Organizations would like to conduct their performance appraisals in an objective manner.  But, a certain bit of subjectivity and bias do seem unavoidable when the manager rates employees on their performance. One main factor for subjectivity in rating is due to difficulty in setting-up number driven goals for all employees and tracking and measuring the same.
The most common rating errors in a performance appraisal are mentioned below:
1)      Halo effect: When an employee performs well in a few areas of his work, then the general tendency of a manager is to rate him well even in un-related areas where his performance was mediocre.  This is called a “Halo” effect.  The opposite of Halo effect is called “Horns” effect in which a manager rates an employee poorly, when he spots a few areas where he has been not performing well.  In other words, in this type or rating error, the overall rating of an employee gets influenced by good or bad performance in a few areas of his/her work.
2)      Recency: Another common error in rating is the problem of recency.  Though performance appraisals are done to rate an employee’s performance through-out a year (or 6 month period etc.), managers tend to remember only the past few months performance, and rate the employee’s performance based on that.  So any performance highs or performance lows of an employee in the last few months, significantly influences the rating.
3)      Central tendency: Central tendency is again a very common rating error. Sometimes, managers with the intention of avoiding conflict, play it safe, by rating employees in the middle of the rating scale.  So it might be a “met expectation” for all performance parameters, irrespective of whether in reality they have not met it or not.
4)      Leniency / Severity: Sometimes, managers tend to rate employees leniently or severely. This may be due to factors like:
    Whether manager’s like an employee or not.  Personal bias or preferences creep in the rating. Their general style is either lenient or strict.
    They compare employee’s performance with their personal standards (instead of standards expected from the role).  This is also called as “Frame of reference” error. Sometimes when employees are similar to the manager, it influences the manager’s rating which will then be lenient.
So in this case the rating is either positively or negatively skewed.
5)      Contrast effect: This occurs when managers rate an employee by comparing him with other employees. So, instead of giving rating based on the standards required for the job, managers rate a person by comparing with other employees performance.  In such cases, a person may be given higher ratings just because he is better than others, but he may still lag behind when compared to the standards expected.
6)      Stereotyping: Stereotyping occurs when managers generalize about employee’s performance based on a group. Grouping can be age-wise, experience-wise, region-wise or university-based and so on.  For instance, managers may generalize or stereotype saying that all young employees do not take ownership of their work, or that freshers hired from a particular university have great technical skills and so on.  But actually, it is important that managers should consider individual differences
HR team can educate managers about these common rating errors, and ask them to be beware of them, when they do the performance rating.  Managers, who are well aware about the performance rating issues, will tend to approach it more objectively and ensure that bias does not creep in their rating.