Podcast: Addressing Bias in Performance ReviewsRead Time: 14 mins
It’s the last quarter of the year, and for some, that means holiday gatherings and New Year’s resolutions. However, many managers and supervisors are instead saying “’tis the season for year-end performance reviews.” While these annual reviews have their shortcomings, companies across the globe, large and small, continue to use them.
In this episode of More with McGlinchey, Labor and Employment Members Camille Bryant and Mag Bickford discuss how bias negatively impacts performance reviews, and steps employers can take to mitigate risks.
Camille Bryant: Hi, I’m Camille Bryant, a Member in McGlinchey’s Labor and Employment practice group. And I’m based in the New Orleans office. I’m joined by my colleague and Managing Member of the New Orleans office, Mag Bickford. Mag, you’re someone that’s had a varied career. You were a plaintiff’s attorney, then you became a management-side labor and employment attorney, and you also spent a good bit of time in-house. And now as the Managing Member of our office here in New Orleans, you conduct performance reviews of others as well. So in your experience, what are some of the primary risks that employers are facing this time of year with performance reviews?
Mag Bickford: That’s a good question, Camille. You know, the biggest problem that I see with regard to performance reviews is the concept that everyone thinks they know how to do them, and how they should be completed and [how they] should be delivered. But in reality, it’s not an intuitive process. It’s a very thoughtful, thought-provoking process. And I think that’s where we have to start. The first big-ticket item that we see regularly in these types of issues is, what role do bias and subjectivity have in the performance review process? Typically we’re all given a form and we’re asked to fill out the form that’s usually pretty generic and not very tailored. And then we fill out those answers with our own biases and subjectivities playing into the answers. We tend to like people who are like us, right? In addition to making performance reviews tricky, “similar-to-me” bias can certainly interrupt the process and make people feel less included, and may even affect how diverse employees are actually treated in the performance review process, as well as the overall makeup of the organization.
So aspects such as [race, color, sex, veterans status, sex, sexual identity, sexual orientation, gender, age, disability, religion, and all the classes] that are protected by state and federal laws. All these [protections] can be unconsciously [triggered] to the detriment of the employee as well as the organization. So we have to know how to assign goals to an employee and how to assess their performance in an unbiased, not subjective, consistent way. Also these unconscious biases regularly play a part in the hiring process. So we want to make sure that we’re fixing that end as well as we go through this process.
Studies have found that gender bias can affect companies on how they write job advertisements. Certain terminology has been found to attract or detract [certain protected groups] such as a gender or sex or race. So we want to make sure that all these things are filtered out of this process [and we apply rules of equity and fairness]. Some things that seem to be very innocuous can actually result in this sort of similar-to-me bias, for example, where did you go to school? Where did you grow up? Do you know so-and-so, and things like that. Whenever you’re conducting a performance review for someone who went to the same school, are you going to treat them better than someone who didn’t? What about if you have an inflated impression of the graduates of that school, or the people who came to that school, or are affiliated with that school? That’s the sort of thing that you see pretty regularly in performance reviews that are a problem and need to be addressed.
Another thing that we see in performance reviews, again, in a similar fashion, are racial stereotypes. These are things that we learn throughout our life through different exposures, but they can be very problematic in the performance review capacity. There are stereotypes: Asian Americans are smart, or maybe they’re passive, or maybe they lack social skills. African-Americans are angry or too aggressive. Latinos are hotheaded and emotional. Older people are not going to be contributing to the workforce for a long time. These sort of things are actually direct discrimination violations that will get you in trouble under Title VII, as well as state discrimination laws. We have to make sure that all of these things are filtered out of the process, and the process becomes objective and constructive.
Another thing is, we have to set up our processes so that these biases and stereotypes and thought processes that are in violation of Title VII and state laws are filtered out. And we’re going to talk a little bit about that in detail in just a few minutes, but things that you want to do to make sure that these things are filtered out of the process of performance evaluation is to make sure that there are things like consistency, that people who are doing the reviewing are trained properly on how to do it. You don’t want too generic of a form because that allows biases to sneak in the back door. You want to make sure that this is not an impulsive process, that you’re just sort of checking off boxes and doing things like that. You want objective, clear information. And you also want to make sure that you check prior documentation, to make sure you’re consistent and respectful of what has happened in the past. Another thing that has to happen is reviewer training. You know, like I said, when I started: this isn’t an innate process. This is something that’s got to be very thought-provoked, very methodical, and very consistent, and very much in line with applicable laws.
