Whether statistics are created in-house or sourced from the internet, their validity seems dubious. Can leaders actually trust the stats they present?
Why does HR needs statistics?
HR is a human-centric function. Top performing functions are no longer merely bureacratic, considering, instead, how employees are impacted by their business. Yet, each worker is different and requires a different approach. So, in a large company, how can HR ensure that it’s doing its best for the majority of staff? The answer for many is to rely on statistical analysis. By reading research, requiring staff to answer surveys and from taking polls, HR can formulate as near to an overarching picture of their workforce’s needs as possible, allowing it to make truly informed and evidence-based decisions.
This isn’t the only power of statistics for the HR community; as a function that requires regular financial investment yet doesn’t actively create incoming profits like sales, HR can often find itself struggling to justify its spending. This, again, puts the onus on statistics to evidence exactly how the money is invested and the long-term benefits of its initiatives. As Katie Arnott, Head of People at Harris + Hoole, told HR Grapevine, statistics help her to evidence the need for investment in HR to the company’s S-suite. She says: “Unless I can evidence what a difference something is going to make with data and statistics, we struggle to justify what we’re doing to our leaders. So yes, I believe in statistics strongly.”
The whole of social science is riddled with these inaccuracies
Should HR trust the stats it has?
To follow Arnott’s thinking, statistics are undoubtedly important to the HR function, but how can they be sure that these numbers are accurate and reliable? In most cases in which HR would be privy to statistics, is there any evidence, bar pure trust, that these numbers are correct and dependable? Many people – such as former British Prime Minister Benjamin Disraeli, who was famously quoted stating: “There are three kinds of lies: lies, damn lies and statistics” – believe that they are, at the very least, extremely questionable.
How can we prove this? Ironically, with research and statistics. A 2009 investigative survey by Dr. Daniele Fanelli from The University of Edinburgh found that 33.7% of scientists surveyed admitted to questionable research practices, including modifying results to improve outcomes, subjective data interpretation, and the withholding of analytical details and dropping observations because of gut feelings.
And whilst at least some of those who are supplying such statistics admit to falsely reporting their research, there’s also the question of complexity; whilst an accurately sourced statistic may be technically correct, it may well still display an eschewed view of the data. The example given by Datapine is this: “Drinking tea increases diabetes by 50%, and baldness raises the cardiovascular disease risk up to 70%. Did we forget to mention the amount of sugar put in the tea, or the fact that baldness and old age are related – just like cardiovascular disease risks and old age?” So, it’s evident that even when statistics are accurate, context is essential.
How can HR trust the stats?
Whilst there will inevitably be many questionable sources of data on the internet, there are also notable names that are considered to be reliable by the general HR community. For example, names such as Deloitte, the Office for National Statistics, Glassdoor and O.C Tanner are notable names that feature regularly in HR reports. To find out more about the validity of the statistics that are regarded as reliable, Ed Humpherson, Director General for Regulation of the Office for Statistics Regulation, explains the following.
He says: “There’s a really simple thing that people should always have in their mind when they hear a statistic like ‘81% of the public wear blue shoes on a Wednesday’, is that they need to go behind the number; go the extra quick. Think about who is providing that information and what their vested interest is in it. If that statistic was created by the ‘Blue Shoes Manufacturing Association, you might want to not put your confidence in that.”
I do think that often you can get a direction of travel, a sense of where the wind is blowing, but I certainly don’t think you can be precise
What’s the solution?
It’s obvious that statistics have their place in seeking direction, yet the majority of statistics that make their way into presentations aren’t truly relied upon; this is the opinion of ex-Twitter VP EMEA Bruce Daisley, who tells HR Grapevine about the research he’s conducted in this area. He says: “If you look at the whole realm of scientific papers, there’s something called the replication crisis. This is where quite often, someone will publish a paper, and then someone will replicate their methodology and won’t be able to achieve the same results. So, the whole of social science is riddled with these inaccuracies,” he states.
But this doesn’t mean that Daisley doesn’t utilise stats when making strategic decisions. His ultimate conclusion – and perhaps one that all HR leaders should follow – is that whilst they may not be ideal for precise reporting, statistics do have value. “I do think that often you can get a direction of travel, a sense of where the wind is blowing, but I certainly don’t think you can be precise,” he concludes.
Humpherson agrees, advising: “My action plan for those seeking accurate statistics would be to always be willing to go the extra quick and do some research into the numbers. Always take note of who created them, why they created them, and how they were created before you trust in them. The second thing you can do is to look for the national statistics designation, or a clear statement by the producer about how the number was ascertained – and the third thing is that no statistic is just a number, they all have context, so think about that.”