A naturalistic observation of gender differences in polite behavior when alighting from a bus

Dafni Markoulaki-Leontidou[1]

A naturalistic observation of gender differences in polite behavior when alighting from a bus



The current study took a sociocultural approach to examine the gender differences in politeness when getting off a bus. Previous research revealed that women smile more than men do in social settings. Further research provided evidence that women are more polite than men are. In this particular research communication methods were measured quantitatively for both men and women. The results showed that there was no statistically significant difference of politeness between men and women. In conclusion both men and women are similarly polite.



Wanting to investigate a communication issue, the study focused in gender differences. A naturalistic observation of politeness was conducted in Bath city. Researchers tried to distinguish gender differences in polite behavior when alighting from a bus. The data collected were analyzed quantitatively using a chi-square test. The results were discussed according to the sociocultural perspective. A review of the previous research on the general topic of gender differences and politeness is provided. Contingency table, chi-square test, materials used and the whole detailed project procedure are explained and also attached in the appendix.


The sociocultural perspective

‘Politeness’ is a major element of everyday interaction. The vast majority of literature studies since the foundation work by Brown & Levinson (1978) reveal that politeness phenomena are not only socially important but also in great psychological interest. More specifically, the task of this paper is to examine whether there is a gender difference in polite behavior. According to the sociocultural perspective different social experiences are responsible for the different social behavior between men and women (Cross & Madson, 1997, cited in Balliet, 2011). More specifically, the structure of society forms different social roles for each sex. This structure emanates from biological differences. Women are assumed to acquire a more domestic role, which implies a more interpersonal skill, whereas men are more related to power and status positions (Cross & Madson, 1997, cited in Balliet, 2011). So they have different social roles; women develop more interpersonal skills in comparison to men who develop more argentic skills.

Traditionally, politeness used to be interwoven as a feminine characteristic. Deutsch (1990) examined gender differences in polite behavior in relation to power status. The results revealed that women smile more than men whether they were in the power status position or not. Ten years later, Hall’s (2000) meta-analysis on different data confirmed the same hypothesis that ‘women smile more than men do’.

The idea of how to observe politeness in natural settings was based on the research of Gibbons (2008) who examined the interaction between male and female passengers with the driver on the minibus in Hong Kong. Gibbon found that women are more polite in comparison to men. He particularly examined the way of communication when people from different sex request the driver to make a stop. The idea for this particular research was to preserve Gibbons idea of examining politeness on the bus and make two changes on the design. First, passengers in England were not obliged to request a stop (stops are predefined), which means that there was no compulsory contact with the driver. So, this context is more constrained than Gibbon’s. Conversely, the new contexts provided place for the examination of non-verbal expression of politeness. Second, passengers were observed when alighting from the bus and not when they enter the bus as in Gibbons research. The scope of the research, as it was mentioned before and based on the relevant research, was to investigate gender differences in polite behavior. But the intuition was that there will be a difference. So the research hypothesis was that ‘There will be a difference in politeness behavior according to gender’. Because several conflicting studies revealed no significant difference between men’s and women’s polite behavior, a two tailed hypothesis was chosen. According to Hobbs (2003) who measured politeness by examining voice mails in Portugal, men were equally polite as women. In fact the findings of Hobbs are matching with the null hypothesis (Banister, 2007) here, that ‘There will be no difference in politeness behavior according to gender’. Literally that both men and women are equally polite.



While the researchers wanted to investigate the association between gender and politeness they employed an observation in naturalistic setting design. The researchers, psychology students at the Open University, analyzed data collected from the bus station in Bath. More specifically, politeness was measured by recording the interaction between passengers and the bus driver when alighting from a bus.  The method was selected because it enabled the researchers to investigate an association between the variables. The role that the observers adopted was that of a complete observer (Gold, 1958, cited in Banister et.al, 2007). The variables were categorical and the data measured at the nominal level. There were two categories for the variable ‘sex’ (‘male’ and ‘female’) and there were four categories for the variable ‘politeness’ (‘no-communication’, ‘verbal’, ‘gesture’, ‘both’).


