Skip to main content
SearchLoginLogin or Signup

The Future of Federal Statistics: World Statistics Day Webinar—20 OCTOBER 2020

An overview and reaction to perspectives voiced in 2020 World Statistics Day Webinar
Published onOct 28, 2021
The Future of Federal Statistics: World Statistics Day Webinar—20 OCTOBER 2020

1. Introduction

The World Statistics Day webinar on the future of federal statistics provided a vital and timely reminder of the critical importance of good-quality, trustworthy data to guide decisions of governments, business, and individuals. With a brilliant array of speakers and an engaging discussion, the webinar gave some profound insights into what we should be aiming to achieve with data and statistics and how we should proceed. It also highlighted why this issue matters to everyone: we should all care about the future of federal statistics wherever we live in the world.

2. Background

John Bailer, the president of the International Statistical Institute, reminded us, in his introduction as a webinar panelist, that statistics should be considered a human right. Bailer’s comments echoed my own that “without official statistics, democracy cannot function. How do we decide who to vote for, unless we can check out the claims that politicians are making about what is going to happen to the economy, or society, or the environment? How can we judge whether they have been successful? How can we look at whether their manifestos actually stack up? Official statistics are the lifeblood of democracy” (Pullinger, 2020).

Xiao-Li Meng, Editor-in Chief, Harvard Data Science Review, started the webinar discussion with reference to Teresa Sullivan’s 2020 article, “Coming to Our Census: How Social Statistics Underpin Our Democracy (And Republic).” Sullivan’s article highlighted that democracy requires numbers for its proper functioning and described the constitutionally mandated census as the keystone. She went on to discuss the implications of mistrust of the government and the lack of checks and balances needed to protect the integrity of the census.

Reference was also made to the Fundamental Principles of Official Statistics (United Nations, 2014). The need for a set of principles governing federal statistics became apparent at the end of the 1980s when countries in central Europe began to change from centrally planned to market-oriented democracies. The principles first developed in Europe were recognized as of more universal applicability and were ultimately discussed and endorsed by the United Nations General Assembly in 2014. At the webinar, Wesley Yung, Director of the Economic Analysis Division at Statistics Canada, reminded us that the first of the Fundamental Principles of Official Statistics positions official statistics as an indispensable element of a democratic society.

Developing the debate, Erica Goshen, Visiting Senior Scholar at the Cornell University School of Industrial and Labor Relations and former Commissioner of the U.S. Bureau of Labor Statistics, noted that high-quality, impartially compiled statistics are especially important in times of conflict and partisanship.

3. Who Does It Matter To?

This is not just an issue for governments, as Ola Awad, the president of the Palestinian Central Bureau of Statistics, argued. It is fundamental for the private sector, universities, and civil society too. Sir Ian Diamond, United Kingdom National Statistician, stressed this argument further, saying that it matters to every citizen in society, and as Rob Santos, American Statistical Association President-Elect said, statistics can represent those who have historically been excluded, especially the most vulnerable. Through statistics, their voice can be heard.

4. What Do They Need?

Throughout the webinar, a central theme was trust. Straightforwardly, Yung convincingly told us of the requirement for information that can be trusted. This question of trust was elaborated into the broader and emerging concept of good data stewardship. As well as being trustworthy, statistics must provide, as Goshen put it, information that can be used. At the heart of this, in Diamond’s words, is information that is well-communicated. To be useful, statistics need to be granular, fast, consistent, and relevant to the issues of the day.

5. What Does This Mean?

At its simplest, this means holding on to the Fundamental Principles of Official Statistics while embracing the new world of data and data science.

As Yung demonstrated, this means developing new methods to complement the traditional federal statistics toolkit. Awad gave us several vivid examples from the fields of tourism, prices, transport, and agriculture to help paint a picture of the opportunities for methodological development and the benefits that can be delivered from them.

As well as new methods, there was a call for utilization of new data sources. Awad spoke about the value of geospatial data and Yung mentioned crowd-sourced data and scanner data. With an open mind, there is a multitude of potentially valuable sources that can help deliver useful insights, often in ways that would not previously have been possible. As new technologies develop, we can expect even greater data potential in future.

A critical theme running through the webinar discussion was the importance of relationships. Santos noted that boundaries are breaking down between official statistics and the rest of the statistics discipline. Relationships with citizens and organizations concerned about privacy were essential to ensure a strong license to operate and an inclusive and sustainable statistical system. As Yung put it, there is no monopoly, and official statisticians should embrace the power of “and.”

6. How Do We Do It?

There is much valuable infrastructure on which we can build. The webinar highlighted several core elements of this:

  1. Legislation that safeguards independence and integrity. In some countries this might mean organizational change. Supporting this aspect of statistical infrastructure, Diamond noted that independence must not mean isolation.

  2. Leadership and skills. The debate included the needs both inside and outside the statistical system, especially statistical literacy and education of the public.

  3. Innovation. In many statistical institutions, delivery of the new agenda would require quite a radical mindset change

  4. Partnership and coordination. Even the best and best-resourced institutions could not succeed on their own. Mechanisms are needed to enable all potentially useful sources of data to be utilized and all necessary skills to be applied to the novel problems that will be encountered.

  5. Finance. Stronger advocacy is needed to demonstrate the value of federal statistics and convince those holding the purse strings that investment in federal statistics is more worthwhile than competing calls for scarce funds.

7. It Is Urgent

Effective responses to the COVID-19 pandemic and decisions about what happens post-COVID-19 require trustworthy and useful statistical data. Goshen showed us a clock ticking. She said that if we just try and muddle through there will be a downward spiral. Bailer quoted Tim Harford in exhorting us: “don’t take it for granted.”

Yung was clear that we cannot keep on doing what we have been doing and Awad encouraged us to think outside the box. Diamond concluded the webinar with words that struck a chord with all presenters: “let’s embrace this enormous opportunity with a strategy for federal statistics that is radical, ambitious, inclusive, and sustainable.”

Disclosure Statement

John Pullinger has no financial or non-financial disclosures to share for this article.


Pullinger, J. J. (2020). Statistics are a human right. Stats and Stories Episode 133.

Sullivan, T. A. (2020). Coming to our census: How social statistics underpin our democracy (and republic). Harvard Data Science Review, 2(1).

United Nations. (2014). Fundamental principles of official statistics. United Nations Statistics Division.

©2021 John Pullinger. This article is licensed under a Creative Commons Attribution (CC BY 4.0) International license, except where otherwise indicated with respect to particular material included in the article.

No comments here
Why not start the discussion?