Research shows that women who speak up at all are risking more than men. Something as minor as telling observers that a CEO “tends to offer his (her) own opinions as much as possible,” and that, “Compared to other CEOs, Mr. (Ms.) Morgan talks much more than others in power,” caused observers to respect Mr. Morgan more and Ms. Morgan less.1 This approval or disapproval was based on gender alone. It isn’t fair.
Speaking up in forceful, assertive ways is even more risky for women. They are burdened with cultural stereotypes that typecast women as caring and nurturing.2 Speaking forcefully violates that cultural norm and women experience a more punishing backlash than men.
In a landmark study, Victoria Brescoll and Eric Luis Uhlmann asked the question, “Can an angry woman get ahead?”3 Their study documented the unequal penalty women experience for showing anger at work, but then went further to explore the reasons behind this gender effect. Their results suggest that the penalty occurs because observers attribute women’s anger to internal characteristics (“she is an angry person,’’ or ‘‘she is out of control”) while attributing men’s anger to external circumstances ("he's having a bad day," or "things were out of control so someone had to take charge").
What this previous research, along with our own, confirms is that emotional inequality is real and it is unfair. And while it is unacceptable and needs to be addressed at a cultural, legal, organizational, and social level—individuals can take control. We wanted to develop specific skills women can use on the job to be forceful, assertive and honest—without experiencing social backlash. Our first step was to recreate the social backlash and emotion-inequality effects in a controlled laboratory setting. We wanted to demonstrate the effects in a reliable way, so we could test ways to reduce them.
We created videotaped interactions so we could control what observers would see. The videotaped interactions featured either a male or female actor and took place in a meeting room seated at a table. The actors used identical scripts and we coached them so that their performances were as similar as possible. The only difference was that one actor was male and the other was female.
In this first study, 4,517 participants played the observer role. Each saw a single 30-40 second performance, and then rated the “manager” using a 20-item survey. The chart below illustrates the social backlash and emotion-inequality effects we observed. The bars represent the percentage drop averaged across status, competency, and worth, in that order.
Next, we decided to test whether brief framing could reduce the emotion-inequality effects. We tested three frames: a Behavior Frame, a Value Frame, and an Inoculation Frame.
• Behavior Frame: The actors described what they were about to say before saying it: “I’m going to express my opinion very directly. I’ll be as specific as possible.”
• Value Frame: The actors described their motivation in value-laden terms before making the statement of disapproval: “I see this as a matter of honesty and integrity, so it’s important for me to be clear about where I stand.”
• Inoculation Frame: The female actor suggested it could be risky for a woman to speak up the way she was about to: “I know it’s a risk for a woman to speak this assertively, but I’m going to express my opinion very directly.”
In this second study, 7,921 participants played the observer role. Each saw a single 35-45 second performance, and then rated the “manager” using the 20-item survey from Study 1.
Each of the frames worked. The chart below illustrates the positive impacts of the different frames.
This study shows that framing statements can help to solve social backlash and emotion-inequality effects. We believe that each frame works in a different way.
Behavior Frame: “I’m going to express my opinion very directly. I’ll be as specific as possible.” We think the Behavior Frame works by setting an expectation. It makes sure the statement that follows doesn’t come as a surprise. Without the frame, observers are blindsided by the force of the emotion and may assume the worst—that the person has lost his/her temper. The frame works by preventing this negative conclusion.
Value Frame: “I see this as a matter of honesty and integrity, so it’s important for me to be clear about where I stand.” We think the Value Frame works by giving a positive reason for the emotion. In fact, it turns the emotion into a virtue by turning it into a measure of commitment to a shared value.
Inoculation Frame: “I know it’s a risk for a woman to speak this assertively, but I’m going to express my opinion very directly.”
We think the Inoculation Frame works by warning observers that they may have an implicit bias. It causes them to try hard to be fair, or to adjust their judgment in an effort to be fair.
