Three quarters of women labelled ‘emotional’ in performance reviews, study finds, suggesting lack of progress on gender-based language bias

Three quarters (78 per cent) of women have been described as ‘emotional’ in performance reviews, compared to just 11 per cent of men, new data has revealed.

The study by Textio, which analysed performance reviews for more than 25,000 people at 253 organisations, revealed that almost three in five (56 per cent) women received feedback labelling them as ‘unlikeable’ in performance reviews, a criticism faced by only 16 per cent of men. 

And while two thirds (67 per cent) of men were described as intelligent, only 32 per cent of women said the same.

More than a decade ago, in 2014, a Textio report found three quarters (76 per cent) of high-performing women had received negative feedback, compared to 2 per cent of men. Descriptors relating to personality, such as ‘abrasive’, ‘difficult’, ‘friendly’ and ‘helpful’, were commonly used to describe women, but rarely for men.

Gifty Enright, diversity and wellbeing speaker and author, said the figures were “disheartening but sadly not surprising”.“A decade on we’re still battling the same old issues. It’s frustrating to see how little progress has been made,” she said. 

Elizabeth Willetts, founder and director of Investing in Women, deemed the findings “deeply concerning”.

“Despite years of conversations about workplace equality, it’s clear that language bias in performance feedback remains a significant barrier for women and marginalised groups to achieve their full potential,” said Willetts.

“This bias not only impacts individual confidence and career trajectories, it also contributes to systemic inequities such as the gender pay gap and underrepresentation in leadership roles.”

The gender divide
Regardless of performance level, women received around 22 per cent more personality-based feedback than men, regardless of whether they were low or high performers.

The study also found that while women were seven times more likely to internalise negative stereotypes about themselves, men were 2-4 per cent more likely to internalise positive stereotypes. 

The term ‘overachiever’ – applied to good performers transcending low expectations – was most frequently applied to women of colour.

Men were three times more likely to recall being labbelled as an underachiever. 

The gender of the manager was not significant in the findings, as both male and female managers gave similar biased feedback. 

According to the 2024 Women in the Workplace survey from LeanIn.org and McKinsey & Co, more than a third (36 per cent) of women thought their gender played a role in missing out on a promotion, pay rise or a chance to get ahead, compared to 15 per cent of men. 

Willetts said the language used in reviews held “incredible power” because it “shapes how employees see themselves, how their colleagues perceive them, compensation decisions and, ultimately, how their careers progress”.

“When women and marginalised groups are disproportionately described as ‘unlikeable’ or ‘emotional’, it reinforces damaging stereotypes, creating additional hurdles for them to overcome,” she explained. 

Enright added that the language in performance reviews was so powerful because it “doesn’t just reflect bias, it reinforces it”. 

“Words such as ‘unlikeable’ or ‘emotional’ don’t evaluate performance; they attack character and create barriers. It is not right to be attacking women's character. For women and marginalised groups, it’s exhausting to constantly navigate these damaging labels,” she said.

Enright warned that this kind of bias in language could diminish confidence, hinder career growth and push talented people out of organisations entirely. “It’s a waste – not just for the individuals, but for companies that lose out on the brilliance and innovation that diversity brings,” she added. 

Michelle Gyimah, pay gaps strategist at Equality Pays, said: “The fact that language about likeability is still used in performance reviews is problematic because these statistics show that likeability can be, and is being, used as a weapon or a reward depending on the whim or biased behaviours of the person doing the evaluation.”

Racial discrepancies 
The Textio study found white people were more than twice as likely to recall being called ‘likeable’ compared to almost every other group.

More than 60 per cent of Hispanic and Black people recalled being called ‘emotional’, compared to 21 per cent of white people.

Additionally, about half of white and Asian people recalled being described as ‘intelligent’, compared to just 18 per cent of Black people and 17 per cent of Hispanic and Latino people.

There were discrepancies among Asian populations, with nearly a third (30 per cent) of South Asian people recalling being described as ‘difficult’, compared to just 9 per cent of East Asian people. 

White and Asian men received the highest-quality feedback overall, and were more likely to be described as ‘brilliant’ or a ‘genius’. 

Willetts said it was "particularly troubling” to see how gender bias intersected with race. “The fact that Hispanic and Black employees are far more likely to be described as ‘emotional’ highlights how deeply entrenched these stereotypes are – and how they can magnify existing disparities in the workplace,” she added. 

According to the Women in the Workplace survey, 27 per cent of Black women have had their judgement questioned within their area of expertise, compared to 17 per cent of men. 

The report also found three in five (61 per cent) Black women thought their race or ethnicity had played a role in them missing out on a raise, promotion or chance to get ahead. For Asian women, this figure was two in five (39 per cent).

Only 38 per cent of Black women said their employers promoted their contributions to others, compared to 48 per cent of white women. 

Meanwhile, the Fawcett Society’s Broken Ladders report found three in five (61 per cent) women of colour had changed something about themself to ‘fit in’ within the workplace, compared to 44 per cent of white women. The report also found three quarters (75 per cent) of women of colour had experienced racism at work. 

How can organisations combat language bias? 
Enright urged organisations to take language bias in performance reviews seriously. She suggested managers be trained to spot and challenge their biases, and tools be used to check for biased language in reviews. An objective framework can help ensure performance evaluations are fair.

“Most importantly, hold people accountable. Change won’t happen without real commitment at every level… Ultimately, this isn’t just about women or marginalised groups – it’s about creating workplaces where everyone can thrive,” she said. “The evidence is right there. Now it’s time to act.”

Gyimah said performance reviews should focus on the outputs provided, not on personal traits or characteristics that have been decided by the person doing the review. “Skills, outcomes and measurable contributions are what is important here. Taking this approach can help eliminate subjective descriptors that do more harm than good in the workplace,” she explained. 

Willetts said organisations must take “proactive steps”, including training managers on the impact of bias in feedback and equipping them with the tools to provide constructive reviews. She urged organisations to standardise performance reviews with clear criteria “tied to measurable outcomes rather than subjective impressions”. 

She said: “Bias in feedback doesn’t just hurt individuals, it hurts organisations by limiting the potential of diverse talent who could have a real positive impact on your business's bottom line. Addressing this issue is not just a moral imperative – it’s a business one.”

The orginal article was posted in People Management in 2025 and can be found here.