This article discusses common misinterpretations of statistical tests, P values, confidence intervals, and power. It highlights that these concepts are often misunderstood, leading to incorrect conclusions in scientific research. The authors emphasize that correct interpretation requires understanding the underlying assumptions of statistical models and the limitations of P values, which are not probabilities of hypotheses but rather measures of compatibility between data and model assumptions. They also explain that P values do not indicate the probability that the test hypothesis is true or false, nor do they indicate the size or importance of an effect. Instead, they provide a measure of how unusual the data are under the model assumptions. The article also discusses the importance of confidence intervals, which provide a range of plausible effect sizes, and the limitations of power calculations, which are not a measure of the compatibility of alternatives with the data. The authors conclude that statistical tests should not be the sole basis for inferences or decisions, and that estimation of effect sizes and uncertainty is often more important than statistical significance. They also warn against the misuse of P values as dichotomous indicators of significance and emphasize the need for proper reporting and interpretation of statistical results.This article discusses common misinterpretations of statistical tests, P values, confidence intervals, and power. It highlights that these concepts are often misunderstood, leading to incorrect conclusions in scientific research. The authors emphasize that correct interpretation requires understanding the underlying assumptions of statistical models and the limitations of P values, which are not probabilities of hypotheses but rather measures of compatibility between data and model assumptions. They also explain that P values do not indicate the probability that the test hypothesis is true or false, nor do they indicate the size or importance of an effect. Instead, they provide a measure of how unusual the data are under the model assumptions. The article also discusses the importance of confidence intervals, which provide a range of plausible effect sizes, and the limitations of power calculations, which are not a measure of the compatibility of alternatives with the data. The authors conclude that statistical tests should not be the sole basis for inferences or decisions, and that estimation of effect sizes and uncertainty is often more important than statistical significance. They also warn against the misuse of P values as dichotomous indicators of significance and emphasize the need for proper reporting and interpretation of statistical results.