We were a bit surprised at how well it worked and we are skeptical that the Inoculation Frame will work if used repeatedly. It could be seen as “playing a card”—in this case the “gender card." Our concern is that it may create short-term benefits, but damage a user’s reputation.
Explain Your Intent Before Stating Your Content
Speaking forcefully creates a social backlash for both men and women—though it’s more severe for women. This backlash occurs when observers use the emotion to draw negative conclusions about the speaker’s intent. The backlash is reduced when the speaker takes a few seconds to explain his/her positive intent before stating the content.
We tested three of the statements a person could use to explain his/her intent—Behavior, Value, and Inoculation Frames. We can conclude that the Behavior and Value Frames are effective and are safe to use repeatedly. The Inoculation Frame works in the short term, but we won’t recommend its repeated use until we’ve tested it more thoroughly.
If not acknowledged or managed well, emotional inequality and social backlash can adversely affect an individual’s career and can prove costly to an organization’s effectiveness. We believe the implications of this research will empower individuals and leaders to engage in and encourage candid discussion while minimizing negative impacts.
1Victoria L. Brescoll, “Who Takes The Floor And Why: Gender, Power, And Volubility In Organizations,” Administrative Science Quarterly, 56, no. 4 (2011): 622-641.
2Alice H. Eagly and Steven J. Karau, “Role Congruity Theory of Prejudice Toward Female Leaders,” Psychological Review, 109, no. 3 (2002): 573.
3Victoria L. Brescoll and Eric Luis Uhlmann. (2008). “Can an Angry Woman Get Ahead? Status Conferral, Gender, and Expression of Emotion in The Workplace,” Psychological Science, 19,no. 3 (2008): 268-275.
Sincerely,
David
This is great! Now I want to see the videos so I can see it in action! This feels like breaking news, but I’ve watched people frame statements all my life. I want to master this skill because I believe it would make a huge difference in my career and home life.
Thanks for the comment, and here’s the video, Eric (also found by clicking on the previous post, “One Simple Skill to Curb Unconscious Gender Bias”): http://www.crucialskills.com/2015/08/one-simple-skill-to-curb-unconscious-gender-bias/ –Steve at VitalSmarts
I really like the research on the utilization of the statements but would have enjoyed taking the gender bias out to see what would happen in the innoculation phase. By removing the gender, it would be possible if acknowledging that a statement itself might be risky not just the fact that it is risky for a female.
Brilliant! I am a huge proponent of what I refer to as ‘qualifiers’ (ie – behavior/value/inoculation framing).
Our tendancy, as humans, is to put our own spin on what someone else says unless that invididual thinks ahead and puts qualifiers in place to stop other individual’s propencity to ‘fill in the gaps’ with their own preconceived notions.
You also added excellent qualifiers on inoculation as well. Great article!
Interesting results! What if the woman mitigated ‘playing the gender card’ in the Inoculation Frame by simply stating “I know that I have strong opinions sometimes, but this is something I am passionate about..”? That would take gender out of it, and just make it part of a PERSONALITY trait, not necessarily a GENDER trait. I myself have been accused of ‘always thinking I’m right/strong opinion’, and am a woman. I find myself using this statement periodically.
This is VERY helpful. I have paid dearly for being an “assertive” woman in my career (actually, just being like an average man, IMHO…) and these are some very specific ways to minimize the inevitable backlash.
Excellent article, clearly stating the issue — and providing solutions! I have sent this article to several women I know. Thanks for the helpful tips!
Could you supply some statistics? Are the differences between the male/female bars significant, especially where they are close? What is the variance or standard deviation? Also, the first figure is confusing; the text states “The bars represent the percentage drop averaged across status, competency, and worth, in that order.” The bars are labeled with different levels of forcefulness. How are the 3 concepts “status, competency, and worth” related to the 3 levels of forcefulness? What does “in that order” mean?Overall, excellent, provocative, and pragmatic article; these added details would strengthen interpretation. Thanks!
Hi Sheryl,
We will release the entire study–with the data tables and statistics in about a week. It will answer all of your questions. I’ll answer a few of your specific questions here.
