UCAR > Communications > Staff Notes > October 1996 Search

Delphi Questions: Performance/salary changes, site disruptions

Question #424 (28 August): I'm not very happy with the way raises are handled. It used to be that I was given a rating (1-4) by my supervisor and then my raise was based upon that and the available funds in the raise pool. Nowadays, I no longer get a rating, just a "pass/fail." And then my raise appears to be based primarily on compa-ratios.

I find it strange that we now have a system where employees are not given quantitative feedback on their performance and raises are not based on merit, but on some ratio that they have no control over. The level of performance seems to be a nonfactor in this new scheme.

I feel this system does not provide a good way to motivate or reward employees. Why can't we go back to the previous system, where employees got a rating and raises were based on merit and available funds?

Answer (11 September): There are two issues involved in this question: performance appraisal and salary increases. Obviously, the two issues are intertwined. I will try to address each issue separately and then explain how the two work together.

First, let's look at the reasons that UCAR, or other institutions, appraise employees' performance. This is a time-consuming process that few people enjoy. So why do we go through it? Many organizations have begun to reexamine the goals of performance management systems. Historically, performance appraisals have been used primarily to determine annual pay increases. Broader goals such as employee development, improving performance, team building, or providing effective feedback to employees are probably more valuable in an effective performance management system than just determining salary increases.

When the primary goal of an appraisal system is to determine annual increases, it is common to see multilevel quantitative rating scales. The four-level rating in our previous appraisal system is fairly typical, although I have seen rating scales with seven or more levels.

A general problem with multiple levels is the need to justify the difference between each level. Objective measures and ratings can be very difficult to achieve in any appraisal system, and in institutions with groups of high-performing employees, the separation between various levels of performance may be artificial. Thus the trend is to reduce, or eliminate, the number of rating levels in appraisal systems. If the primary goal of an evaluation system is not to determine pay increases, then the use of multilevel rating scales may be unnecessary or even counterproductive.

Our performance management system was modified a couple of years ago to focus more on coaching and developmental planning. When evaluating each employee's performance of key activities, supervisors identify suggested changes or improvements. Goals and actions for the upcoming year are developed to encourage the employee's continued professional growth. Other than a determination that performance meets job requirements, no overall rating is necessary to accomplish these goals; therefore, there is no need for multiple levels.

As to pay determination, UCAR is committed to recognizing sustained performance in annual salary increases. While there is no quantitative scale, the performance appraisal provides detail on each employee's performance. Directors and managers establish increases based on employee performance and compa-ratio. Linking pay to performance seems to make some sense, so why do we "complicate" things by introducing compa-ratio?

The compa-ratio is simply the ratio between an employee's pay and the midpoint of the salary range for the employee's job. The midpoint is tied to the median paid by similar employers. It reflects the pay for fully competent employees doing comparable jobs.

Compa-ratios indicate where the employee is paid within our pay range (and, on a global basis, where they are paid within their fields). In an ideal pay-for-performance system, higher-performing employees would move to higher compa-ratios and lower-performing employees to lower compa-ratios. By reviewing compa-ratios, a manager can determine the relative pay level of employees in different salary ranges.

UCAR, like all other institutions, must work with limited resources in determining pay increases. For the June 1996 increases, we worked with a 3.7% salary increase budget. This is a fairly competitive number--the range for most other institutions for the same period was 3.5% to 4.0%. When determining increases, managers consider relative performance levels, current compa-ratios, and the available budget.

For employees with similar performance levels, the manager will allocate greater increases to those with the lower compa-ratios to bring them closer in line with those with higher compa-ratios. Conversely, for employees with similar compa-ratios, relative performance will be the overriding factor in determining salary increases.

In the past year we completed our job evaluation program. This program realigned many of our jobs to the market and changed the compa-ratio for many employees. This caused some higher-performing employees to have lower-than-normal compa-ratios and some lower performers to have higher compa-ratios. As a result, compa-ratios had a greater impact on this year's salary increases than would be normal. While compa-ratios will always be important in determining salary increases, I would expect to see the balance shift toward performance as our system matures.

Does this system provide a way to motivate employees? It is arguable that no pay-increase system really motivates employees. People are motivated by many factors, most of them internal. Our system is designed to recognize sustained individual performance. Over time, higher performers will move to higher pay levels as measured by compa-ratios.

