Skip to main content

Masculine men do not like feminine wording: The effectiveness of gendered wording in health promotion leaflets in the UK

PLOS ONE, 2022

Katherine Baxter, Barbara Czarnecka, Bruno Schivinski, Maria Rita Massaro

PLOS ONE, volume 17, issue 10, article e0273927, 27 October 2022

  • Open access
  • 3 studies in the UK
  • 247 + 599 participants
  • Survey experiments and interviews
  • Mixed methods
  • MANOVA
  • United Kingdom
Download the full text (PDF)Published version

Free full text: the open access article (CC BY 4.0), re-typeset as a clean PDF with the same content as the published version. Please cite the published version.

Keywords
  • gendered wording
  • agentic wording
  • communal wording
  • endorser gender
  • health promotion
  • health communication
  • social marketing
  • physical activity
  • walking
  • leaflets
  • gender role identity
  • masculinity
  • femininity
  • gender stereotypes
  • message-audience congruence
  • homophily theory
  • Gender Trait Index
  • attitude towards the ad
  • advertising credibility
  • behavioural intention
  • mixed methods
  • sequential design
  • survey experiment
  • mall-intercept survey
  • semi-structured interviews
  • directed content analysis
  • MANOVA
  • United Kingdom

Whether a health leaflet uses ‘masculine’ or ‘feminine’ wording matters mainly for one audience: masculine men. Across two survey experiments and interviews in the UK, masculine men rated leaflets with communal, caring wording poorly, while women and less masculine men rated all versions positively.

Key findings

  1. Masculine men dislike communal wording. In Study 3 (n = 599), men whose dominant gender role identity was masculine gave the communal-wording leaflets the lowest ratings of any group (attitude 3.38 and 3.65, against 5.3 to 6.2 for everyone else), whoever the doctor endorsing it was (Baxter et al., 2022).
  2. Everyone else liked every leaflet. Women and feminine men evaluated all four leaflets positively, whether the wording was agentic or communal and the endorser male or female; their ratings differed, but none was negative (Baxter et al., 2022).
  3. Men preferred the male doctor with agentic wording. In a mall-intercept experiment with 247 adults (Study 1), men rated the agentic leaflet with a male endorser more positively than women did (4.32 vs 3.34, p < .001) and reported higher intention to walk (3.33 vs 2.60, p = .012); most men interviewed in Study 2 preferred this leaflet too (Baxter et al., 2022).
  4. Most women had no preference. In 20 semi-structured interviews, most women did not prefer either leaflet, and in Study 1 women’s attitude towards communal and agentic wording did not differ significantly from men’s in most conditions (Baxter et al., 2022).

At a glance

Design
Sequential mixed methods: a 2 (endorser gender) × 2 (agentic vs communal wording) × 2 (respondent gender) survey experiment, semi-structured interviews and an online survey experiment adding dominant gender role identity
Stimuli
Leaflets promoting walking 30 minutes a day, five days a week, endorsed by a male or female hospital doctor (‘Dr James Baxter’ or ‘Dr Julie Baxter’) of the fictitious campaign ‘Life training’; about 3.5% gendered words
Study 1
247 adults in Hertfordshire, England (49.4% women), pen-and-paper mall-intercept survey, November 2017 to March 2018
Study 2
20 semi-structured interviews (10 men, 10 women) with Study 1 volunteers, March to July 2018; directed qualitative content analysis
Study 3
599 UK adults (50.9% women, mean age 38.8), online Qualtrics survey, January to June 2019; gender role identity measured with the Gender Trait Index
Outcomes
Attitude towards the ad, behavioural intention to walk and, in Study 3, advertising credibility
Analysis
Three-way (Study 1) and four-way (Study 3) MANOVA with Bonferroni pairwise comparisons, IBM SPSS 26
Ethics
University of Bedfordshire Ethics Committee (BMRI/Ethics/Student/201718/001 for Studies 1 and 2; BMRI/Ethics/Staff/2018-19/006 for Study 3)

