Decoding bias coded words in Job Descriptions

truly inclusive blog - bias coded words in jds - banner


A boy got hurt on the playground and was immediately rushed to the school clinic. The nurse there was visibly upset and cried “that’s my son” but the nurse was not his mother. Why did the nurse call him that?

I am sure you correctly identified that the nurse was indeed the father of the boy. This famous puzzle, or rather a version of it, was often used in DEI training to expose our bias blind-spots, especially our gender schema, which ascribes gender to certain social roles, behaviors and traits. For instance,  if I say the perpetrator of a crime was ‘tall and strong’- many would assume it was a man. If I were to say my colleague is ‘too bossy’-   a lot of people would assume it was a woman.

Although the words, ‘tall’, ‘strong’, ‘bossy’ are inherently devoid of any gender or racial indicators, it is almost as if such meanings have gotten coded into these words through regular associations and suddenly become glaringly evident in certain contexts.

Coded language

Code words are essentially gate-keeping words, whose alternate meanings or connotations are known only to a subset of people initiated to its double meaning . Eg. “Mary Jane ” is a code word for “Marijuana”. This property of  coded language where you can say something subversive without actually having to, has much appeal in propaganda and politics. Eg. one can publically make a case for seemingly innocuous “religious freedom.” instead of straight-up  “anti-LGBTQ” or talk uncontroversially about  “middle class” rights instead of “white people” rights.

Most of the coded words we mentioned above are deliberately coded with specific intentions in mind. But, coding can also happen at a collective unconscious level based on cultural biases, stereotypes and schemas. For instance, if a word- say ‘thug’, is overarchingly used in the context of  ‘Black man’, they form a word association. That is, hearing the word ‘thug’ automatically primes you to ‘Black man’ and vice versa. And every time you hear ‘thug’, you are also thinking of a black man, so much so that they may seem synonymous and the ‘norm’. Once this association is set, even counter examples such as a black president, get treated as exceptions. Because this happens at an unconscious level and is socially pervasive, such coded words have the potential for covert discrimination.

Coded language in job descriptions

A Job description is one of the first encounters a prospective employee has with a company. Ideally, every individual, regardless of their gender, sexuality, race, age, ability etc. should see themselves being reflected in it. Even the slightest impression of not being a ‘cultural fit’ can make a highly qualified candidate not apply and have a less favorable opinion of the brand.

As we saw earlier, people already have schematic/prototypical conceptions of who is most suited for a role. Eg. Coders are typically heterosexual young males. A CEO or a professor is an older white male; a nurse is a woman, a basketball player is a black male etc. The more a candidate digresses from this prototype, the more they are culturally and self-assessed to be unsuitable for the job. Eg. A queer woman, or a woman of color is twice removed from the prototype of a coder.  Research  says that women only apply if they feel they are 100% qualified and are a closer match to the prototype. This could also be true for other underrepresented groups as well.


Now the words used in job descriptions can either enable such stereotypes or dismantle them entirely. Hence if true inclusivity is your priority as a brand, it’s good to be aware of some of these coded words. Here we are primarily going to discuss two types – gender and age coded words.

Gender coded words

Gender-coded words are basically character traits to which we assign masculine or feminine values. Research shows that, agentic or leadership-congruent traits are traditionally viewed as masculine while communal or nurturing traits are viewed feminine.

Masculine: competitive, aggressive, challenge, decisive/decision, dominate, champion, driven, fearless, lead/leadership, strong, expert, driven, expert, objective, principles


Feminine : collaborative/collaborate,dependable, honest, loyal, interpersonal, enthusiastic, committed, connect/connected, patient, support share, responsible, understand, together, feel

A seminal research, published in the Journal of Personality and Social Psychology, has found that  advertisements for stereotypically male dominated jobs like Tech, Science, manufacturing and for leadership roles such as director, head, partner, had more masculine wording, which led women to

  1. Think more men worked there
  2. Believe they would not belong in that position
  3. Find the job less appealing

This is in spite of the fact that the masculine-coded words did not affect women’s assessment of their own abilities to do the actual job. That is, they did not feel they lacked the above mentioned agentic traits; they merely felt that they would not get a favorable response or a call-back because they were not the ‘ideal’ the company was looking for. Conversely, men showed no difference in anticipated belonging based on either masculine or feminine wording in male or female dominated industries. 

