The concept of utilizing big data to enable scientific discovery has generated tremendous excitement and investment from both private and public sectors over the past decade, and expectations continue to grow. Using big data analytics to identify complex patterns hidden inside volumes of data that have never been combined could accelerate the rate of scientific discovery and lead to the development of beneficial technologies and products. However, producing actionable scientific knowledge from such large, complex data sets requires statistical models that produce reliable inferences (NRC, 2013). Without careful consideration of the suitability of both available data and the statistical models applied, analysis of big data may result in misleading correlations and false discoveries, which can potentially undermine confidence in scientific research if the results are not reproducible. In June 2016 the National Academies of Sciences, Engineering, and Medicine convened a workshop to examine critical challenges and opportunities in performing scientific inference reliably when working with big data. Participants explored new methodologic developments that hold significant promise and potential research program areas for the future. This publication summarizes the presentations and discussions from the workshop.

Our ability to observe and forecast severe weather events has improved markedly over the past few decades. Forecasts of snow and ice storms, hurricanes and storm surge, extreme heat, and other severe weather events are made with greater accuracy, geographic specificity, and lead time to allow people and communities to take appropriate protective measures. Yet hazardous weather continues to cause loss of life and result in other preventable social costs. There is growing recognition that a host of social and behavioral factors affect how we prepare for, observe, predict, respond to, and are impacted by weather hazards. For example, an individual's response to a severe weather event may depend on their understanding of the forecast, prior experience with severe weather, concerns about their other family members or property, their capacity to take the recommended protective actions, and numerous other factors. Indeed, it is these factors that can determine whether or not a potential hazard becomes an actual disaster. Thus, it is essential to bring to bear expertise in the social and behavioral sciences (SBS)â€"including disciplines such as anthropology, communication, demography, economics, geography, political science, psychology, and sociologyâ€"to understand how people's knowledge, experiences, perceptions, and attitudes shape their responses to weather risks and to understand how human cognitive and social dynamics affect the forecast process itself. Integrating Social and Behavioral Sciences Within the Weather Enterprise explores and provides guidance on the challenges of integrating social and behavioral sciences within the weather enterprise. It assesses current SBS activities, describes the potential value of improved integration of SBS and barriers that impede this integration, develops a research agenda, and identifies infrastructural and institutional arrangements for successfully pursuing SBS-weather research and the transfer of relevant findings to operational settings.

Mathematics Research Center Symposium: Scientific Inference, Data Analysis, and Robustness focuses on the philosophy of statistical modeling, including model robust inference and analysis of data sets. The selection first elaborates on pivotal inference and the conditional view of robustness and some philosophies of inference and modeling, including ideas on modeling, significance testing, and scientific discovery. The book then ponders on parametric empirical Bayes confidence intervals, ecumenism in statistics, and frequency properties of Bayes rules. Discussions focus on consistency of Bayes rules, scientific method and the human brain, and statistical estimation and criticism. The book takes a look at the purposes and limitations of data analysis, likelihood, shape, and adaptive inference, statistical inference and measurement of entropy, and the robustness of a hierarchical model for multinomials and contingency tables. Topics include numerical results for contingency tables and robustness, multinomials, flattening constants, and mixed Dirichlet priors, entropy and likelihood, and test as measurement of entropy. The selection is a valuable reference for researchers interested in robust inference and analysis of data sets.

Logic and scientific inference; Probability; Sampling; Errors; Physical magnitudes; Mensuration; Newtonian dynamics; Light and relativity; Miscellaneous questions; Statistical mechanics and quantum theory.

Inductive inference and "experimental error"; A population sample model: local inference; Expansion of the model: inference in the large; Expansion of the model: inference in the large; Interpretation of results; Random variables and distributions; Variance and related topics; Problems of sampling in physical situations; Randomization; Restricted randomization and experimental designs; Regression or curve fitting.