Value of Employee Engagement Over-estimated By Lack of Understanding Of The Research
Why What You Believe About The Employee Engagement Research Is Wrong By Robert Bacal
One of the reasons that employee engagement efforts are not justified by their return, AND why companies are spending so much on them anyway (est. one billion dollars a year), is that the research is almost always misinterpreted, some times in subtle ways, sometimes in very extreme ways.
Not only are readers and decision makers not reading the source research, but they rely on journalists to provide summaries, but there is a lack of understanding of how to read studies, basic statistics, and even, sadly, logic.
In this section we’ll look at examples of research that is communicated in ways that make it appear that employee engagement is far more important than it actually is. Before we look at those examples, here are some things to keep in mind.
Three Things To Keep In Mind About Employee Engagement Research And It’s Interpretation
- Since employee engagement is such a “common sense” kind of concept, there is a strong tendency to apply a confirmatory bias to what we read. That means that when we come across research reports that seem to confirm our pre-existing thoughts and conclusions, we tend to believe and remember them. Readers need to guard against this tendency.
- Almost all employee engagement research involves correlations — this thing tends to appear (or not appear) with that thing. For example, it’s often suggested that companies with higher employee engagement scores are more profitable and productive. That is, there IS a correlation. However, whether high employee engagement CAUSES business success, or whether business success causes employee engagement is unknown. Usually, people interpret better performance to employee engagement, but that is almost certainly not the case. It’s probable that business success DOES make for better engagement, AND that BOTH are caused by things like superior management and strategy.
- There are so many correlative studies about that simply by chance, one will find positive correlations. For those statistically minded, if one thousand correlations are calculated, and we want to rule out chance at a 0.95 confidence level, we will find about FIFTY correlations that come up as statistically significant simply BY CHANCE.
- For the most part, when you read summaries of research, you are relying on the ability of the writer of that summary to understand the research methodology, and have sufficient skill to identify when the findings of the research need to be questioned. Journalists, and even those writing summaries of research for the companies that DO the research are not generally statistically competent, or sadly, even able to apply basic logic. So, if you read anything but the original studies, you are likely getting a non-critical interpretation of the data.
Examples of Misinterpreted Employee Engagement Research, and Critiques
Employee Engagement: Putting the Cart Before the Horse? | CustomerThink By Bob Thompson- Well written and intelligent critique of the employee engagement movement that includes the suggestion that business success causes employee engagement, and not the other way around.
Misinterpreted: Coaching Increases Communication, Employee Engagement, and Productivity, Study Finds By ASTD- Great example of how the industry overstates findings. The TITLE claims that coaching INCREASES engagement, but in fact, this study only uses survey data about what respondents “think” about coaching.
Must Read: The (Statistical) Confidence Game of Employee Engagement Surveys By Converge- A must read on employee engagement surveys, and the title says much of what is inside this article.
Positive Correlation: Employee Engagement vs Productivity By Helen Wallis- Yes, there are correlations between employee engagement and productivity (sort of), but this is the most common research mistake. Do EE scores cause productivity? Or is it that people in productive workplaces BECOME more engaged. Nobody likes to work in a loser organization.
Prime Example Of Mis-Stated Employee Engagement Research By Bruce Rayton- In this apparently reputable literature review, the authors immediately show their inability to understand research by implying causation — that higher engagement causes improvements in various other variables, as noted: “This paper sets out the evidence for the effectiveness of employee engagement in raising performance and productivity across the UK economy.”
Schindler lifts safety through employee engagement – Or Does It By Towers Watson- Download this Towers Watson Case study that looks at the CORRELATION between engagement and safety. BUT, it’s a correlation, and it’s just as likely that lower accident rates and fatalities actually cause higher engagement, not that engagement causes higher safety levels.
Why Employee Engagement? (These 28 Research Studies Prove the Benefits) By Kevin Krause- The title does NOT say it all. In this attempt to justify money spent on employee engagement, the author cites 28 studies that “prove” the value of employee engagement efforts. First, science doesn’t involve proving anything, so one has to question the credibility of the author. Second, most are correlative studies, that do NOT imply cause and effect. Third, where there appears to be a cause-effect relationship in “time”, there are no control groups used, so it’s just as likely that other variables caused improvements.