Can we go back to the old system? We don't plan to. Is the new system perfect? No, but there is no perfect performance or pay system. We believe it is a way to meet the goals of the performance appraisal process and recognize sustained performance with differentiation in pay.

Bob Roesch
UCAR compensation and benefits manager

Question #425 (28 August): I would like to request that employees be notified when activities that may be disruptive are scheduled at NCAR/UCAR. For example, in July there were actors and a filmmaking crew at the Mesa Lab for several days. It was disruptive not knowing what was going on and what all the commotion was. Also, the parking lot was filled with cars and semitrucks for crew and equipment.

As another example, I was working at ML on a Saturday (3 August), and when I left to go home, I found myself in the middle of some kind of biking event. I had to wait ten minutes for some people controlling traffic to let me drive down the hill.

I am not necessarily against such activities at NCAR. All I'm asking is that you let employees know ahead of time. That way we can know what is going on, and perhaps adjust our schedules a bit to avoid disruption. (Note: if the above events were announced and I just missed it, then it is my fault, and I apologize!)

Answer (19 September): In most cases, UCAR/NCAR management makes every attempt to notify staff of events occurring at the Mesa and Foothills Labs that may be disruptive. Joan Chiszar (Facilities Support Services) distributes a weekly copy of the ML and FL conference room schedule that lists events Monday through Sunday. Dale Kellogg (NCAR Director's Office) prepares a calendar of institutionwide activities, including management meetings, outside scientific meetings, Human Resources events, blood drives, parties, and the like. This calendar is sent to each NCAR division and UCAR program. Staff Notes Monthly and the announcements and calendar sections of This Week at UCAR are two other vehicles of communication that alert staff to upcoming events. We hesitate to send out announcements to all local users about every event that takes place, as we think staff would find this intrusive.

Since I coordinated the film crew in July I can respond directly to your concerns about that event. First, the crew was not here for several days, but only one day for approximately ten hours. We had three celebrities on site and we did not want fans to make a special trip to NCAR to catch a glimpse of their favorite stars, so we deliberately did not publicize the filming so as to minimize the disruption to the staff as well as to the film crew. We asked that all the crew's trucks and cars be parked at the end of the parking lot so that they would not disrupt our staff. However, we didn't move them as far down as we could have and we did, in fact, have a slight parking problem. We've noted this and will make every possible attempt not to have the problem repeated.

As for the bike race, it would be wonderful to have a calendar that would include events similar to this in a list of what's going on at ML and FL so that staff can see everything on a daily basis. We're certainly moving in that direction. This Week's calendar is now Web-based, and the room reservation system is in the process of being transferred to the Web. The institutional calendar also may move to the Web. If you know of an event that may be of general interest to staff or may cause staff disruption, you may add it to the announcements section of This Week by sending an announcement to thisweek@ucar.edu or contacting Jacque Marshall, ext. 8616, jacque@ucar.edu. Seminars, meetings, and the like may be entered into the weekly calendar through the following Web address:

Nita Razo
NCAR Visual Communications manager

Whither Delphi?

To judge by the dearth of Delphi questions and answers in Staff Notes Monthly over the past year, you might conclude that we've stopped asking questions of the powers that be. Not so. The service, established in 1975 to allow staff to anonymously submit questions and suggestions to management, is alive and well. In fact, 14 questions have been submitted over the last 21 months, says Delphi coordinator Rene Munoz. However, except for the two in this issue and two previously published in This Week at UCAR, none of these were deemed of general interest.

Rene forwards for publication any Delphi Q&As that appear to be of sufficiently broad interest to staff. "Since I became Delphi coordinator in January 1995, many of the questions have been of a much more specific nature than in the past. In return, many respondents have provided very frank and forthcoming answers. That's what the confidentiality option allows--private, two-way communications that would not work if questions and answers were always published."

Rene is now working on some new angles for publicizing and archiving the Delphi process. Stay tuned for details. In the meantime, we'll continue to publish questions and answers of general interest in Staff Notes Monthly. Any staff may ask a Delphi question and submit it in confidence to Rene (ext. 1171, ML room 135) in written form; it must be signed. Rene asks that you put your question in a sealed envelope marked "personal" should you use interoffice mail. Detailed procedures are given in the UCAR Policies and Procedures Manual, section 1-1-13. •BH

In this issue

Other issues of Staff Notes


Prepared by Jacque Marshall, jacque@ucar.edu, 303-497-8616