May 27th, 2013
Research: Employers conned by bullies with political skills
Most research on workplace bullying uses bullied targets as participants. That is true of our online studies here at WBI. Those are the people who visit the website and can complete surveys. When we do national surveys, we contract with pollsters who can randomly sample Americans so we can draw conclusions about the entire population. Research on, or about, bullies is rare.
Now comes a clever new study (Treadway, et al. 2013 — see below for full citation) that investigates the relationship between bullies’ perceived political skill and perceptions of the bullies’ job performance.
Clever is how the researchers identified bullies. Genuine perpetrators certainly do not volunteer to be researched. So, employees in a mental health facility (only 69 eventually participated) checked who they considered to be a bully from a list of all employees. Workers were considered bullies if they were observed “engaging in unfair criticisms of co-workers, withholding work-related information, excluding someone from social interactions, and/or attempting to intimidate others” Using a mathematical formula, bullies were identified as well as their connections to other bullies in a network. Thus, each person had a “bullying” score of some sort based on the closeness to a bully’s network.
Everyone rated their own political skill on a questionnaire of 18 items with a maximum score of 7. The average score was 5. Skill was defined as “the ability to effectively understand others at work, and to use such knowledge to influence others to act in ways that enhance one’s personal and/or organizational objectives.” The definition closely matches the one for Machiavellianism or political intelligence, but the researchers chose to use “political skill.”
The measure of job performance was a score between 10 and 50 (average 30.67) provided by the employer that was based on performance evaluations.
Bullying (as scored in this very peculiar way in this study) was not correlated with political skill. Neither were political skill alone or bullying correlated with job performance.
The more important analysis was a stepwise regression model. Simply put, the task was to explain job performance scores. Initially, age and education explained (accounted for) 27% of a person’s performance rating. When bullying and political skill were added as single variables, they did not add significantly to knowing a person’s age and education. However, when researchers added the interaction (combined effect) of both bullying and political skill, an additional 9% was gained in predicting the job performance score. That 9% was statistically significant. And that is the payoff.
The major finding
The payoff in the study was the effect on perceived job performance by the interaction between political skill and bullying. Only bully-employees with a high level of political skill were evaluated as being high performers. In other words, bullies use their skill to “con” evaluators (their managers) into believing they are good performers.
Bullies with little political skill fared less well, being perceived as lower performers.
Shortcomings of the study
Political skills were scored by self-report. That is, people evaluated themselves. But the identification of bullies was done by others. It would have been better to match the uniqueness of the bully variable and also have people rate the political skills of others.
The bullying measure was actually an individual’s centrality with the bullying network, using a procedure called UCINET 6 to calculate an eigenvector index. Researchers ignored the gold standard of behavioral scales to define bullying — the NAQ (Negative Acts Quesionnaire by Einarsen and Hoel, available in a short version) and may have misrepresented the degree of “bullying.”
The failure to directly correlate bullying with political skills suggests problems with the operational definition of bullying. And the failure of bullying alone to significantly accounted for variance in job performance scores is explained by the choice of the weak measure of bullying. The regression analysis itself relied on only scores from 54 participants.
The researchers themselves cited the unique characteristics of a sample of employees from a mental health facility that might not extrapolate to a more general workplace.
The researchers tried to explain a rather simple notion — linkage between bullying and political skill. The study provided support for their hypothesis. However, it is clear they know too little about the workings of a real organization. They stated:
“This study marks the first attempt to measure the relationship between being a bully and job performance and offers an initial explanation of why bullies thrive in the workplace despite organizational attempts to sanction such behaviors.“
The academics make the assumption that because bullying is so negative and harmful that organizations are doing all they can to stop it. Oops. Better talk to targets for a dose of reality.
D.C. Treadway, B.A. Shaughnessy, J.W. Breland, Jl Yang & M. Reeves. Political skill and the job performance of bullies. Journal of Managerial Psychology, 2013, 28 (3), 273-289.
This entry was posted on Monday, May 27th, 2013 at 10:36 am and is filed under Bullying-Related Research, Social/Mgmt/Epid Sciences. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.