Performance reviews are frequently the topic or subject in Title VII and state law discrimination cases. Camille, what litigation have you seen lately regarding these issues?
Camille Bryant: Well, Mag, it’s like you said, you know, these biases lead to direct evidence of discrimination under Title VII. And so it’s not a surprise that we’re seeing litigation coming out of this. And it’s also not a surprise that we’re seeing public scrutiny, even though it may not necessarily be full-blown litigation. We’re seeing employees being more vocal and they’re expressing their thoughts and opinions on social media, and that’s also having a direct effect on company branding.
The one that comes to mind is Amazon. Earlier this year, they actually faced a good bit of scrutiny over their hiring and their promotion practices, particularly because several black employees alleged that they were promoted less frequently. And they also specifically alleged that they were subjected to harsher evaluations as opposed to some of their other, non-black colleagues. So we know that this is an issue that’s definitely affecting companies. And as a result, Amazon actually had to issue a public statement saying that they were going to re-examine how they were actually conducting performance reviews and that they were going to take a look at their attrition rates. And as part of that, they were really going to do a statistical analysis to see where the gaps were in their processes. And so generally statistical analysis is significant with respect to various demographic groups, if there’s a 95% chance that the difference is due to discrimination.
And so we also saw some issues with this with regards to Yahoo a few years back, where they had a male employee that was alleging that females were treated better during the performance review process. And that actually did lead to litigation. And in that case, there were specific allegations that the company was using these ratings specifically to oust people. So they weren’t, you know, genuine, they weren’t actually based on the employee’s performance. They were based on a lot of these subjective things that you just mentioned a few moments ago. And so they were using the ratings as a process to sort of weed out employees that they didn’t feel as valued, allegedly. Those were the allegations that were put forth by that specific employee.
And so while employers certainly need to evaluate employees’ performance, we want to make sure that we have people in the jobs that are effective and efficient and that are respectful. But at the same time, these systems and review of these systems have to be non-discriminatory. And with regards to the bias, as you said, oftentimes we’re unaware of these subtle manifestations that we have with regards to gender or race or religion or any other protected class that may be affecting how we’re doing our jobs, or how we’re viewing other people and how they’re doing their jobs as well.
We’ve seen this renewed push in the DEI space with regards to promoting equity across various demographics and combating gender bias by recruiting and retaining more women and fields that may traditionally be male-dominated. But the issue with that is you increase all of your statistical measures with respect to these different demographics. And so then the question becomes, are these things still an issue? Does equality still exist? How is it achieved? Are bias issues resolved now that we’ve got 50% women and 50% men? So now everything’s equal? Well, the shorter answer is no. Because oftentimes when women or other unrepresented groups become more represented in a specific industry or within a specific company, then those biases can remain evident because the very shift in composition may lead people to more readily infer that the discrimination is now a thing of the past, which oftentimes is an erroneous assumption.
That idea also aligns with a lot of evidence that we’re seeing that’s showing that individuals who sort of hold this type of belief, that for example, women no longer face discrimination, tend to lack awareness of the very things that that particular group may be facing. And as a result, they don’t take any steps to address them. So this is definitely a multifaceted issue. It’s definitely leading to litigation as we’ve seen, and as we continue to see. And so unless employers identify these issues, correct them, and do the training, like you said, it’s going to continue to be an issue. And so, one of the questions that I have for you is, because we know that these are problems that are inherent within these systems, how can employers drill down and identify specifically the issues that they have in their evaluation processes?
