One of the hardest things for state and local governments to do is evaluate employees. Though assessments can be used to help decide whether an employee gets a raise or not, this isn’t customary practice. Assessments are used to help managers and employees improve an agency’s performance and to assist in retaining productive employees. But for governments with so-called pay-for-performance systems, high marks on an evaluation can translate into giving an employee a bonus, a raise or a promotion.

Developing and using robust personnel performance measures can be critical. “Turnover is so high and training costs are so significant that there’s a big advantage to those who can select, train and retain productive employees,” says Michael Brink, senior director at the firm Huron Consulting Group. Unfortunately, once you start evaluating individual employees -- and particularly when those evaluations are tied to that person’s salary -- there’s always the possibility that the men and women whose future is tied to those assessments are going to think the system is unfair. As one state human resources official told us, the three biggest employee issues in her state are “skepticism, uncertainty and fear.”

This is a significant issue. If employees don’t think their evaluations are fair, they’re far less likely to be motivated by them. As in, “Nothing I do makes a difference in my evaluation -- so why try?”

It’s also critical that the managers involved in the process have faith in the evaluation system, since they rely on the assessment when they make promotion or compensation decisions. Often, though, managers find that the evaluation system they are required to use doesn’t meet the needs for the particular employees he or she is overseeing. “If you were a carpenter,” says Howard Risher, editor of Compensation & Benefits Review, “and someone designed a new hammer for you, but never asked what kind of hammer you needed, it would fail” to work for the task at hand.

Tennessee, which has had a history of troubled efforts in evaluating its employees, has been rolling out a new iteration of its system. Until the 1980s, Tennessee had so-called merit pay on its books, but the state discovered that managers weren’t using that system to achieve the intended results. “We heard stories about people who were going to the manager and saying, ‘I haven’t had a merit increase for a while,’ and the manager would just say ‘yes,’” recalls Danielle Barnes, general counsel of the Tennessee Department of Human Resources (DOHR). As a result of this kind of misuse and abuse, the state stopped funding its merit pay program in the mid-1980s. Little progress was made in rebooting the system until 2012. That’s when one of the governor’s initiatives, the Tennessee Excellence Accountability and Management Act, surfaced. A primary part of this new effort is to avoid grading employees on a one-size-fits-all checklist for all agencies. When DOHR started to implement the new initiative, “people wanted us to give them a book with all of their goals in them,” Barnes says. DOHR’s answer to that request, she says, was, “You all have to come up with your own because we don’t know your business.”

That sounds like a sensible start. Consider the difference, say, between a prison guard and a nurse. The kind of “bedside manner” required in one job is dramatically different than what is required in the other profession.

One of the struggles Tennessee continues to face is dealing with people who meet the listed criteria but have a bad attitude. That attitude can easily drag down the capacity of co-workers to meet their targets. “That can be a big issue,” Barnes says. “They’re ignoring our core values.”

But, she notes, the state wasn’t even measuring core values at the time. Now, it’s focused on making sure that agencies inform their employees -- through counseling and coaching -- what the core values are. But the state still hasn’t figured out a way to follow up on the issue.

There have been signs of a significant shift in the way that states and cities are setting up criteria to evaluate their employees. It’s almost a reversal of current thinking. That is, for a number of years now, program measurement has moved in the direction of relying on outcomes or results. The idea of looking at the amount of work done as a primary measure has been critiqued and discarded by many as a secondary effort.

However, since it can be enormously difficult to gauge the quality of an individual employee’s work, the trend in a growing number of places has been to go back to the old-fashioned way and look at quantitative indicators. In explaining the reasoning behind this approach, James Honchar, deputy secretary for human resources in Pennsylvania, says that “simplicity for me would be those standards that insulate the employees from any subjectivity of the evaluator.” The standard, he says “has to be objective. If an employee has to produce five widgets a day, he either produces them or he doesn’t. Employees really buy into it if they are being measured by something that’s quantitative and real.”

That doesn’t mean Honchar and others who follow this logic discard the notion that employees must do quality as well as quantity work. But when it comes to public-sector employees in the real world, measuring quality can be extremely difficult. Frequently it is contingent on the opinions of the supervisor. So, although it is undoubtedly important to consider quality, it’s a whole lot less problematic when you focus on quantity -- and move up from there.