The decline of homophobic attitudes in Britain and the rise of homophobic attacks; a critique of research methods
The Open University
This report evaluates two research methods used in psychology in the context of homophobic attitudes. The methods chosen are surveys and case studies. The research design is a literature review. The link between homophobic attitudes and attacks on LGBT+ people is explored. The report finds that the survey method is flawed due to its reductive approach. The case study method is found to be of limited use as it is not always generalizable. The link between homophobic attitudes and anti-gay attacks is weak. Social factors have more importance. There is a need for more mixed methods research with a focus on perpetrators and social contexts. There is also a need for research into other groups in the LGBT+ community other than gay men.
This report is the result of work commissioned by the British Government into the usefulness of research methods on the subject of homophobic attitudes. This work also seeks to explore the apparent contradiction between increasingly liberal attitudes towards LGBT+ people and an increase in the number of attacks perpetrated on them in recent years.
Research methods are evaluated within a pragmatic approach, which means a concern with how useful these methods are in solving practical problems.
The main focus is a critique of two frequently used methods in psychology. The methods chosen are:
- case studies.
The argument taken here is that the evidence produced from these methods is found to be weak with both methods giving only a partial picture. The relationship between this evidence of homophobic attitudes and homophobic attacks is then explored and it is found that there is no significant link. Homophobic attacks are caused by many factors with social context considered more important than attitude. There is a need for a mixed methods approach focused on action.
Let’s examine some background from which this work arises. According to the latest British Social Attitudes Survey 64% of people now say there is ‘nothing wrong at all’ with same sex relationships (Harding, 2017). Attitudes have become more liberal year on year.
In spite of these increasingly liberal attitudes, an increase in verbal and physical attacks on LGBT+ people has been reported. The proportion of LGBT+ people who have experienced a homophobic incident in the last year because of their sexual orientation has increased from nine per cent in 2013 to 16 per cent in 2017 (Bachmann and Gooch, 2017).
What is a homophobic attitude?
One meaning of homophobia is heterosexuals’ fear of being close to homosexuals as well as homosexuals’ self-loathing (Weinberg, 1972, cited in Koc, 2017). This definition does seem problematic. There are surely degrees of homophobia with varying intensity. Some psychologists reject this term and prefer ‘anti-gay’ (Koc, 2017).
The term attitude is even more troublesome. According to some psychologists, ‘attitude’ refers to a person’s evaluation of any aspect of the social world around them (Bernard et al., 2003). Attitudes are open to change as we have seen from the British Attitudes Survey (Harding, 2017). In addition, attitudes are not always consistent with behaviour (LaPiere, 1934). They tend to be context dependent (Potter and Wetherell, 1987). Again, this concept is simplifying a complex web of thoughts, feelings and behaviours.
The evidence in this report is based on a review of international literature on the subject of homophobic attitudes and then on homophobic attacks. Online searches of academic research studies were conducted using The Open University Library and Google.
- Critique of the Survey Method
The survey often takes the form of a written questionnaire. It is a method for collecting information as self-reported by individuals.
An example is Parker and Bhugra’s (2000) paper “Attitudes of British Medical Students towards male homosexuality”. This study assessed medical students attitudes to gay men. A 20-item questionnaire was completed by 428 students, 50% of whom were male. It was found that 10-15% held very negative views. The authors hold that this finding is worrying as a gay male patient has a high chance of seeing a homophobic doctor.
Let’s have a closer look at the method. The participants were medical students in London who were invited to fill out a one sided questionnaire after lectures. They completed the task there and then and left them on the bench. Their responses were anonymous. The questions consisted of 20 statements concerning male homosexuality, bisexuality and HIV/AIDS. Beside each statement the participants were asked to indicate their opinion by placing a tick in one of the boxes denoting: strongly agree, generally agree, unsure, generally disagree and strongly disagree. The statements phrasing, positive or negative with regard to attitude, was balanced across the statements. The evidence is translated into numeric form.
Firstly, the advantages will be examined. One advantage of the survey is high representativeness. This means the participants have the same characteristics as the whole population of people being studied. The Parker and Bhugra (2000) study has a large sample size, which suggests it is representative of all medical students in the UK. 428 students took the test. This implies strong evidence.
A second advantage is the attempt at objectivity. Researcher bias is minimised. In the Parker and Bhugra (2000) study the students are answering standardised carefully phrased written questions on their own. They are not influenced by the presence of the researcher or the responses of others. This makes the evidence appear trustworthy.
