SOLD BY: From a positive perspective, it should be reminded that there is a great deal of good statistics out there. They can be seen as one of the most valuable assets that modern journalism might have to proactively approach events and issues of the day. A basic competence in statistical reasoning would enable and empower journalists to harness the power of good data – namely their capacity to summarise patterns, to depict and predict trends and to identify possible causes and effects – for news purposes. Used wisely and properly, good data allow journalists, among other things, to overcome the sometimes tyrannical rule of intuition, to link anecdotal evidence to the big picture, to bust myths, to challenge prejudices, to connect seemingly distance incidents/events and to turn ostensibly boring, soulless numbers into vivid, meaningful representation of the world.
And if they manage to go beyond basic statistical reasoning to be able to generate their own data, journalists will have wide-open opportunities to produce deep-digging,
Statistics, without exaggeration, are part of the fabric of the contemporary world. They are a pervasive feature of daily life, shaping the way we think and behave in ways that many of us no longer realise. Almost every key aspect of modern life – the quality of the air we breathe, the severity of a medical condition we have, the performance record of a soccer player we like, the potential price of the house we sell, the condition of the social care system we use, the health of the economy in which we operate, the national leader we want to vote, so on and so on – is statistically measured in one way or another. As such, statistics have long been a staple of daily news – from ‘hard news’ about politics, economics, business, finance, science and education to ‘softer’ categories such as health, crime, sports and entertainment, community or other areas of social life. And it is likely to become even more so in the future, when the ‘big data’ society is gradually normalised.
Yet when it comes to the quality of such news reporting, the media often ‘get a bad press’ – to the extent that some scientists and experts have come to assume that journalists seldom get numbers right (Utts, 2010). Meanwhile, many journalists, suffering from a ‘blind spot’ for numbers, tend to dismiss statistics altogether.
STATISTICS, NOT MATHEMATICS
First, it is not hard to make use of statistics for news purposes. Journalists’ traditional ‘number phobia’ is not because of statistics per se, but because their nature is either vastly misunderstood or too narrowly understood. The job of handling numbers for the news is often wrongly perceived as that of measuring, calculating and analyzing things with eye-numbing formulae. Statistics and mathematics are two different things: It is not necessary to be adept at mathematics to be able to use statistics effectively.
However frightening they might look, statistical analyses are about the application of valid reasoning, not calculation. Mistakes are often made in the news, but few involve getting the math wrong: Most are due to flaws in the logic applied to data and their context (Moore and McCabe, 2003).
In other words, the journalist’s job is not to learn and use eye-numbing formulae or calculate complicated things. For the most part, journalists deal with pre-processed data packages from other sources and what they need is a permanent determination to question data and a basic level of statistical reasoning. ‘You don’t need to be a nerd to improve your reporting of news with numbers’, says Deborah Potter (2009). ‘You just need to remember one basic, journalistic question: does this make sense?’ Some basic knowledge of statistics is essential but what journalists need the most is not a set of skills to calculate or create their own data but one to use logical, valid reasoning and journalistic scepticism to (a) find and acquire data, (b) explore and evaluate their real meaning in context, (c) investigate non-numerical factors shaping them and (d) report them in a balanced, fair, accurate, accessible and engaging manner. All this does not require any special math skills. If one can add, subtract, divide and multiply, he/she can learn to handle statistics for the news, as long as he or she is willing to apply to data the same probing and enquiring mind that is essential for any other newswork.
Few journalists, for example, would need to know the scary look of the formula for a correlation co-efficient: what they need is to understand what those numbers between −1 and +1 mean in practice (what can be considered a strong, moderate or weak positive/negative association) and, more importantly, to understand that such association is not necessarily a cause–effect relationship. Even in computer-assisted reporting and data journalism, where journalists do perform software tasks to create their own data, data production skills are not necessarily the most demanding part of the process.
The central task is still to find and assess existing raw datasets before using them, and then to apply logics and reasoning to both software-based processes and their resulting computer output. Software skills can be acquired quickly – for example, some free data visualisation apps on the Internet today require users no more than a few days to learn. But these skills are meaningless – and could be dangerous – if the ability and habit to question and evaluate data are absent.
If journalists have the ability to handle these metrics, they can harness them into an excellent, unique tool for more pertinent, more engaging and perhaps more viable news products. But if they do not have the statistical reasoning skills to understand and use web metrics wisely and calmly, journalists might professional judgement, unless. In this new ‘click-thinking’ culture, web metrics could easily deepen one of journalism’s already severe crisis – the dumbing down of news – and bring newsroom tensions and conflicts to a new height (see Nguyen, 2013, for a review of these issues).
For another thing, which is of exclusive focus in the rest of this article, journalism is operating in an increasingly chaotic world of ‘lies, damn lies and statistics’. Statistics, of course, do not lie. They cannot. It is those using statistics that lie, intentionally or unintentionally. Numbers are neither as neutral nor as objective as they look: they are a human invention to describe and represent the world out there and thus are subject to human reasoning and human capacity. In other words, the real problem is not numbers per se but the way people produce, use and assign meanings to them. Statistics-based lies, whenever they occur, are often because data are inappropriately produced or improperly interpreted for all sorts of benign or malicious, objective or subjective reasons. And this is why good news reporting of statistics is badly needed.
People tend to place more faith in numbers than in words, and the way they are presented in the news plays a crucial role in reinforcing or challenging such faith. Without journalists’ help to understand data, as Gigerenzer et al. (2007) said, ‘the public is susceptible to political and commercial manipulation of their anxieties and hopes, which undermines the goals of informed consent and shared decision making’.
Such manipulation is becoming more sophisticated than ever. Today, realising the deep and powerful penetration of statistics into the way people think, believe and behave, all major social, economic and political institutions have integrated numbers as a central part of their public communication – including ‘news management’ – strategies. Often, they mobilize complex, resourceful public-relations machines to pump into the newsroom all sorts of data that work to their advantage. One result of this is a deluge of ‘bad statistics’ out there, that is, those that are derived from a deliberate manipulation of the data collection and/or analysis processes.
In some cases, they are from ‘research projects’ in which the ‘researcher’ knows the conclusion before he or she starts. In others, it is about manipulating and ‘massaging’ data to advance some interests at the expense of their rigour. In some more shocking cases, the data arriving at the news desk are completely fabricated and do not exist. It is in part because of this chaos that the UK parliament passed the Statistics and Registration Service Act in 2007, giving birth to the UK Statistics Authority, an independent quasi governmental body that oversees the ONS and scrutinize all official statistics produced in the country.
And it is not just serious data that constantly seek the limelight: seemingly silly or bizarre things do, too. Consider, for instance, the following topics that Kevin Peachey (2010), a BBC consumer-affairs reporter, compiled from press releases:
• An average British woman walks the same distance as that between London and Hull as a result of shopping every year, according to research by one of the UK’s largest retail chains.
• Liverpudlian women have the largest breasts in the United Kingdom, the same retailer announces on another occasion.
• Batman is the superhero boss that most UK employees would like to have, says a recent campaign to raise awareness of employees’ hidden potential.
REFERENCES
Dunwoody S and Griffin RJ (2013) Statistical reasoning in journalism education. Science Communication 35(4): 528–538.
Esposito JL and Mogahed D (2007) Who Speaks for Islam? What a Billion Muslims Really Think. New York: Gallup Press.
Geier K (2012): On the importance of statistical literacy. Washington Monthly, 12 May. Available at: http://www.washingtonmonthly.com/political-animal-a/2012_05/on_the_importance_of_statistic037307.php (accessed 10 March 2013).
No comments:
Post a Comment