Opinion
Official statisticians must learn to communicate better
Summary
Upfront public engagement will go a long way in winning back some of the trust our system has lately lost
The UK’s Office of National Statistics (ONS) recently made headlines for sharply rewriting its economic history. It had earlier reported that the UK’s economy was 1.2% smaller than its pre-pandemic size by 2021-end. Its latest update suggests that the economy was in fact 0.6% larger by then. The ONS attributed the large revisions to detailed data from business surveys that was unavailable earlier.
The ONS’s official explanation was followed by a Financial Times article by the UK’s national statistician. Ian Diamond pointed out that early survey responses, collected in the middle of the pandemic, were far less optimistic compared to what later data revealed, leading to those revisions. Diamond acknowledged the need to improve the accuracy of early estimates. Importantly, he underlined the need for an independent review of the revisions process; “...we have invited the official watchdog, the Office for Statistics Regulation, to review our analysis of these revisions, how we deal with sharp discontinuities in economic performance and, equally, how they are communicated" he wrote. “This will show if there is anything we could do differently."
Diamond’s response showed what statistical leadership should look like, and how statisticians should communicate. The contrast with Indian authorities couldn’t be starker. Consider the recent debate on India’s gross domestic product (GDP) numbers, for instance. The debate was initiated by an economist unfamiliar with India’s national accounting procedures. A clear statement from the ministry of statistics and programme implementation (Mospi) could have ended the debate. Instead, officials from the finance ministry and the Economic Advisory Council to the Prime Minister (EAC-PM) waded into the debate, creating the impression that Mospi lacks answers to the questions being raised.
A lack of timely and clear communication has been a key weakness of India’s statistical system for a long time. When the GDP series was revised in 2015 (with 2011-12 as base year), it created a controversy because of Mospi’s opacity. Mospi was unwilling to subject the GDP revision process to an independent review or to open up the new corporate database that it had plugged into the national accounts system. Mospi’s failure to produce either a back-series or a detailed method note added fuel to that fire.
Three years later, an attempt by the National Statistical Commission (NSC) committee on real sector statistics to produce a back-series came under attack from government officials since it showed higher growth in the pre-2011 period (when the Congress-led regime was in power). Eventually when the ‘official’ back-series was produced in 2018, it downgraded growth in the pre-2011 period. The fact that it was the Niti Aayog chief rather than the chief statistician who released the figures at a press conference ensured that the back-series remains contentious.
The chief statistician’s silence on that occasion was in marked contrast with his public disavowal of National Sample Survey (NSS) findings on toilets in 2019. The NSS report was based on a 2017-18 survey that showed many rural households lack toilets. It appeared soon after India’s sanitation ministry had declared the country “open defecation free," seemingly embarrassing the government. The chief statistician co-authored an article with the sanitation ministry secretary discrediting the NSS findings, causing great discomfiture among rank-and-file statisticians.
India’s current chief statistician has chosen to avoid the media altogether. Since he took charge in 2021, he hasn’t given a single interview to any media outlet. Other senior Mospi officials are also reluctant to speak on the record. In a recent conference on national accounts statistics, reporters were warned not to identify any Mospi officer in their reports. When an EAC-PM member raised questions about NSS population projections recently, it was a statistical advisor (and a former chief statistician) who posted a quick rejoinder. It turned out to be incorrect, and stoked further controversy. Mospi officials chose not to intervene.
Mospi’s silence is a defeatist response to the criticisms it has faced in recent years. A confident statistical leadership would talk openly about the strengths and failings of the system. It would communicate the errors and uncertainties involved and the steps being taken to minimize them. The lack of a proactive communication strategy has made India’s statistical system prone to periodic controversies. Relying on statistical advisors to clarify issues also has its limits. Senior Mospi officials need to speak out on their own. Ultimately the buck stops with Mospi’s top brass. They are accountable to Parliament and to the public at large. They need to explain their own work.
How official statisticians engage with the public plays a major role in shaping trust in official statistics in all parts of the world today. As a 2017 research paper by Per Nymand-Anderson of the European Central Bank pointed out, communicating statistics has become a fundamental responsibility of official statistical organizations. Effective communication can help crowd out statistical misinformation and low-quality statistics. In the absence of such communication, trust in official statistics could be eroded, Nymand-Anderson argued.
The bottom line: A statistician who is unable to explain his or her work clearly is unlikely to be taken seriously.