Now, the disadvantages will be examined. The first issue is that the survey method is reductionist. It is over simplifying very complex phenomena. In this example, Parker and Bhugra (2000) are using a questionnaire, which reduces the many facets of homophobic attitudes down to a series of statements. They are not capturing the nuance, the variability and the context of the individual participants’ attitude. For example, 15% of the students agreed with the statement: ‘Male homosexuals should not be employed in schools’. The researchers take this as evidence of extreme homophobia. But is it? All we can really know is that on a particular date at a particular time a participant put a tick in a box. We don’t know why he ticked this box. He may have limited direct experience of gay men. He may not hold anti-gay views about other aspects of LGBT+ life. The whole context of human interaction is missing. Maybe his attitude is not stable over time but varies according to the situation and the conversation he is having at the time. Attitudes may not be enduring thoughts, but constructed through interactions. They can be contradictory (Potter and Wetherell, 1987). This survey is giving us too narrow a picture.
The second disadvantage of a survey is lack of validity; it is not measuring what it says it’s measuring. In this example, the researchers say they are measuring homophobic attitudes. They claim to be getting to the personal opinions of the students. All they are really doing is measuring explicit attitudes, not the implicit ones that are of more interest. The explicit attitudes are what the students were willing to state publicly in an official survey, not what they are thinking, which we can never really know. They could be lying or at least modifying their views to fit in with what other people think. They are probably aware that homophobia is now frowned on. The private homophobic attitudes could be higher than 15 % or lower. We don’t know. We can only speculate. Attitudes can’t really be measured at all due to their constantly changing nature (Potter and Wetherell, 1987).
Now, moving the scope of our argument more widely, it seems there is a problem with the whole paradigm that this method is situated in. A paradigm means a way of looking at the world. The research is framed within this worldview. The survey is in the tradition of scientific enquiry, assuming there is an objective reality which can be measured and which follows logic. Many researchers argue against this. They hold that psychology is the study of complex human interactions, which people actively construct, in different contexts (Flyvbjerg, 2001). There may be no one truth but many subjective truths. Generalisations cannot easily be made on complex human mind states. Parker and Bhugra (2000) suggest that the survey indicates 15% of medical students are homophobic. This is presented as if it is a scientific fact. We cannot know this. We can’t say anything about the implicit attitudes, just interpretations of them, which may be in flux. This makes the survey evidence seem weak.
Finally, this study only concerns attitudes towards gay men. It can provide no evidence about attitudes towards different groups within the LGBT+ community. As attitudes are so variable there could be significant differences here.
It seems reasonable to conclude that the survey method is extremely flawed so the evidence it provides is not of any great use in solving social problems. It provides more of a starting point which other research could build on.
Critique of the Case Study Method
A case study is a detailed look at a particular phenomenon, an individual or group of people. It grew popular as a method as a reaction to the limitations of approaches like surveys. More particularly, case studies use observations and interviews (Flyvbjerg, 2001).
In order to critique this approach let’s look at an example of a case study on the subject of homophobia: “‘It’s Just Not Acceptable Any More’: The Erosion of Homophobia and the Softening of Masculinity at an English Sixth Form” by McCormack and Anderson (2010). This research examines the relationship between sexuality, gender and homophobia and how they impact on 16-18 year old boys in a sixth form in the south of England. The boys were found to have pro-gay attitudes and there were no instances found of homophobic language or behaviour.
Let’s have a closer look at the methods used. Two researchers were employed to observe the students in the common room, in lessons and in interest clubs at the sixth form for five months. They then researched attitudes through 17 semi-structured interviews. The data produced was coded thematically. The sixth form was in the south of England and was considered to be typical. The evidence is the statements of the boys and the notes of observations arranged into themes.
Firstly, the advantages of a case study will be examined. This method gives rich and complex data about one particular community, which allows us to feel we are getting a detailed picture of the interpretations of the participants. In the McCormack and Anderson study (2010), the boys’ attitudes can be captured with all their complexity as they discuss homophobia. The research is in depth for five months covering a wealth of interactions and responses. One gay student expresses how everybody is ‘cool with it’ and ‘nobody is bothered’ (McCormack and Anderson, 2010 p.851). The open nature of the research gets to this student’s own interpretation, it is not reduced by general statements like in the survey.
