Leadership Assessments are Subject to Bias

As part of leadership, it is important to identify and assess the next generation of leaders.  In large organisations, it is not always possible to simply identify them through direct interaction, or a limited mentorship programme.  As such, organisations are moving towards tools such as leadership surveys.  These surveys are a preferred tool to determine whether a person is a potential leader, and whether his values are congruent with the organisation. 

These surveys are generally put in the form of an assessment sent to leadership candidates, their managers, and their colleagues.  Some companies implement a 360-degree feedback system.  The assumption is that the data is accurate, and objective.  But this is not how we should identify leaders.  This is how we identify popular people.  People are biased, and there is a reason why eyewitness accounts in trials are known to be faulty,  In general, people are terrible judges of situations and character.  For example, people evaluate those who share their culture better.  Those who speak the lingo have an advantage.  Men are judged more favourably than women, even when they have obvious flaws in character.  Taller men are judged more favourably.  Attractive women tend to be disadvantaged.  Fairer people have an advantage.  This follows with political affiliation, religious association, gender identity, nationality, and so forth.  This system is fundamentally flawed. 

These surveys measure the perception of what people imagine a good leader is, not what a good leader actually is.  Because perceptions of good leadership are shaped by the media, these surveys end up looking at the wrong parameters to judge the qualities of good leadership.  This means that organisations utilising these feedback tools actually manage with data on what people perceive good leadership is, not the reality of it.  This means that ideas of good leadership get subverted, and effective leadership diminishes in these organisations.  Due to the influence of the media, narcissistic and sociopathic behaviour is perceived as good leadership.  Flamboyance and perception have taken the place of quiet efficiency and ingrained effectiveness. 

In recent years, studies in leadership have written enormous amounts of literature on this phenomenon, and devoted entire sections in journals on the question of identifying and quantifying leadership through various means of research.  Personally, I think the underlying assumption that the average person knows what good leadership is happens to be greatly exaggerated.  People know the fruits of good leadership; they do not know what a good leader is.  Good leaders are often only recognised by the fruits of their decisions and policies, and this could be long after they are gone. 

360-degree feedback is a tool to quantify the efficacy of management, not leadership.  Management is something apparent, and being the subject of management, people can tell the difference between effective management and ineffective management.  It is a question of values, and whether they felt enfranchised and empowered.  Leadership is more than that.  Leadership is about seeing beyond the horizon, and putting in place the foundations to meet those challenges then.  The average employee, or follower, is not privy to that level of decision-making and planning.  How is he going to make a judgement on that? 

Ultimately, judgements of leadership performance can only come from peers, and the people above.  Nurturing of such leadership performance comes through mentorship programmes to inculcate values, and create a learning environment.  That does not mean we cannot use these feedback tools to have a better understanding of a candidate leader’s performance.  We need to recognise the limitations of these tools, and refine how we use them. 

Firstly, people need to be educated on the role and nature of good leadership.  Based on this, we can have a conversation, and construct the ideal leader.  This ideal leader, by shaped consensus, is what we rate leaders against.  This makes the rating system more useful, and less arbitrary. 

Secondly, we need to drill down on specific rated behaviours and values, and look for qualitative and referenceable examples.  This makes the ratings more useful, and we can benchmark certain behaviours. 

Finally, the feedback should be constructed in such a manner that people are forced to slow down, and think.  If we create it like a multiple choice question, people will simply tick boxes at random and go through it as quickly as possible.  This renders the data useless.  On the other hand, we also want to avoid confirmation bias.  This comes back to the need for confirmation, and examples. 

When done properly, and paired with a mentorship system, we can use this data to help the organisation identify potential leaders, better judge identified leaders, and shape our engagement and development programme.  People, left to their own devices, do not know what is good for them due to inherent and media bias.  But they are still a useful means to understand effective application of leadership behaviour, and how to refine it.