Attitude towards the leaflet in Study 3, by gender and gender role identity

Masculine menFeminine menMasculine womenFeminine women1234567Male doctor, agentic wording, Masculine men: 6.136.13Male doctor, agentic wording, Feminine men: 4.424.42Male doctor, agentic wording, Masculine women: 4.694.69Male doctor, agentic wording, Feminine women: 5.275.27Male doctor,agentic wordingMale doctor, communal wording, Masculine men: 3.383.38Male doctor, communal wording, Feminine men: 5.345.34Male doctor, communal wording, Masculine women: 6.116.11Male doctor, communal wording, Feminine women: 6.216.21Male doctor,communal wordingFemale doctor, agentic wording, Masculine men: 5.805.80Female doctor, agentic wording, Feminine men: 6.046.04Female doctor, agentic wording, Masculine women: 4.964.96Female doctor, agentic wording, Feminine women: 5.465.46Female doctor,agentic wordingFemale doctor, communal wording, Masculine men: 3.653.65Female doctor, communal wording, Feminine men: 5.365.36Female doctor, communal wording, Masculine women: 6.106.10Female doctor, communal wording, Feminine women: 6.226.22Female doctor,communal wordingAttitude towards the ad (mean)Study 3, n = 599 UK adults; observed means from Table 6 of the article.
Observed means from Table 6 of the article (drawn for this page; the article shows these values in Fig 3). Masculine men rate both communal-wording leaflets low; every other group rates every leaflet positively.

Communal wording: how each group rated the leaflet

Male doctor, communal wordingMasculine men3.38Feminine men5.34Masculine women6.11Feminine women6.21Female doctor, communal wordingMasculine men3.65Feminine men5.36Masculine women6.10Feminine women6.22
Study 3 attitude towards the ad (observed means, Table 6 of the article; n = 599). Agentic-wording leaflets were rated 4.4 to 6.1 by all groups.

Tables

Table 1. Main effects and interactions of respondent’s gender, wording condition and endorser’s gender

Source of variation Pillai’s trace F p Partial η²
Main effects
Endorser’s gender 0.01 1.71 0.18 0.01
Wording 0.02 2.66 0.07 0.02
Respondent’s gender 0.04 4.60 0.01 0.04
Interactions
Endorser’s gender x Wording 0.01 1.68 0.19 0.01
Endorser’s gender x Respondent’s gender 0.00 0.48 0.62 0.00
Wording x Respondent’s gender 0.08 9.75 0.00 0.08
Endorser’s gender x wording x Respondent’s gender 0.02 2.54 0.08 0.02

df = 2, error df = 238.

Table 2. Pairwise comparisons (estimated marginal means) with Bonferroni adjustments by gender, endorser’s gender, and wording condition

Endorser’s gender Wording Respondent’s gender M SE p 95% CI lower 95% CI upper
Behavioural intention
Male Agentic Male 3.33 0.29 0.012 0.16 1.30
Female 2.60
Communal Male 3.47 0.30 0.427 −0.35 0.84
Female 3.23
Female Agentic Male 3.28 0.29 0.394 −0.32 0.82
Female 3.03
Communal Male 3.33 0.29 0.059 −0.02 1.14
Female 2.76
Attitude towards ad
Male Agentic Male 4.32 0.22 0.000 0.55 1.41
Female 3.34
Communal Male 3.86 0.23 0.056 −0.89 0.01
Female 4.30
Female Agentic Male 4.13 0.22 0.209 −0.15 0.70
Female 3.85
Communal Male 4.17 0.22 0.444 −0.61 0.27
Female 4.35

Estimated marginal means; SE and p refer to the difference between men and women in each cell (Bonferroni-adjusted).