Another advantage of the case study is that the context is not ignored. It is part of the data. As Flyvbjerg (2001) argues it is not possible for psychology to proceed like natural science so this method offers a way forward, capturing a particular example. In this study example we get close to the constructed realities of the boys at this particular sixth form as they are observed in different situations. An example is a gay student is sitting alone on a coach and three boys go and sit with him for the rest of the trip. This is powerful evidence of pro-gay behaviour. They are not just saying it, they are doing it.
However, some psychologists would still argue there are disadvantages of the case study method. Firstly, for it to be really useful the case chosen has to be typical. Untypical cases cannot be generalised to the whole population of school children. In this example McCormack and Anderson (2000) are at pains to point out that the school is typical but this can be contested. It seems predominantly middle class and white and is in the south of England. As Stonewall have pointed out, homophobia may be at its most prevalent in the northeast of England (Bachmann and Gooch, 2017). The finding here of no homophobic language is not supported by other research in schools (Norman and Galvin 2006, Douglas et al., 1999), which suggests the study is not a typical case. However, this criticism does not have to be devastating. A series of such case studies in different locations around the country with different demographics would give varied data which, when looked at together, would give a detailed picture of homophobic attitudes in the UK.
Secondly, the problem of researcher bias is an issue for some psychological approaches. This means the participants are influenced by the presence of the researcher and behave and speak differently, and that the researcher brings their own preconceived ideas into the study. In the McCormack and Anderson (2000) study there is an attempt to blend in with the participants and live among them. However, they are not participants. The age difference must be apparent. It seems the students like the researchers. They will be concerned to express attitudes socially acceptable to the visitors. The investigators may be finding exactly what they want to find. In addition, they are coding and interpreting the information while classifying it into themes. They are being selective in the statements they pick out for comment. In this way, without reading the transcripts the reader is not getting all the data but a highly edited version of it. Due to the way the study is written up we are not given many examples of the students’ actual words. One such example is: ‘He wouldn’t keep at it long’, (McCormack and Anderson, 2000 p.851). This statement is referring to anti-gay epithets, which might be used by other students. This phrase suggests a concern with what other people think. Is it the case boys do not express anti-gay attitudes not because it is what they really think but because of the reaction of other people? We still do not know what the boys’ intrinsic attitudes are, only what they say and do in front of their peers. We are seeing an interpretation of the students’ views, not the students’ views themselves.
Moving more widely, some psychologists would criticise this research for still not being nuanced enough. Discourse analysis, where transcribed conversations or texts are examined in detail might give us more sense of how the students’ attitudes are very complex and contradictory depending on the context (Potter and Wetherell, 1987). There is not much sense of this here, no idea that attitudes constantly shift and change according to the subject, no capturing of the complexities of categorisation and the irrational and emotional nature of conversation as shown in more discursive studies (Coates, 2013). It seems all too neat and easy. Perhaps, the researchers were asking the wrong questions or not going deeply enough.
Finally, this study only gives information about attitudes towards gay men. It can provide no evidence about attitudes towards different groups within the LGBT+ community.
In conclusion, the evidence from the case study method is often weak. The study must be designed very carefully if it is to be helpful in solving social problems. The evidence provided in this particular study is not useful for understanding homophobic attacks.
- Is There a Link between Homophobic Attitudes and Attacks on LGBT+ people?
The weak link between attitudes and behaviour means attitudes research is not useful in finding out why homophobic verbal and physical attacks occur. Social factors are more significant.
In support of this view, Walters and Brown (2016) suggest there are multiple social, psychological and structural causes of homophobic crime. Attackers are influenced by the people around them, the environment, the conflicting ideas in their minds, and the inequalities in society. Attitude is only one factor and not the most important one.
Support for this comes from Williams and Tregidga (2013, cited in Walters and Brown, 2016), who suggest a number of contextual factors, including opportunity, proximity, intoxication, housing issues and existing relationships between the victim and the perpetrator all come into play. Furthermore, Chakraborti and Garland (2015, cited in Walters and Brown, 2016) have emphasised the role of threat, which results in some men policing the boundaries of male heterosexuality through violence, reinforcing their own masculinity. Again, support comes from Franklin (1998, cited in Herek, 1998) who interviewed men who had assaulted LGBT+ people who did not recognise themselves as particularly homophobic and reported no hatred towards this community. Factors such as gender norms, peer dynamics, youthful thrill seeking, and economic and social disempowerment were more important than attitude.