The first coding for the variable ‘politeness’, before the pilot observation (Appendix 1), contained three categories (‘no communication’, non-verbal communication’, ‘verbal communication’). A table of how these categories first defined is presented below:

Βehaviours (Measurement of politeness)  
No communication Straight off the bus
Non-verbal communication –physical gesture Hand gesture, head-nodding.
Non-verbal communication–facial expression Smile
Verbal communication Thank you, thanks, see ya, see you later, buy, cheers, have a good day, ta, etc…

But the pilot observation (Appendix 2) revealed that many passengers used both non-verbal and verbal communication. Furthermore, it was clear that if two different categories for non-verbal communication were used, then the number of participants could not be measured since the categories in the variable ‘Politeness’ were not mutually exclusive. For these reasons a simplification of the original categories was decided. The final categories were:

Behaviours (Measurement of politeness)  
No communication Straight off the bus
Non-verbal/Gesture communication Hand gesture, head-nodding, smile
Verbal communication Thank you, thanks, see ya, see you later, buy, cheers, have a good day, ta, etc…
Both ( Verbal + Non-verbal / Gesture communication Hand gesture+ thank you, smile + buy, and all possible combinations. 


Two hundredth twenty eight people randomly selected were observed. No one of them was needed to be informed to provide consent nor about the purpose of the research because of the design of the project. The vast majority did not understand what was being observed. Because they were not asking to participate rather they were just observed in a naturalistic setting there was no need for any incentive to be used. The only limitation that the researchers followed according to the people being observed was that everyone was observed apart from children. So the interest was focused in particular behavior of the adults. Generally, the final sample contained 88 males and 140 females.


One clipboard and a pen were the main materials used. The clipboard was useful because it enabled the researchers to write their observations more easily and accurately in the up write position.


The exact position for the observation was the bus station in Bath city in England. Researchers were given permission by the manager of the station to stand in the frond of each bus when it approached the bus station. For safety reasons researchers were told to wear safety jackets provided by the manager of the station.  The routes and the destinations of the bus were not taken into account. The researchers then observed the interaction between each adult passenger with the bus driver when they left the bus. More specifically, they observed whether there was no-communication between them, whether there was a gesture, a verbal communication or both. Using their clipboards researchers wrote an ‘n’ for ‘no-communication’, a ‘v’ for ‘verbal’, a ‘g’ for ‘gesture’ and a ‘b’ for ‘both’. The four observers after having tested their inter-observer reliability (Αppendix 3) in the pilot observation they did the previous day of the final observation, there were divided in pairs were each one of them observed either the men or the women for the first five buses arrived. Later, they changed their focus and the one who at first observed women in the next five buses he observed the men. So every observer had observed in the end of the study both men and women.


Contingency Table

No-communication Verbal communication Gesture communication Both (non-verbal          and verbal communication) Total
Male     23 (23.2)      8 (10.0)       7 (5.8)     50 (49.0)   88
Female     37 (36.8)     18 (16.0)       8 (9.2)      77 (78.0)   140
Total      60      26       15       127   228

The research hypothesis was that there will be a difference in politeness according to gender. The results were analyzed using a chi-square test (χ=1,118, df=3, p=0.773, Gramer’s V=0.773) (Appendix 4). As the analysis doesn’t revealed a statistically significant association, the null hypothesis was accepted which means that there will be no difference in politeness behavior according to gender. Both men and women are equally polite.



The results of the observation showed that there was no difference in polite behavior according to gender. Both men and women were equally polite towards the bus driver. One interesting found was that more women than men totally ignored the bus driver when alighting from the bus. The label ‘no-communication’ was thought to be a sign of an impolite behavior, so the fact that more women fall in that category was at least unexpected according to the previous research such that of Gibbon (2008). This difference though, is balanced from the fact that more women than men fall in the category of ‘both’, which means that apart from a verbal communication with the driver made also a gesture. The category of both was thought to be the more polite type of interpersonal communication a founding that if it was isolated from the rest of the study results totally confirms Gibbon’s claim that women are generally more polite than men.