These researches have also found that substituting masculine coded words for feminine coded or neutral words see a significant difference in applications from women.


This research has also been replicated by Harvard’s Gender Action  and Appcast. You may find the complete list of the  identified gender coded words on the gender decoder tool by Katmatfield. 

We also need to acknowledge a caveat in the research, in that it didn’t touch on non-binary-coded words or include non-binary individuals.

Age coded words

Just like gender, age bias can also get coded into our JD vocabulary. It is clear that Tech hires a higher proportion of younger workers than non-tech. . According to Visier the average tech worker is 38-years-old, compared to 43-years-old for non-tech workers. Another research by Payscale,  says that the median age of a tech worker is as low as 31. 

The discrimination underlying this can be rephrased in the biased  words of Mark Zuckerberg himself, “I want to stress the importance of being young and technical; Young people are just smarter”. Research published by the National Bureau of Economic Research confirms that language used in job advertisements- eg. ,young’, ‘energetic’, ‘tech savvy’ etc.  can act as gate-keeping mechanisms that deter applicants aged 40 and older from applying.  Once again the age-coded words reflect stereotypes related to age and fitness; age and technology; age and obsolescence etc.


Given below are a few age-coded words and phrases that JDs are replete with:

Young-age coded: Smart, Innovative, adaptable, technologically savvy, social media savvy, digitally native, energetic, innovative, creative, flexible, adaptable, fit, high-energy, high-motivated, high-potential, fresh, dynamic, enthusiastic outgoing, sociable, comfortable with change,   ability to learn new technologies,  up-to-date with current industry jargon, super fun work environment, fast-paced environment 


Words appealing to all age-groups :  knowledgeable, dependable, expert, hardworking, loyal, dedicated, experienced etc. 

Just as in the case of  gender-coded words, the respondents did not rate themselves low on technical capabilities etc., just that they did not find themselves fit for the role or the company. It was also found that  using language which appealed more broadly to older people does not deter younger applicants.

Common concerns in substituting coded language with neutral language

1. It merely treats the symptoms and not the root cause of gender/ age parity. 

Yes, gender or age parity is a systemic problem, and  needs to be addressed at multiple levels. If not complemented by strong and consistent DEI effort, this alone may not yield results. It may be hard to believe that simple word changes could lead to substantial changes in workplace gender/age balance. But to achieve this, we need good candidates from underrepresented people in the first place, and if the job descriptions are deterring them from applying, gender or age parity is bound to remain a pipe dream. The idea, after all, is to increase representation of women in traditionally male and young people dominated fields and vice versa, so that the agentic or communal traits will eventually stop being gender-coded, and the notion of a male nurse and a woman surgeon is also the cultural/psychological norm rather than the exception.


2. Certain words like ‘leader’ or ‘objective’ are crucial and irreplaceable

This is a status-quo problem– the tendency of continuing things the way they are because they feel easy, hence ‘natural’. Yes, using the gender neutral word ‘mentor’ instead of the masculine coded ‘leader’ may sound unnatural or even frivolous at first. But, constant usage usually naturalizes new words and usages.  And if neutral words encourage underrepresented people to actually apply for positions and see themselves represented, this is one trade-off that is worth having.


3. Recognising such words and finding unbiased alternatives is too tedious and time consuming

This is true indeed. This is where research based assistive technologies like’s Inclusive Language Screener can help. The screener  not only detects unconscious bias and coded language, but also offers real-time inclusive language recommendations and insights on the nature of errors. You can even track your degree of inclusion over time, and measure your #trulyinclusive transformation.


Begin your #trulyinclusive transformation today. Your journey towards true inclusivity starts with a single click!

Sign up now! 

Comments are closed.

Thank you!

Click the button below to download the audit report.