At present, there are many gaps in our knowledge in this area and further research is needed (Walters and Brown 2016).
Evidence from both surveys and case studies concerning homophobic attitudes should be treated with caution as it can be seriously flawed. Surveys can give too reductive a picture and case studies can suffer from researcher bias and cannot always be generalised. Attitudes research is not useful when exploring the reasons for the rise in homophobic attacks. Analysis of assailants’ speech could be more fruitful. A mixed methods approach is needed with data from surveys providing a starting point and then further research conducted in context.
- Conduct research into the causes and motivations of homophobic verbal and physical attacks, with an emphasis on perpetrators and the social contexts they find themselves in, using a mixed methods approach.
- Conduct research concerning behaviour and speech towards other groups in the LGBT+ community other than gay men.
- Ensure research is solution focused, paying attention to disadvantaged groups within society.
Bachmann, C. L. and Gooch, B. (2017) LGBT in Britain Hate Crime and Discrimination, London, Stonewall and YouGov (Online). Available at http://www.stonewall.org.uk/sites/default/files/lgbt_in_britain_hate_crime.pdf (Accessed on 3rd January 2018).
Bernard, M.M., Maio, G.R., and Olson, J.M. (2003) ‘The vulnerability of values to attack: inoculation of values and value-relevant attitudes’, Personality and Social Psychology Bulletin, vol. 29, no. 1, pp.63–75, London, Sage (Online). Available at https://doi.org/10.1177/0146167202238372 (Accessed on 10th January 2018).
Coates, J. (2013) ‘The discursive production of everyday heterosexualities’, Discourse and Society, Vol. 24, no.5, pp. 536-552, London, Sage (Online). Available at https://doi-org.libezproxy.open.ac.uk/10.1177/0957926513486070
(Accessed on 10th January 2018).
Douglas, N., Warwick, I., Whitty, G., Aggleton, P., and Kemp, S. (1999) ‘Homophobic bullying in secondary schools in England and Wales‐ teachers’ experiences’, Health Education, Vol. 99 Issue: 2, pp.53-60, Bingley, Emerald Publishing (Online). Available at https://doi.org/10.1108/09654289910256914 (Accessed on 10th January 2018).
Herek, G. M. (ed.) (1998) Stigma and Sexual Orientation: Understanding Prejudice against Lesbians, Gay Men, and Bisexuals, London, Sage (Online). Available at http://dx.doi.org.libezproxy.open.ac.uk/10.4135/9781452243818 (Accessed on 10th January 2018).
Koc, Y. (2017) DD801, Week 14 Interrogating Methods: Measuring Attitudes and Homosexuality, Milton Keynes, The Open University (Online). Available at https://learn2.open.ac.uk/mod/oucontent/view.php?id=1214131 (Accessed on 3rd January 2018).
LaPiere, R.T. (1934) ‘Attitudes vs. Actions’, Social Forces, Vol.13 (2), pp.230-237, Oxford, Oxford University Press (Online). Available at http://heinonline.org.libezproxy.open.ac.uk/HOL/Page?public=false&handle=hein.journals/josf13&page=230&collection=journals (Accessed on 3rd January 2018).
McCormack, M. and Anderson, E. (2010) ‘’It’s Just Not Acceptable Any More’: The Erosion of Homophobia and the Softening of Masculinity at an English Sixth Form, Sociology, Vol. 44(5) pp. 843-859, London, Sage (Online). Available at https://doi.org/10.1177/0038038510375734 (Accessed on 3rd January 2018).
Parker, A. and Bhugra, D. (2000) ‘Attitudes of British Medical Students Towards Male Homosexuality’, Sexual and Relationship Therapy, Vol 15, No.2, pp. 141-149, London, Taylor and Francis (Online). Available at http://dx.doi.org/10.1080/14681990050010736 (Accessed on 3rd January 2018).
Potter, J. and Wetherell, M. (1987) Discourse and Social Psychology, London, Sage.
Walters, M.A. and Brown, R. (2016) Causes and motivations of hate crime, Equality and Human Rights Commission, New York, SSRN (Online). Available at https://ssrn.com/abstract=2918876 or http://dx.doi.org/10.2139/ssrn.2918876 (Accessed on 3rd January 2018)