Table 3. Descriptive means (Study 1)

Endorser Wording Respondent Intention M SD Attitude M SD
Male Agentic Male (N = 31) 3.33 1.28 4.32 1.00
Female (N = 33) 2.60 1.13 3.34 0.93
Communal Male (N = 30) 3.47 1.30 3.86 0.82
Female (N = 28) 3.23 1.06 4.30 0.94
Female Agentic Male (N = 33) 3.28 0.98 4.13 0.88
Female (N = 31) 3.03 1.14 3.85 0.92
Communal Male (N = 31) 3.33 1.23 4.17 0.76
Female (N = 30) 2.76 1.09 4.35 0.69

Table 4. Four-way multivariate analysis of variance (MANOVA) for the three sets of outcome measures: Behavioral intention, ad credibility, and attitude towards ad

Source of variation Pillai’s trace F p Partial η²
Main effects
Endorser’s gender 0.02 5.13 0.002 0.02
Wording 0.01 3.01 0.029 0.01
Respondent’s DGRI 0.06 12.29 0.000 0.06
Respondent’s gender 0.09 20.12 0.000 0.09
Interactions
Endorser’s gender x Wording 0.02 5.41 0.001 0.02
Endorser’s gender x Respondent’s DGRI 0.05 10.12 0.000 0.05
Endorser’s gender x Respondent’s gender 0.03 6.92 0.000 0.03
Wording x Respondent’s DGRI 0.09 20.58 0.000 0.09
Wording condition x Respondent’s Gender 0.25 64.65 0.000 0.25
Respondent’s DGRI x Respondent’s gender 0.00 .72 0.539 0.00
Endorser’s gender x Wording x Respondent’s DGRI 0.02 4.01 0.008 0.02
Endorser’s gender x Wording x Respondent’s gender 0.02 4.24 0.006 0.02
Endorser’s gender x Respondent’s DGRI x Respondent’s gender 0.04 8.89 0.000 0.04
Wording x Respondent’s DGRI x Respondent’s gender 0.13 30.75 0.000 0.13
Endorser’s gender x Wording x Respondent’s DGRI x Respondent’s gender 0.11 24.05 0.000 0.11

df = 3, error df = 581.

Table 5. Pairwise comparisons by respondent’s gender, dominant gender role identity (DGRI), endorser’s gender, and wording condition
Endorser’s gender Wording Respondent’s DGRI Respondent’s gender M SE p 95% CI lower 95% CI upper
Attitude towards ad
Female Communal Femininity Male 5.36 0.25 0.001 −1.36 −0.35
Female 6.22
Masculinity Male 3.65 0.25 0.000 −2.95 −1.95
Female 6.10
Agentic Femininity Male 6.04 0.26 0.029 0.05 1.09
Female 5.46
Masculinity Male 5.80 0.25 0.001 0.34 1.34
Female 4.96
Male Communal Femininity Male 5.34 0.26 0.001 −1.40 −0.35
Female 6.21
Masculinity Male 3.38 0.26 0.000 −3.25 −2.21
Female 6.11
Agentic Femininity Male 4.42 0.25 0.001 −1.35 −0.33
Female 5.27
Masculinity Male 6.13 0.26 0.000 0.92 1.95
Female 4.69
Advert credibility
Female Communal Femininity Male 5.12 0.27 0.000 −1.70 −0.64
Female 6.29
Masculinity Male 3.36 0.26 0.000 −2.87 −1.82
Female 5.71
Agentic Femininity Male 5.74 0.27 0.34 −0.28 0.81
Female 5.47
Masculinity Male 6.05 0.26 0.000 0.54 1.58
Female 4.98
Male Communal Femininity Male 5.24 0.28 0.001 −1.46 −0.35
Female 6.14
Masculinity Male 2.83 0.27 0.000 −4.21 −3.11
Female 6.49
Agentic Femininity Male 4.44 0.27 0.004 −1.31 −0.24
Female 5.22
Masculinity Male 6.37 0.27 0.000 1.37 2.45
Female 4.46
Behavioural intention
Female Communal Femininity Male 3.37 0.25 0.614 −0.36 0.61
Female 3.24
Masculinity Male 2.33 0.24 0.004 −1.19 −0.22
Female 3.05
Agentic Femininity Male 1.81 0.25 0.000 −1.89 −0.88
Female 3.20
Masculinity Male 3.36 0.24 0.003 0.25 1.22
Female 2.62
Male Communal Femininity Male 3.62 0.26 0.942 −0.49 0.52
Female 3.60
Masculinity Male 2.31 0.25 0.702 −0.60 0.40
Female 2.41
Agentic Femininity Male 3.92 0.25 0.000 0.72 1.71
Female 2.69
Masculinity Male 3.03 0.25 0.503 −0.33 0.67
Female 2.86

Estimated marginal means; SE and p refer to the difference between men and women in each cell (Bonferroni-adjusted). Behavioural intention was measured on a 1–5 scale.

