Report from one of the sessions at the RSS 2017 Conference. More reports of conference sessions are listed here.
Mathematics professor Rebecca Goldin has called for statisticians to 'have some empathy' for journalists who write about statistics, acknowledging the many challenges they face in reporting on the subject.
Journalists are trained in language, she said, but statistical language often 'obscures rather than clarifies', especially when it assigns unique meanings to words in common usage, such as ‘significance’ or ‘error’.
Goldin, a George Mason University professor and director of STATS, was delivering the President’s Invited Lecture at the Royal Statistical Society Conference in Glasgow, during which she highlighted the ways in which media accounts of statistics can be both misleading and confusing: whether they use the wrong average, mistake correlation for causation, or struggle with the concept of confidence intervals.
But in her experience, she said, most journalists want to get the story right, and only a few will be wilfully misleading. She recommended that statisticians take the time to explain their work to journalists, but always follow-up a phone interview with a written explanation to ensure their points are properly understood.
'Isolate the takeaway and be very precise about what that is,' she said.
- Rebecca Goldin’s talk was titled The Media’s Love-Hate Relationship with Statistics: Challenges in Communication.