By Amy O’Hara, Research Professor in the Massive Data Institute and Executive Director of the Federal Statistical Research Data Center at the McCourt School for Public Policy
Like a booster shot of sanity, the Committee for National Statistics (CNSTAT) recently released the seventh edition of Principles and Practices for a Federal Statistical Agency. This book is full of good ideas, organized around five principles. Five?! The earlier editions of the purple book only had four! Yes, CNSTAT has added Continual Improvement and Innovation, a reminder that the best methods of the 1980s are not necessarily the best methods for the 2020s. If you need reminded of the other four principles, see below.
The other four principles are Relevance, Credibility, Trust, and Independence. The public servants at our nation’s statistical agencies live by these things. They maintain the highest standards in their numbers and publications despite political and budgetary headwinds. Their independence gives us reliable economic indicators, national reporting on crime, and Wonder-ful public health data. The newest principle comes at a great time, with the ongoing implementation of the Evidence Act and ambitious Action Plan in the Federal Data Strategy.
But wait, there’s more. Principles and Practices describes ten practices for federal statistical agencies that protect public data users. These include authority and procedures to protect independence (especially important since our mortality and population numbers were in jeopardy due to political interference), focusing on quality, meeting users’ needs, and being open about limitations of data produced.
This book is required reading even if you are not a federal statistical agency. This book is an incredible resource, it lists all the stat agencies, and has an appendix of pertinent legislation and regulations. Principle 2 is all about you, data user. Understanding what the federal statistical system aspires to can make you a more informed consumer of its data products.