How we build and test our synthetic population, and what published research says about the surveys it is measured against.
We check our synthetic population against prediction markets where people bet real money. Follow one market from start to finish, then see the 18-axis map that shows how far to trust a new question before the event settles.
The science behind reading audience emotion from video: film reliably produces measurable feelings, those feelings follow cues on the screen and in the soundtrack, who is watching changes them, and a small panel's moment-by-moment response predicts a large audience. Then how TestScreen puts those findings into a system a buyer can check.
How the population is built from US Census and ACS microdata, calibrated to about 1 percent average error on the census margins, and checked against prediction markets and published Gallup and Pew results. It also sets out the known limits, and what a reviewer can inspect.
What published studies say about survey research: falling response rates, how respondents behave, how much bad data gets into panels, how well the standard quality checks work, and how the questionnaire itself shapes the answers.
We walk buyers through our sources, methods and validation record on every call.