At the same time, the review I read made me realize that some of the issues the author faced, are the same as those we face as researchers.
The premise is that dating websites can be gamed to find the right mate. Having never used one (my marriage pre-dates them), I assumed that these sites use complex algorithms to match compatible people. The trouble is that while this is true, these algorithms can break down.
First off, many people are not honest in their profile. They might be looking for someone to sit around with and watch television but admitting that is tantamount to saying “I’m really lazy” so they fudge a bit. Some go beyond this and tell whoppers like “I’m not married”. Obviously any bad data will lead to bad matches.
Second, aligning profiles is only a first step…it determines which profiles an individual sees. At that point the individuals are free to contact each other or not. Thus, how that profile reads is more important than the questions that determine the “match”.
The author, Amy Webb, decided to gather her own data. After crunching the numbers she was able to both better attract invitations from the right men AND figure out which of them she should be talking to.
Of course, by law I have to tie this back to Market Research so here goes. It strikes me that we face the same issues as these dating web sites. We rely on respondents to tell us the truth, but sometimes they don’t. Some lie to qualify for surveys and get an incentive. Others, shade the truth because they don’t want to admit it. We bring some of this on ourselves. Can we blame respondents for tuning out on long surveys? Is it any wonder that we get bad data when we ask respondents questions they are not sure how to answer?
As we continue to advance the analytics and techniques we use in our business, it is critical that we remain vigilant in our efforts to engage respondents and make sure the data these techniques rely on is reliable itself.
Rich brings a passion for quantitative data and the use of choice to understand consumer behavior to his blog entries. His unique perspective has allowed him to muse on subjects as far afield as Dinosaurs and advanced technology with insight into what each can teach us about doing better research.