Table 6. Observed means for attitude towards ad, ad credibility, and behavioural intention (Study 3)
Endorser Wording DGRI Respondent Credibility M SD Attitude M SD Intention M SD
Female Communal Femininity Male (N = 29) 5.12 1.10 5.36 0.84 3.37 1.44
Female (N = 45) 6.29 0.99 6.22 1.11 3.24 1.20
Masculinity Male (N = 48) 3.36 1.19 3.65 0.77 2.33 0.69
Female (N = 29) 5.71 0.70 6.10 0.71 3.05 0.60
Agentic Femininity Male (N = 26) 5.74 1.02 6.04 0.64 1.81 1.18
Female (N = 46) 5.47 1.48 5.46 1.36 3.20 0.86
Masculinity Male (N = 44) 6.05 0.52 5.80 0.54 3.36 1.00
Female (N = 31) 4.98 1.44 4.96 1.17 2.62 1.05
Male Communal Femininity Male (N = 25) 5.24 1.12 5.34 1.40 3.62 0.84
Female (N = 47) 6.14 1.00 6.21 1.08 3.60 1.00
Masculinity Male (N = 49) 2.82 1.36 3.38 1.12 2.31 0.90
Female (N = 25) 6.49 0.44 6.11 0.66 2.41 0.79
Agentic Femininity Male (N = 25) 4.44 1.00 4.42 0.48 3.92 0.95
Female (N = 56) 5.22 1.30 5.27 1.66 2.69 1.25
Masculinity Male (N = 48) 6.37 0.74 6.13 0.78 3.03 1.27
Female (N = 26) 4.46 1.71 4.69 1.43 2.86 1.24

The question

Health campaigns often try to reach men and women with different messages, and research on job adverts suggests that ‘agentic’ (stereotypically masculine) and ‘communal’ (stereotypically feminine) words can attract or put off each gender. Drawing on message-audience congruence and homophily theory, we asked whether gendered wording and the gender of the endorsing doctor change how British adults respond to a leaflet promoting walking, and whether a person’s gender role identity matters more than their gender.

Implications

Practical implications

Public health bodies, social marketers and commercial providers such as gyms need not tailor leaflet wording to women, who responded well to all versions. When targeting masculine men, they should avoid communal wording and consider agentic wording with a male endorser. Anyone using gendered language should also weigh efforts to promote gender-fair language and the risk of reinforcing gender expectations.

Theoretical implications

Gender-based congruence between message and audience was not needed for a leaflet to work, except for masculine men. Dominant gender role identity explained preferences better than gender, which challenges the assumptions of message-audience congruence and homophily for women, and suggests that changing gender roles shape how women respond to gendered content.

Limitations and future research

  • Survey experiments and interviews raise questions of external validity, and the cross-sectional design and purposive sampling limit causal inference.
  • The leaflets were seen without competing material, and only a doctor endorser was used; the interview sample was small.
  • The findings come from the UK and may not generalise to other countries, languages, ethnic groups or behaviours.

Questions this paper answers

Does gendered wording make health leaflets more effective?

Mostly not. In three UK studies, women and feminine men rated leaflets positively whatever the wording; only masculine men responded negatively, to communal wording (Baxter et al., 2022).

What is agentic and communal wording?

Agentic wording uses words stereotypically linked with men, such as decide, active, challenging, ambitious and self-confident; communal wording uses words linked with women, such as choice, pleasant, gentle, cheerful and communally.

How should health campaigns talk to masculine men?

