One of the devilish problems with statistics in market research is that we deal with probabilities rather than simple yes/no answers. But you, as a business person, need to make clear-cut decisions: Should you launch the product, yes or no? Which one name should you choose for your new service? Which three areas should you…
Two paragraphs can be exceptionally thought provoking when they get right to the point. In a simple introduction to the New York Times Book Review some time ago, the editors highlighted two ideas about statistics and stories worth sharing.
Election season polling always highlights the danger of bad pollsters giving the market research and polling industry a bad name. While there are lots of reasons that polls might predict outcomes incorrectly (polling is far more difficult and risky than many people think) there is no excuse for deliberately bad pollsters making up data to…
Making sense of data and statistics is not just a matter of being clear and precise, nor is it just a matter of keeping things simple (though all of these are important). To make numbers really compelling, we need to relate them to people’s experiences. If we want our audience to grasp the importance of…
In an article published some years ago entitled Data-Driven Journalism, Walker Sands, a Chicago-based marketing and PR agency, outlined two ways to get media placement for your company or your client with research. Their premise was that journalists and readers are hungry for interesting stories, and in today’s data-driven world, some of the most interesting…
Today I saw my first “dumb election statistics” article of the season, with AP News offering this dramatically titled story: Kamala Harris’ election would defy history. Just 1 sitting VP has been elected president since 1836. We are bound to see tons of headlines and stories like this during the next several weeks. It brought…
An article in The New Yorker claimed that a hundred years ago most Americans died in their mid-fifties. Alas, they were fooled by a statistic and the claim is not true.
The U.S. Census is one of the most useful tools for research and marketing because it tells you how many people fit specific demographic profiles and where they live. It started out as a simple count of how many people live in each state in order to allocate congressional seats. But it has since become a rich…
In basic stats class, all of us learned about the importance of random sampling. It provides the foundation for the iron-clad mathematics of estimation, statistical significance, and margins of error. But the problem in social science, market research, and opinion polling is that random sampling (sometimes referred to as probability sampling) almost never exists. So…
Customizing MaxDiff exercises by piping in text from previous survey answers might seem like a good idea, but analyst beware: if you are estimating individual-level scores using hierarchical Bayes or latent class analysis, you must split the data into subsets before you calculate the scores. Your results will be nonsense otherwise. Here is the scenario…