Forums
Forums » Topic Forums » General Forum » RATING SCALE For Performance Evals?
|
RATING SCALE For Performance Evals?
posted at 6/5/2012 6:25 PM EDT
on Workforce Management
|
Posts: 4
First: 2/13/2012 Last: 6/5/2012 |
We are completly overhauling our performance review system and in the process of assessing the competencies required for each position, job descriptions, interview questions, you name it. I am hoping to implement a robust scale which differentiates between weak, average and strong performers. There is a rating scale that I can't rememer the name of right now...instead of 1-2-3-4-5 it uses more of a spread like 0-2-5-8-11. Can anyone remember what this is called? Have you used it in your organization and how is it working out?
|
|
|
Re: RATING SCALE For Performance Evals?
posted at 6/12/2012 11:30 PM EDT
on Workforce Management
|
|
Posts: 1
First: 6/12/2012 Last: 6/12/2012 |
Hi, Heather: I can't help you with the name of the rating scale, but if you type "rating scales" into a search engine search box (I used Google), you will find an item which provides a number of examples of various rating scales (images of rating scales); possibly you will find what you are looking for. One thing would be to put a line with intervals marked, but which would allow the assessor to place a tick between indicated intervals. One problem with these sorts of scales is that it is up to the assessor to determine what characterizes good, average, and poor performance. You might want to consider a descriptive ranking which clarifies what characterizes each level of performance for key performance indicators. There are examples of such rating scales available in the "images or rating scales" noted above. Such ranking systems are called BARS (behavioral anchored ratings); you can find articles providing discussion and examples by entering "BARS rating systems" in a search engine search box. Hope this helps. |
|
|
Re: RATING SCALE For Performance Evals?
posted at 6/21/2012 2:06 PM EDT
on Workforce Management
|
Posts: 1
First: 6/21/2012 Last: 6/21/2012 |
Hi Heather, You're right. There are many different rating schemes companies come up with for performance appraisals, but few are as goofy as the example you cited. Here are the most important things you need to know about rating scales: q Most performance appraisal systems use either three, four, or five levels to rate performance. There are some systems that use a two-level (pass/fail) approach. There are some that use seven or nine (or even more) levels, and some that have no final rating. These are rare. q The majority of organizations in America, Europe and Asia use a five-level system. q Most managers believe that they can accurately discriminate among five levels of performance: 1 = truly unacceptable 2 = in need of improvement 3 = fully successful 4 = superior 5 = genuinely distinguished. q There are advantages and disadvantages for each alternative to the number of rating levels. However, a consistent problem is that appraisers rarely use all of the levels available to them, no matter how many are offered. q Appraisers will often add plusses (+) and minuses (-) to a rating in order to increase the level of granularity without actually moving to the next higher or lower rating. q The “2=Needs Improvement” rating typically does not allow appraisers to distinguish between those employees whose performance is below par because they are new or recently promoted, and those employees whose performance is marginal and should be reassigned or terminated. Providing an opt-out rating alternative like Learning or Insufficient Data or On Target can allow managers to accurately assess (and compensate) people who are not yet fully successful q Offering more than five rating levels actually tends to reduce the amount of differentiation rather than increase it. q If the final performance appraisal rating is determined by averaging individual item ratings and allows the use of a decimal point, then the final rating of almost everyone in the organization will be between 3.4 and 3.6. q No matter how many rating levels are provided, the lowest level is almost never used. Hope this helps,Dick Grote |
|
|
Re: RATING SCALE For Performance Evals?
posted at 6/21/2012 6:57 PM EDT
on Workforce Management
|
|
Posts: 174
First: 9/29/2011 Last: 2/12/2013 |
Dick makes good points. Because managers as a rule don't like confrontation, any 5 tier system is heavily skewed towards the top. "Outstanding" can often have 15% or more of the employees; the "above average" tends to be about 40 - 60%, "average around 25% and "needs improvement" less than 10%. As Dick notes, the lowest level is never used (hopefully because the terrible performers are gone long before it's evaluation time). There are a couple of ways around this to be sure. One is that you limit the number of employees who can be in the top 2 categories. Another, which I've used with success, is to give managers a target percentage of their employee's total salaries for performance increases. Provide them with a salary increase matrix for position in the grade range and performance. Managers will soon figure out the performance management distribution on their own. I have also used a 3 tier system and find it to be a vast improvement over the 5 tier. "needs improvement", "satisfactory" and "excellent" are about all you really need. You can certainly use the salary management approach to indirectly influence performance rating distribution as well. Cheers!
|
|
|
Re: RATING SCALE For Performance Evals?
posted at 6/29/2012 1:49 PM EDT
on Workforce Management
|
|
Posts: 1
First: 6/29/2012 Last: 6/29/2012 |
With regard to scaling, I agree with Dick Grote's points. In addition, I find the desire within an organization to change the scales is primarily due to a lack of trust in the raters and their ratings. This is usually not a problem caused by the use of a 5 point scale. Rather, it is because the scale is being used for two different purposes that compete. Consider whether the purpose for the evaluation is for determining (1) how well an employee's performance measures up to performance criteria (such as those provided as behavioral anchors on a BARS) or for (2) determining the rank order of performance of individuals within a business unit or group. The first purpose follows a criterion-referenced approach and is usually provided to employees with feedback about their positive performance and with needed improvements that are identified and discussed with planning for interventions. The second purpose listed above is usually used for determining distribution of pay raises. The difficulties come in when both purposes are to be fulfilled at the same time on the same form using the same ratings. When supervisors rate employees according to criteria and provide feedback, but also have to use those scores to determine pay raises, difficulties arise when distinctions are to be made between employees using a criterion-referenced scale. By definition, criterion-referenced scales compare a person against a standard, not a person against a person(norm-referenced). A norm-referenced approach determines the average performance of employees in a group and the standard variance around that average. If the same form that is criterion referenced is used for both purposes, the same score can yield different pay effects without specifying the method or reason for the difference, increasing perceptions of pay inequity. Depending on how the pay determination is made, a supervisor may artificially inflate everyone's ratings so as not to be so responsible for the pay differences. Or, a supervisor can artificially rate employees to facilitate the pay determination by creating a rank order amongst employees, likely resulting in some ratings that are not accurate descriptions of employee performance. This allows the supervisor to stay within budget and have a metric for making their decision, but it also delivers inaccurate performance feedback to an employee. This is not a good scenario and is counter to the purposes of performance evaluation. Here is a solution. Perform the first type of evaluation at a time (and hopefully more than once) during the year that is separate from the pay determination. I agree that a five point scale is not only typical, but advisable and sufficient for the first purpose. The feedback of the supervisor should be authentic in its evaluation of the employee's performance against the standards. A separate form or process at a separate time would be used to rank order employees within the department. This form could be created to provide categories of importance for determining the rank order. Alternatively, some employers determine pay using other tools (e.g., profit sharing, meeting development plan goals, statistical models). The point is, separate the pay determination from the evaluation of the employee. This will allow the scale to be used more accurately and for the results to be viewed more authentically. Eric Vincent |
Forums » Topic Forums » General Forum » RATING SCALE For Performance Evals?
