For many people, the start of a new year represents a chance to wipe the slate clean, to start over, to do better, and not to repeat the mistakes made in the year (or years) just past. Inevitably, though, it doesn't take long for reality to bite, for idealism to give way to realism (or, worse, pessimism), and for even the most hopeful of us to acknowledge that our imperfect world won't be made perfect overnight. Statisticians, like many people, might wish for a perfect world. But where others might despair at the flaws they see, statisticians may be less perturbed than most. After all, as the astrophysicist Adam Frank wrote back in 2016, statistics is about “the imperfect world of imperfect equipment taking imperfect data” (n.pr/2snV6Fn). This is a world that Michael Wallace explores in detail in our cover feature on measurement error (pages 14–19). It is about the mistakes that can be made when trying to measure something, the consequences that can come from assuming our measurements are perfect when they are not, and how statistics can help bring us a little closer to the truth. Wallace's article leads nicely on to two other features this issue, both of which concern the difficulties of accurately measuring things. First, on pages 26–29, Jason Oke and Tom Fanshawe explain how biases introduced by diagnostic practices, disease definitions and screening protocols can affect our understanding of cancer survival rates, particularly when we look to compare rates in one country to those of another. Then, on pages 30–33, Joseph J. Salvo, Annette Jacoby and Arun Peter Lobo explore just some of the ways in which the accuracy of the decennial US census is affected by different forms of data collection and imputation. They argue that a high rate of self-response is key to a good count of the population. But even when measurements are the best they can be, they will likely still be far from perfect. To quote Adam Frank once again, data is only ever “a partial representation of the thing you are trying to understand”. Therefore, knowing what questions to ask about the limits of data, measurements, research and evidence is a crucial skill for surviving and thriving in an imperfect world. On pages 42–43, statistics education ambassador Christine Franklin talks about her drive to help society develop that skill and her goal of making statistical literacy a priority in education. So, here's hoping for a better-informed, more statistically literate world in 2020 – even if it is no less imperfect than it was in the year just gone! Our imperfect world won't be made perfect overnight Photo: Elyse Marks Imaging/RSS