1. The dependent measures were Status, Competency, and Worth. Status and Competency were scales composed of 4 or more questions that were highly reliable (Cronbach’s alpha of .95 and .96). Worth was determined by asking what salary the person should be giving (within a set range). The bars were an average of the 3 scales–which was also highly reliable (Cronbach’s alpha = .97)
2. The “percentage drop” was calculated by subtracting the scores given to the actors in the three Forceful Conditions (where the actor expressed a disagreement on what was clearly a high-stakes issue–with increasing levels of forcefulness) from the scores given the same actor in the Neutral Condition (where it was not a crucial conversation–the actor gave a status update that didn’t include any disagreement, high stakes, or emotion). This drop was then converted into a percentage.
3. The phrase “in that order” doesn’t make any sense. It’s my mistake. It was a simple average of the scores for Status, Competency, and Worth.
thanks,
David
Thanks for the tools; I will definitely put them to work in my job. The data is interesting in that with qualifiers or without qualifiers women are not perceived as equal to men in status, competency or worth. Women do seem to benefit the most from using a qualifier when being only mildly assertive, while men benefit more from qualifiers when being moderately assertive. Very interesting summary of this work. How can we use it to change culture in the workplace?
Cheers to both Diane and Michelle! We did try an inoculation frame that didn’t mention gender. But we only tried it with the male actor. We knew we couldn’t have him say, “I know it can be risky for a man to be this assertive…” So we had him say, “I know it can be risky to be this assertive…”
Essentially, we were having him inoculate against the assertive statement itself. And it worked as well as the Value Frame!
Our next step is to test it with the female actor. I think it’s an inoculation statement that doesn’t call on the “gender card’, and yet gets the job done.
thanks,
David
Who provided the subjective determination of what would be considered “forcefulness” or not, and its varying levels? Appears flawed in its implied methodology.
Great question! Chase McMillan and I worked with the actors to get three levels of “forcefulness”. By “forcefulness” we meant increasing levels of anger, frustration, and disgust. While we coached the male and female actors to be as similar in emotional level as possible–we knew it would never be perfect. That’s why we systematically varied it from mild to moderate to strong. That allowed us to see the effect across a wide range of emotion–so we could be sure it wasn’t an anomaly of one particular performance.
Thanks,
David
This is extremely helpful, especially being a female Exec in a mostly male dominated environment. Frequently struggling in communicating effectively in meetings, and trying all sorts of “different” language. I will be utilising these tools, especially the first two! Thanks.
We’ve received a few excellent questions that I want to post and respond to. Here is the first:
I just read your article on Social Backlash in the Workplace, and had an incident where I spoke up and had the same backlash. Ironically, I was speaking to a female superior. Do we, as females, create these same backlashes for each other? Just wondering if your study delved into this aspect.
Our study did not explore this gender interaction. However, other studies have found that both men and women have similar implicit biases–and these biases have similar effects on their actions. This suggests that your experience is the norm, not an exception. Sad, isn’t it!
“There is nothing more painful to me at this stage in my life,” Jesse Jackson once told an audience, “than to walk down the street and hear footsteps and start thinking about robbery—then look around and see somebody white and feel relieved.” (Quoted in Chicago Sun Times, Nov 29, 1993).
Isn’t this frustrating! No one would accuse Jesse Jackson of conscious bias, but we all hold unconscious, implicit stereotypes that lead us to take actions that run counter to our values.
I applaud Facebook and Google for taking the lead on addressing these implicit biases. They are working to make our implicit biases discussable and to create venues where we can work to reduce them.
David
Here is another great question:
The article on How to Avoid Social Backlash in the Workplace was very timely and most interesting. I have worked in the Corporate environment for more than 25 years and have seen first hand how women have been penalized for being direct. Did your study take into account what impact did the Social Backlash and Emotional Inequality have on minoirties in the workplace?
Our study did not. Our actors were white and in their early 30’s. You can imagine that minorities, differently abled, elderly, and other groups would also suffer from society’s implicit biases.