According to the sociocultural perspective learning is not only a matter of education. Instead as Crook & Light (1999) explain, we need to focus “in the particular institutional and cultural contexts in which learning occurs”. People learn behaviors when they interact with others, learn social roles according to the culture they grow up a process which is called enculturation. More specifically enculturation is “a process through which people adopts specific cultural practices and act in accordance with cultural norms” (Littleton, 2007, p. 211). So, from that view the results from the particular study apart from sawing equal polite manners between men and women they reveal many conclusions for that British society. As it can be assepted from the above, British society lacks dichotomies in gender politeness behavioral roles.  Indeed, Gibbons measurements had taken place in Hong Kong a generally conservative society with a bit different culture from the west ones. Moreover, Hobbs foundlings from Portuguese people are more close to the ones found here which partially confirms the role of culture in politeness behavior. Though biological origins traditionally formed specific behaviors between the two genders, enculturation as a conscious or unconscious procedure in a great degree determines gendered social behavior and of course polite behavior too.

Based on the above, the null hypothesis was accepted because in the first place it was thought more possible to find a difference in politeness according to gender something that never proved. This happened first because cultural difference between previous research and the particular one were underestimated. Most research which previously claimed women to be more polite than men were measured east societies and not west ones as it was done here. Another reason though, might have to do with the procedure of the particular experiment which might have worked as a confounding variable to the results. The fact that passengers who examined know that they were observed, since the researchers were standing in front of the bus with their clipboards. Although, they did not know what the observers measured, only the fact that they knew they were observed may force them to behave differently.

In conclusion, the results of the naturalistic observation reported here offer some support for the Hobbs founding and for the sociocultural perspective’s claim of enculturation. However, future research conducted in the area of politeness should attempt to confirm these results by studying other cultures and observing if there would be any difference in the particular results when the driver instead of being a man is a woman.



Banister, P., et. al. (2007). Testing Hypotheses (Eds.) Exploring Psychological Research Methods. Milton Keynes: The Open University.

Brown, P. & Levinson, S. (1978). Universals in language use: politeness phenomena. In E. Goody (Ed.). Questions and Politeness (pp.53-311). Cambridge: Cambridge University Press.

Crook, C. & Light, P. (1999). Information technology and the culture of student learning. In Littleton, K., Toates, F. and Braisby, N., (Eds.) (2007). Three approaches to learning. Mapping Psychology. Milton Keynes: The Open University.

Cross, S.E.& Madson, L. (1997). Models of the self: Self-construal and gender. Psychological Bulletin, 122: 5–37. In Bulliet, D., Mackfarland, S.J., Li, N.P., Van Vugt, M. (2011). Sex differences in cooperation : A meta-analytic review of social dilemmas. Psychological Bulleting, vol. 137, no. 6: 881-909.

Deutsch, F. M., Coll, H. & Hadley, S. (1990). Status, Sex and Smiling: The effect of role on smiling in men and women.  Personality and Social Psychology Bulletin, vol 16, no. 3: 531-540.

Gibbons, J. (2008). Mitigation on the minibus. Journal of Asian Pacific Communication, vol. 18, no 2: 157-165.

Gold, R.L. (1958) Roles in sociological field observations. Social Forces, vol. 36: 217-23. In Banister, P., (2007). Testing Hypotheses (Eds) Exploring Psychological Research Methods. Milton Keynes: The Open University.

Hall, J.A., Carter, J.D., & Horgan, T.G. (2000). Gender differences in non-verbal communication of emotion. In Fischer, A.H., (Eds) Gender and Emotion: Social Psychological Perspectives, pp. 97-117. Cambridge University Press, vol. 6: 331.

Hobbs, P. (2003). The medium in the message: politeness strategies in men’s and women’s voice mail messages. Journal of Pragmatics, vol. 35(2): 243-262.

Littleton, K., Toates, F. and Braisby, N. (2007). Three approaches to learning. Mapping Psychology, Milton Keynes: The Open University.