Avoid communal, caring wording, which masculine men rated poorly (means of 3.38 and 3.65, against 5.3 or more for every other group), and consider agentic wording presented by a male endorser (Baxter et al., 2022).

Do women prefer a female endorser?

Not clearly. Most women interviewed had no preference between a male and a female doctor, and women in Study 3 rated all four leaflets positively (Baxter et al., 2022).

Key concepts

Gendered wording
The use of words stereotypically associated with men (agentic) or women (communal).
Agentic and communal traits
Agentic traits, such as ambition and assertiveness, are stereotypically linked with men; communal traits, such as warmth and care, with women.
Dominant gender role identity (DGRI)
Whether a person scores higher on femininity or on masculinity on the Gender Trait Index.
Message-audience congruence
The idea that messages work better when their content matches the characteristics of the audience.
Homophily theory
The idea that people are more persuaded by, and relate more to, sources similar to themselves.

Open science and materials

Study materials

Agentic leaflet (male endorser)
‘Hi, I am Dr James Baxter. I have been a hospital Dr for 15 years and I am now an avid walker. I recommend walking 30 minutes a day for 5 times a week as this can bring you lots of health benefits. Start walking soon! Making the decision to keep active is good for your mental health. Exercise promotes good mental health because it stimulates chemicals in your brain that leaves you feeling more self-confident and more relaxed than you were before you worked out. Exercise is also good for your physical health and for your weight. When you are physically fit you burn off calories. The more you exercise, the more calories you burn. This helps you maintain your ideal weight and keeps you looking good. Building walking into everything that you do is one of the best ways to get the workout you need. Challenge yourself and take the stairs at every chance you get. When you are driving, be ambitious and park in the car park or space that is the farthest from your destination. If you have been sitting for a long period of time, be determined to take some time to get up and walk around. The benefits are yours for the taking. All you need to do is start moving now either individually or even better walk with colleagues or friends! Start walking 30 minutes a day, 5 times a week!’
Communal leaflet (female endorser)
‘Hi, I am Dr Julie Baxter. I have been a hospital Dr for 15 years and I am now an avid walker. I recommend walking 30 minutes a day for 5 times a week as this can bring you lots of health benefits. Start walking soon! Making the choice to keep fit is good for your mental and emotional health. Exercise promotes good mental health because it stimulates chemicals in your brain that leaves you feeling more cheerful and more relaxed than you were before you worked out. Exercise is also good for your physical health and for your weight. When you are keeping fit you burn off calories. Understandably the more you exercise the more calories you burn. This helps you maintain your ideal weight and will give you a flatterable physique. Building gentle walking into everything that you do is one of the best ways to get the exercise you need. Take the stairs at every chance you get. When you are driving, park in the car park or space that is the farthest from your destination. If you have been sitting for a long period of time, be sure to take some time to get up and walk around. The benefits and pleasant effects are yours for the taking. All you need to do is start moving now on your own or even better walk communally with colleagues or friends! Start walking 30 minutes a day, 5 times a week!’
Leaflet design
Both leaflets carried the headline ‘Walking Campaign: Benefits of Walking’, a body diagram listing benefits (mental health, lifespan, heart health, weight, immunity, bones, balance) and the slogan ‘Walking not only changes your BODY, it changes your MIND, your ATTITUDE and your MOOD’. Wording transcribed from S1 and S2 Fig of the article.
Measures
Attitude towards the ad: six semantic differentials (irritating, boring, good, informative, objective, appropriate). Behavioural intention: likelihood of walking 30 minutes a day, 5 days a week (unlikely/likely, improbable/probable, impossible/possible, uncertain/certain). Advertising credibility: three items. Gender role identity: Gender Trait Index (8 femininity and 8 masculinity items).

How to cite

Baxter, K., Czarnecka, B., Schivinski, B., & Massaro, M. R. (2022). Masculine men do not like feminine wording: The effectiveness of gendered wording in health promotion leaflets in the UK. PLOS ONE, 17(10), Article e0273927. https://doi.org/10.1371/journal.pone.0273927

Open access under a Creative Commons CC BY 4.0 licence.

Related publications

All publications