There have been many studies that look at implicit bias and race. Here are links to a few:
1. Can An Agentic Black Woman Get Ahead?
http://pss.sagepub.com/content/23/4/354.short
2. Correspondence testing of recruiter bias and the Implicit
Association Test
http://www.psychometricsforum.org/uploads/Psyche/Psyche%2067_LR%20%282%29.pdf#page=5
3. The Impact of Implicit Racial Bias on the Exercise of Prosecutorial Discretion
http://digitalcommons.law.seattleu.edu/cgi/viewcontent.cgi?article=2082&context=sulr
Implicit bias is a tough problem that remains, even after legal and other more explicit barriers have been removed.
David
Hi David,
With this thought in mind, do you have plans to replicate the study with actors of diverse racial and ethnic backgrounds? If not, do you have plans to take racial and ethnic diversity into account in your future studies? The nation is increasingly becoming more diverse and performing studies such as these without taking this diversity into account renders the data less useful.
You are correct to ask about the broader context of the data. We held many things constant: ethnicity, age, clothing style, attractiveness–and the list goes on. Research is like this: in order to examine one factor under a microscope, you have to hold all other factors constant.
Others have looked at the backlash African Americans experience. Of course they had to hold everything except ethnicity constant. As a result, it still doesn’t reflect the true diversity of the workplace.
So, are these studies all too narrow to be meaningful? I don’t think so. Our results document a powerful force that interacts with many other factors–but it’s still very real.
David,
Thank you for the summary of this research and its appropriate application. Do you know if additional research is currently underway on the extensive use of the Inoculation Frame? I tend to agree with your assessment about its limited use. Even as I imagine playing out those scenarios, I can feel the dismissal that repeated use of the Inoculation Frame would likely trigger. I also appreciate your providing the sources. I will continue to explore.
Hi Kymm, No, I don’t know of any additional research on inoculation frames. However, last week I had the chance to share the inoculation video with a group of about 350 HR and Training specialists. I asked them if they would recommend using the strategy. About three-quarters said, No! Good to know, right?
Hello Kymm. See my comment to David below. You might find it interesting and related to your comment. I run a non-profit trying to drive gender balanced leadership in the S&P 500.
Elba Pareja-Gallagher
I wish I had learned about this 10 years ago – would have changed my career path. Love that you’ve provided concrete solutions. Here’s to the next phase!
Also – David, is this article published anywhere besides the blog?
Hello David. This is research we can apply in our daily lives, thank you for the summary. I too have concerns about the inoculation frame. I recently came across this study by the Academy of Management that found that women and minority corporate executives are penalized for engaging in “diversity-increasing” behaviors. It’s frustrating. The research found that “women leaders’ engagement in diversity-valuing behavior may be viewed as selfishly advancing the social standing of their own low-status demographic group.” Here’s a link to the report:
http://aom.org/News/Press-Releases/Women-and-minority-corporate-executives-are-penalized-for-fostering-diversity,-study-finds.aspx
The researchers addressed a phenomenon that women live with in the work world: their comments are perceived and evaluated more negatively and more exactingly than are men’s. The experiment using three different communication strategies to mitigate or lessen negative perceptions produced some interesting findings. It was surprising to me that of the three–behavior [expectation], value [turn comment into virtue/connection], and inoculation [preview the possible negative response and state that concerns are being stated directly]–that inoculation received the most favorable ratings in the experiment. I agree with the author that this strategy might work in the short term, but if frequently used, could damage the woman’s credibility among her peers. The strategies themselves are useful communication techniques. Using them all and varying which ones are used might be the best approach. One reason any of these strategies may be effective is that in using any of them, one must first slow down, back up, decide how to open with the concern and then how to phrase it. Stopping to cool down a little before giving a response is a standard recommendation. Cooling down is an excellent way to reduce the negative emotional content and to try to find the most persuasive or easiest to listen to way to present one’s reaction or reservations.