APPENDIX 1: Pre- Pilot Thoughts

Issues identified:

a)       How we are going to ensure that we all were observing the same people. We will consider this during the pilot, checking how people get off the bus, i.e., one at a time or more.

b)      We agree that we will not be observing children, purely young adults and older.

c)       Whether we will be observing from on the bus or outside. We will use our journey into Bath on the bus to consider the benefits of staying on the bus to observe and pilot 5 minutes of observations from a) the bus stop and b) from a vantage point behind the bus driver.

d)      We have decided that if the observations are from the outside to locate ourselves at Bath bus station where the frequency and numbers of participants will be greater.

e)       We are targeting local route bus services rather than open-top tourist buses to give a broader range of cultural considerations.


Appendix 2: During the Pilot Observation Thoughts

Material used: clipboard, paper and pen

On the journey to the bus station

a)       We considered the use of a less-obtrusive form of measuring than the clipboard, eg. note book

b)      The group were situated on different seats on the bus, the person sat directly behind the driver having the best view and able to see or hear if the participant smiled or had any verbal communication.

a)       Observed from the bus stop in front of the bus.

b)      Decided to simplify the categorization process to only have a ‘no communication’ category, one non-verbal communication category and another category for both verbal & non-verbal. This enabled us to retain details about how many participants we had measured, as opposed to when we gave a score for every gesture. Thus, equal measure of men and women. This changed the method of data analysis to possibly that of chi-square, should we obtain the minimum of 5 participants in each category.

c)       Enquired about busiest route buses to enable a good flow of participants.

d)      Looked at positioning outside the bus which enabled clearest view. Due to the nature of the bus station layout (health & safety) we obtained verbal consent to be able to stand the other side of the locking doors when the buses come in.

e)       We then went to observe other bus stops in town to obtain our inter-reliability rating and decided on pilot of 10 male & 10 women.

f)       We adjusted the way we collected our data to a more efficient method. Using symbols, ‘N’ for no communication, ‘V’ for verbal communication, ‘G’ for non-verbal communication gestures and ‘B’ for people who used both ‘G’ & ‘V’ together. We also decided we would measure men first and then women so that we would not have to let our eyes leave the page whilst we scribed.

g)       We negotiated the order that data was collected in order to compare findings for inter-observer reliability.

h)      Positioning of observation was agreed.

i)       We discussed whether the categories could be simplified further to communication/no communication but decided it would lack the richness of detail the other categories offered.

j)       Our tutors agreed that once we had established inter-observer reliability we could observe individually on the actual day, we also notified them of the development of our project and the consequent change in analysis from our original proposal.

k)      From a further 20 minutes at the campus bus stop, we clarified that if a person were to give a verbal communication without a facial gesture but turning their head & giving eye contact to the bus driver it would be counted as ‘B’. We also noted that at the campus there are two types of buses, those with doors by the driver and others with doors mid-bus! However, the buses in Bath main bus station are all the same with only one exit by the driver.


Coding details of behaviors to be observed


‘N’ No communication Straight off the bus

‘G’ Nonverbal communication         Physical gesture, hand gesture, head-nodding, friendly facial expression

‘V’ Verbal communication      Thank-you, thanks, see ya, see you later, bye, cheers, have a good day, ta, etc…

‘B’ Both verbal and nonverbal         This included eye contact given to the bus driver when head-turning and saying thank you.

The pilot session took 1 hour 50 minutes in total and a further 20 minutes when returning to campus.


Appendix 3: Inter-observer Reliability

Inter-observer reliability table

Participants Observer 1 Observer2 Observer 3 Observer 4 Agreement
1 B B B B
2 B B B B
3 B B B B
4 B B B B
5 V B N N
6 N B B B
7 N N N N
8 N N N N
9 N N N N
10 N N N N

Inter-observer reliability showed 80 % reliability on collected data


Appendix 4: Chi-Square Test and Symmetric Measures

Chi-Square Tests



Asymp. Sig. (2-sided)

Pearson Chi-Square




Likelihood Ratio




Linear-by-Linear Association




N of Valid Cases


0 cells (,0%) have expected count less than 5. The minimum expected count is 5,79. 

Symmetric Measures


Approx. Sig.

Nominal by Nominal Phi



Cramer’s V



N of Valid Cases




[1] With the collaboration of Melanie Harper, April Collinge.