Thanks to traditional and social media, an error in one document can echo outwards.

This article looks at how the erroneous claim that “4 in 10 dads will not see their children on Fathers’ Day” spread, and its corrections.

‘4 in 10 Fathers’

In the Fathers4Justice press release, the first sentence contains a misrepresentation:

A third of dads (37 per cent) will not see their children on Father’s Day according to a major new poll from ComRes which was commissioned by Fathers4Justice.

The question was:

Have you or someone you know experienced any of the following?

37% of the sample responded that they — or someone they know — had experienced “Not seeing my child(ren) on Father’s Day”. The survey estimated 21% of fathers had experienced not seeing their children.

About 1 in 5 fathers had not seen their children on some past Father’s Day. (Source: ComRes/Fathers4Justice)

Consequently, there are three errors seen in headlines and articles:

  • ‘4 in 10’: The 37% figure includes fathers who knew other fathers who had experienced that event — much higher than the 21% estimated to have experienced it directly themselves;
  • Experience, not expectation: The question was about experience of past Fathers’ Days: no claim can be made about the percentage of fathers who would not be seeing their children this Fathers’ Day;
  • Reason: There is no subsequent question to ascertain the reason for these misses, so it cannot be asserted that these misses were due to “separation or family breakdown”.

The Echoes

I contacted each of the four organisations directly. The Telegraph and the Mail Online were very swift in their responses. The Mirror and Sunday Express took longer, and meant I contacted the Independent Press Standards Organisation.

Knowing someone who has experienced something is different from experiencing it yourself. (Graph: R Pubs)

I am grateful for all of the corrections undertaken, and for IPSO handling my complaints.

This is the online correction from the Sunday Express. (Source: Sunday Express)

Also, I contacted ComRes. Market Research Society companies are required — by their Code of Conduct — to:

Members must take reasonable steps to ensure that findings from a project, published by themselves or in their employer’s name, are not incorrectly or misleadingly presented.

In my correspondence, I was assured that ComRes had sought to guide their client away from that headline, as it did not accurately portray their survey estimates. Unfortunately, Fathers4Justice proceeded against their recommendations.

It should be noted that the Fathers4Justice website contains multiple misrepresentations of official statistics. For instance, the ONS Families and Households 2012 report did not find that ‘1 in 3 children live without their father’. The proportion of dependent children living in lone parent families was estimated to be 23.8% in 2012.

For survey data, journalists should ask for the underlying data tables — which British Polling Council members are required by transparency rules to publish. Survey research companies should be wary of partisan campaign groups, and how their research is used by clients.

We should stand up for accuracy in statistics.

Note: the code for the article’s graph in ggplot2 is available on R Pubs.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store