Mag Bickford: That’s a good question, Camille. So there are things that employers can do to try and weed out the things that put them at risk in the performance evaluation process. The first place to start is sort of a coordination of metrics. So you’ve got this form or this document online that you’re filling out. There’s always going to be a sense of subjectivity to that process to some degree. But what you can do is use specific analytics of performance and cross-reference those with the performance review to make sure they’re consistent. So for example, if you know that person has to make 15 widgets, and 15 is an acceptable performance, and they made the 15 widgets, but then you say that they aren’t such a good employee. If you compare that data, you’re going to say, “well, what is that about? Why is it so inconsistent?” So using analytics to cross-reference with the actual performance review can be very helpful and help employers to have a more objective process. HR can help you with this. They can do the comparison, and that should be a good place to start.
Another thing you want to do is look at the questions you’re asking on your performance evaluation. You want to evaluate that very closely to make sure that the questions are artfully drafted so that they give you relevant information. Those little boxes that you check are often very generic and make things a little bit difficult in terms of really identifying key factors. You want to develop such strong questions that they give you the answers you need. So maybe you want to ask the manager to identify three strengths of that particular employee in a specific area. That might be a more thought-provoking, sincere process. You’re not just going to throw a bunch of lists on the page that maybe fit the situation, or maybe not. You want them to dig deep for real answers rather than allowing their preconceptions to just shade the answers.
So one useful tool for combating bias is having calibration meetings. Calibration is the act of reviewing and adjusting employee performance reviews by more senior staff, or a specially selected team, or an outside vendor, to ensure that it’s an equitable process. It’s an additional measure to combat bias and organizations can use the calibration process to ensure manager reviews of employees are objectively verified. It allows for the comparison of the performance reviews to the different metrics of the diverse groups in your organization to combat bias. It’s a way to ensure that managers are making quality, verified performance reviews. If you’re going to do that, you want to make sure you do it before you give out the money, because it may require adjustments.
Another thing that’s important to note in your policy – it’s often a good idea to stress [that] just because you’re having a performance review does not mean you’re necessarily getting a pay raise. That’s something you want, because sometimes that’s the case, right? And you don’t want people to say, “oh, it’s December, I’m getting my performance review and I can’t wait to get my new raise,” when in fact they’re not. That’s not a good platform for employee development, when the [correct] expectation is not set.
You want to make sure that you’re calibrating at the right time. You’re coordinating with senior management. And when this is done correctly, it’s really an effective tool to make sure that the performance evaluation process is not biased in any way. Think about things – get away from group-think. We don’t want to just rubber stamp things into existence. We really want to focus on achievements or conquering issues, or really having an individualized assessment, as opposed to just some sort of rubber stamp of things that are going on.
So, Camille, what do you think the options are? If you do find yourself in a situation where you have room for improvement in the performance evaluation process?
Camille Bryant: Well, Mag, like you said, these performance reviews are always going to have a level of subjectivity to them. And by default, more often than not, they are going to have some issues. And so, you know, when we think about how to identify these areas and how do we improve, the first step is identifying that there’s a problem. And once we know that there’s a problem, where do we go from here? How do we move forward? And that’s really where your prevention strategy really has to take hold.
So to really reduce the effect of similar-to-me bias, you really want to require specificity in your manager’s training. You really want to dig deep and get real answers to the questions. And so part of that’s going to revolve around making sure that the right questions are asked as part of the evaluation process.
You know, there’s been a lot of studies and data out there that’s shown that assessments are more effective when there’s actually a set criteria of data that managers are working on consistently. And so while oftentimes we want to give employees that continuous feedback throughout the year, which definitely has its place and its value. You shouldn’t remove that entirely, but you do want to make sure that when you give that sort of unstructured feedback, that you’re checking to make sure that there aren’t any of those biases that are creeping in. Because without some set criteria, people will likely shape their feedback based upon their life experiences. And so that’s where those stereotypes come in, that you mentioned earlier. So I was thinking, you know, you mentioned that certain groups are considered aggressive or angry, and more times than one, I can recall someone telling me I need to smile. I wonder if that’s something that would have been said to someone of a different demographic. So we really have to be mindful of some of these things.
