Rater Errors

The biggest problem in evaluating employees is the ability of the rater to accurately perform evaluations (Peak et al, 2010). One of the largest failures of supervisors is documentation as many forget to write an officer’s actions down, whether they be bad or good. “No one incident (unless extreme) makes or breaks an evaluation, but it’s the patterns that are created by an officer’s conduct and performance that do” (Murgado, 2012, par. 5). I encourage my Sergeants to do two things to prevent failure to document. First, I suggest each keep a working diary of the good and bad that happens each day. Second, before the end of each evaluation period, I ask that each Sergeant have their officers give them a list of their training, accomplishments, …show more content…
The halo effect is where a supervisor rates an officer high or low on everything based on one dimension (Peak et al, 2010). A good example is if a Sergeant believes in a high number of arrests. Officers who make very few arrests will make low scores in all categories because of it. A regency problem is similar and occurs when a recent event affects an officer’s entire rating (Peak et al, 2010). For example, an officer arrests a serial bank robber and it is his only high risk arrest of the year, but receives high marks on everything because of it. Rater bias is simply when a rater gives high or low marks simply because of personal feelings (Peak et al, 2010). Constant error problem occurs when a rarer rates a group too low, too high, or too much in the middle without any clear logic (Peak et al, 2010). Last, unclear standards when there is no agreement across the board on rating standards and each supervisor may rate differently (Peak et al, …show more content…
The largest downfalls are the rater errors mentioned above. Another method is officer surveys as they “provide insight into the level of professionalism at an agency” (Davis, 2012, p. 4). However, these are one sided and the community’s view is lost. “Community surveys involve interviewing a random or representative sample of the community to gauge people’s opinions of police effectiveness or police misconduct” (Davis, 2012, p. 4). The problem with these are that they can be “influenced by respondent demographics, neighborhood context, and media coverage of the police” (Davis, 2012, p. 4). Contact surveys look at the interactions between police and citizens using calls for service lists or citations (Davis, 2012). These surveys do not ask global opinions, but ask questions about certain behaviors and research has shown that biases from the community surveys do not seem to be there (Davis, 2012). CompStat, a computerized tool for tracking crime, is another way to evaluate performance as it will show crime rates in different period (Peak et al, 2010). Last, agencies could simply use stats such as ticket, arrest, and accident numbers to show success of officers in the communities.