Also, it’s really important that companies help their raters or their managers by giving them the format and then allowing them to fill in the blanks. And so nudging managers into specifically thinking about the situations and the behaviors rather than the personality types. So you really want to move away from the personality type, because that’s where the similar-to-me bias comes in, and rather look on “what are the terms and conditions of this job? What are the behaviors that are needed for this employee to do this job? And how is this employee responding in the particular situation with which they’re faced?” and less so on personality fit or culture fit, which is a lot of what we like to use in the HR world.
One other thing that’s really important to remember is that we talk some about gender bias, but gender bias can also have a huge impact on the experiences and assessments of non-binary and other transgender individuals. Historically we’ve thought of gender bias as, as male or female and very binary. But we really have to expand our thinking and make sure that those raters and those who are evaluating within our companies are also expanding their thinking to make sure that their ratings properly encompass the different situations and scenarios that some of your other employees may be facing. And so, although these biases may manifest differently for transgender individuals, in some ways, you know, they are similar. So it’s important to stay alert and make sure that your managers have their eyes open to all of those different situations.
The other thing that’s really important to remember is, you either need to make sure to give everyone the benefit of the doubt or don’t give anyone the benefit of the doubt. You know, there’s always the situation where we follow the rules until we don’t follow the rules. And so, so much about performance reviews is about consistency across the board. And while you will certainly have those extenuating circumstances from time to time where something very unique is happening, you really want to make sure that you’re applying your rules consistently, and that if you do give employees the benefit of the doubt, you do that for everyone.
The other thing that’s really important to remember is to seek input from other individuals who work with that employee. And that can be done perhaps as part of the calibration process, but you want to talk to not only direct reports, but to peers and to clients. If you have an employee that you think is receiving a harsher evaluation than may be necessary, you want to do a complete 360 review to figure out where is this review really coming from? Where is this assessment really coming from? And part of that may be by asking others within the department how that employee is performing, and not solely relying on the evaluation of one person solely.
With that also comes not making assumptions. Oftentimes we think that, you know, mothers are doing one thing and then fathers are doing another. And along those same lines, oftentimes in the performance review process, there are numerous studies that have shown that women receive harsher performance reviews when they come back from maternity leave. Even though their work performance may have been the same before and after going on leave, the studies have shown that they are treated more harshly. And so it’s really just important to make sure that you’re identifying these issues, that you’re taking preventative steps to train your managers and your supervisors, and to calibrate, you know, Mag, as, as you said, once the review process is over, to make sure that some of those preventative steps are actually coming to fruition and eradicating biases. And to the extent that they’re not, then we have calibration to level things out to make sure that everyone’s being treated equitably.
Mag Bickford: Camille, a lot of times we see statements on performance reviews that are sort of conclusory and don’t really drill down to the facts. Statements like, “oh, she’s not a team player” or, “oh, she has a bad attitude.” They seem to be statements that the reviewers often use, but they don’t really address the behaviors. What do you think that there’s a better way to do that?
Camille Bryant: Well, like you just said, you have to drill down. So if you get a performance review that says, “oh, she’s not a team player,” what about this particular employee makes them not a team player? Are they missing deadlines? Are they not attending company functions? Are they not willing to pitch in and burn the midnight oil on a project when the rest of the team is staying late to work on that project? Those are some of the questions that I would want to know. When you also say the employee has a bad attitude, you know, what about that employee’s attitude is off-putting? I would also want to know, are there other employees in that same demographic or in a different demographic that are not being treated the same way? Because like you also said earlier, you know, there are certain groups that get stereotyped as aggressive or as angry. So is part of that, “oh, she has a bad attitude” some of our own unconscious biases that may be creeping in? Or is it truly an attitude problem? Because it may be. But you really have to dig deep and ask those questions to really figure out whether or not this is bias-related or is this, in fact, a problematic employee that needs to be disciplined at the end of the day for their conduct.
So those are just some of the high-level thoughts that we have on performance reviews. Thanks, Mag, so much for joining me for this conversation. I know that we have these discussions all the time and it never gets old. So with that, I’d also like to thank all of our listeners for tuning in and for joining us for this latest installment of “More with McGlinchey.” If you have any questions regarding your evaluation process, please feel free to reach out to any of us here at McGlinchey.
Subscribe wherever you listen to podcasts: