Educational Psychology and Q Methodology
There has been an active interest in Q Methodology in Educational Psychology at Sheffield University since 2005. Here I provide an overview of the approach, the experience of Q sorting and examples of Doctoral work completed by trainee Educational Psychologists at Sheffield University. The following is drawn partly from my own doctorate (Hughes 2012).
Q in brief
Q methodology is unusual in a number of respects. In general terms, it is both quantitative and qualitative and although it is more than 80 years old, few have heard of it. Three features in particular can be said to characterise Q methodology.
William Stephenson developed Q at a time when factor analysis was being used (by Spearman, for example) to explore correlations between test scores across populations of people. Individual differences related to intelligence, for instance were measured. Factor analysis enables comparison of weight, height and other human traits to be achieved so that assertions about a population might be made. Conclusions might include the idea, for instance, that in a population, very tall people also tend to weigh more. Any differences between variables relate to the whole population. In contrast, Stephenson sought the views of people and then applied factor analysis to their responses. In so doing he was able to explore subjective opinions in relation to a topic. This approach then, emphasises individuals measuring rather than being measured. People are correlated instead of tests.
Contrasted with conventional factor analysis then, in Q, people are the variables as opposed to test scores which provides an understanding of differences between individuals themselves rather than between the variables associated with them. This means that firstly, data has to be gathered in an appropriate form that allows this kind of by-person analysis, namely in the form of the first unusual characteristic of Q, the Q sort.
The Q sort is a method that requires a participant to sort items according to some kind of criterion, such as the degree to which they agree or disagree with, like or dislike an item. Items can be anything that can be sorted but are usually statements written on cards. Items are placed in a position on a grid consisting of columns of different heights, which in total, describe the shape of a normal distribution curve (the outer columns are shorter than the middle columns). It makes no difference where in a column an item is placed, but moving an item to the right or left is determined by how much a participant agrees or disagrees (for instance) with an item. When asking participants to sort items, there is an important distinction between valence and salience. Items are not simply ranked from right to left in descending order of value to the sorter. Instead, when the Q sort is completed, the items are arranged so that they spread out from the middle column to the left and right hand outermost columns with increasing salience. The outer columns thus contain items, about which the participant feels most strongly in contrast to the middle column, where ‘neutral’ items are placed, or those that the participant is not particularly bothered either way about.
Fig 1 Participant engaged in a Q sort using a grid (I based figure 1 on my photograph of my daughter in her teens-a willing participant)
The second unusual or characteristic feature is Q factor analysis. This involves Q sorts being compared with each other so that similarities and differences lead to the identification of factors. Each Q factor represents a pattern of sorting which forms the basis of a participant’s subjectivity or their viewpoint. In Q, a factor represents a particular level of statistical correlation for the assertion to be made that, although the participants providing Q sorts for it differed in certain respects in relation to the way that they sorted the items, their Q sorts were similar enough for them to subscribe to a pattern. This pattern is called a Q factor.
A third unusual aspect of Q is the interpretation of the Q factors. Each of the patterns are scrutinised so as to search for descriptions which enable the factors to be expressed. Such expression can be regarded as a viewpoint which articulates the strongest areas of agreement and disagreement when compared with, or relative to, the other viewpoints.
The Q sort experience
In November 2005, John Bradley’s presentation of his research into the views of young people on going to University (Bradley, 2005) inspired me to start using card sorts (as opposed to Q sorts) with children and young people. At this stage, without wishing to engage Q methodology, I designed and presented tasks to young people that required them to sort cards. I started to explore the response of children and young people to completing a questionnaire, compared with responding to the same items, presented as a card sort activity. One boy, for instance, told me that a questionnaire felt more like he was ‘not welcome...like you’ve done something wrong’. I liked the feel of a card sort and as an EP, its potential for engaging children and young people in a very different way to an interview. I became interested in developing a card sort activity as a way of trying to determine which stage on the six stage Trans-Theoretical model of change (Prochaska et al, 1994) a youngster might be situated. As a potential area of enquiry, I found this satisfying in the way that it might inform practice by developing an assessment tool as well as by understanding better how what children and young people say about their behaviour, might be understood and used to engage them effectively in considering change.
I created an opportunity to complete a Q study with all members of the Educational Psychology Service in the Local Authority (LA) in which I work-my first experience of completing a Q sort which I found instructive. I noted that I ‘felt I knew when I had completion’ and that there was a process of continual comparison. Some of the statements provoked quite deep reflection and I realised, at first hand, that the wording of the statements was crucial. I found myself identifying a theme and then clustering associated statements. Other colleagues’ comments led me to think that doing a Q sort was not something that could be done passively if one wished to preserve one’s integrity. Some colleagues felt constrained by the shape of the grid, although, like me, one EP said, ‘feels like I’ve finished....it matters where it goes’ (with respect to placing the statements on the grid).
My daughter completed a Q sort shortly before her 16th birthday and wrote the following:
‘The Q sort would be good for people that don’t like to talk about themselves or find it difficult. Also it’s easier than writing down information about how you’re feeling because the thoughts are there. And because they are thoughts that relate to the person but aren’t actually volunteered by that person, that makes it easier for the person to describe how they are feeling but not feel as if they are giving too much away or finding it difficult.’
From the Q-Method list, a thread (September 2007) developed around the theme that Q sorting is a valuable exercise in itself. One contributor wrote about having used Q to identify what was important in an after school education programme, stating,
In the discussion ... each of us spent lots of energy championing our own causes - and very little energy listening to the arguments of others, but in the sort activity we were forced to read and think about the points that everyone else brought up. Almost to a person, the comments afterwards were along the lines of ‘whatever statement I was reading at the time seemed the most important to me’. Only through the Q sort activity did we really listen to each other (Downing Wilson, D. 19/09/2007, 00:02:07).
Brown’s response to the Q-list described the procedure as one that forces the Q sorter to take various perspectives as ‘value clarification through representative exposure’ (see Brown, 1994). A number of contributors commented on the value of Q sorting in making personal views and beliefs more explicit to the Q sorter and using the method at planning and training events.
Examples of Doctoral (DEdCPsy) students work at Sheffield University
This approach has appealed to a number of our students training to become Educational Psychologists on the doctoral training programme. Recognizing the value of the means by which different views can be explored and understood, they have investigated a range of topics of concern to young people.
Rachel Massey investigated the views of children with hearing impairments regarding the support they had from their teaching assistants. Rachel’s work is used extensively in a most helpful and thorough introductory guide by Watts and Stenner (2012). Hearing impaired youngsters held different views about what was important to them. For some, it was an adult who used sign language. For others, playing down differences between their own needs and their mainstream peers was important. One aspect of Rachel’s work that I found particularly valuable was her use of ‘child-friendly’ narratives based on her findings, that she used with young people in order to understand better what was important to them in establishing a support contract with their Teaching Assistants (Massey, 2010).
Carol Plummer was keen to challenge the dominant discourse surrounding ‘young carers’, namely that this is a position to be avoided, a ‘bad thing’. In contrast, she found four distinct factors within young carer participants: Factor 1: ‘We’re proud and positive. We feel included and well supported but don’t like being singled out; Factor 2: ‘Caring’s just what we do. We feel mature, but are unsupported and misunderstood’; Factor 3: ‘‘Parentified’ and wanting to care, but we need people to recognise that we’re struggling and worried’; Factor 4: ‘Being mature doesn’t stop us from worrying, although we’re supported, especially in school’, (Plummer, 2012).
Having spent a number of years working in China teaching English, Richard Stollery’s interest lay in understanding the views of young second language learners, regarding what they found helpful about the strategies used by adults to support them in school (Stollery, 2013). Richard identified four viewpoints and discussed the implications for professionals, the need to provide personalised support and the importance of staying in touch with the viewpoints of individual students with EAL in school.
Arguing that existing research into cyberbullying has tended to utilise surveys in order to understand the extent to which cyberbullying is experienced by young people in society, Francine Wint explored the views of young people in relation to what bothered them about Facebook. She wanted to minimise ‘the potential for researcher bias’ and maximise ‘the opportunity for young people to give their personal account’ (Wint, 2013). She found four viewpoints: ‘I want to protect others’; ‘I am worried about the dangers on Facebook’; ‘I know who I am and what I’m doing’; and ‘I don’t want any trouble’. Taking these back to young people enabled her to explore possible actions that young people themselves, schools and EPs could take.
Noting that previous research in the area of meeting the inclusive needs of female Muslim pupils had explored the views of parents and head teachers, April, working with young Moslem girls as ‘co-researchers’ investigated their views of inclusive education in a predominantly white area (Frearson, 2013). She found four viewpoints (one of which was bipolar), an individualist approach to inclusion in a school that creates a sense of belonging based upon valuing and respecting Muslim identity; a collectivist approach to inclusion in a school that supports the process of social integration; (f3 positive): Muslim others as significant to feeling included; (f3 negative): A school that provides an Islamic classroom environment and curriculum extending inclusion to activities and events outside of school.
I have worked as an EP since 1988. Related to Warren’s question concerning the character of the social identity that we ask children to take on when they participate as ‘researchers’ rather than as the objects of classroom investigations, (Warren, 2000), my own doctoral work was also concerned with the position of young people as researchers and has been described elsewhere (Hughes, 2012, 2014). My research question was broadly concerned with young researchers’ experience-what it was like, as opposed to why it was like that or how it should be.
I worked on a project with young people, where I had engaged them as ‘co-researchers’. By drawing on our experiences, initial statements came from them and included :
- Are given the opportunity to take on additional responsibility
- May not fully understand what is going on
- Are accepted as an equal, valid member of the team
- May feel under pressure to complete
- Enable me to see how more experienced people work
- Have a lot of responsibility
- Have to plan very carefully
- Are valuable in offering and sharing new ideas
- Have to share ideas together
A set of items (Q-set), consists typically of between 40 and 80 items (Watts & Stenner 2005; Stenner et al, 2008). I continued to generate statements, aiming to represent the ‘co-researcher’ experience and developed a set of 59 statements which reflected the perspectives of the young people I was working with and the themes found in the literature. For instance, ‘the biggest ethical challenge for researchers working with children is the disparities in power and status between adults and children' (Morrow & Richards, 1996, p. 98), led to these statements:
-experienced frustration over the limits which were placed on them by the adults
- found that power-sharing (or democracy) between adults and young people in research was possible
- found that it was clear that the adults felt that they knew best, and
- found that adults were willing to adopt a learner role
These statements relate to responsibility and decision-making:
- took responsibility for sorting out the ethical issues in the research
- were much more than just assistants to the adults, and
- made some really important decisions
whilst these were examples of statements related to outcomes:
- experienced a different way of adults and young people learning together
- helped to produce better outcomes than work produced by adults alone, and
- were able to learn how more experienced people work
Participants were asked to arrange the statements according to how much they agreed with them, comparing each one with the others. Statements were designed to complete this lead statement: In my experience, young people working with adults on research……..
Q-sort data collected from participants, was entered into PQMethod (see resources) for analysis which led to a solution with a number of the young researcher participants loading on five factors. In other words, the analysis revealed that there was enough similarity between the Q sorts to indicate that there were five distinct patterns (factors) or viewpoints and the young people’s individual Q sorts that were most closely associated with each of them.
The findings from my research can be related to aspects of the literature. For instance, there appears to greater variation in young researchers coming up with the idea for research than that suggested by Bucknall (2012), stating perhaps an ideal position when she writes that, in one study, ‘The young researchers…were unanimous in confirming choice of topic as a crucial motivational factor’.
The results from this study indicate a number of different viewpoints held by young people, based on their experience of having worked as young researchers. In my work it is clear that the situation is more complex than being able to conclude that young researchers are a ‘good thing’.
Bradley, J. (2005) Q-Methodology. Presentation to EdD course, Sheffield University, November, 2005.
Brown, S.R. (1994). Representative exposure and the clarification of values. Policy Sciences Institute, Yale University School of Law, New Haven, CT. (Read at a meeting of the Policy Sciences Institute, Yale University School of Law, New Haven, Connecticut, 28-30 October 1994).
Bucknall, S. (2012). Children as researchers in primary schools: Choice, voice and participation. Abingdon, Oxon [England: Routledge.
Downing Wilson, D. (2007) Q-Method list (QMETHOD@LISTSERV.KENT.EDU) 19/09/2007, 00:02:07).
Frearson, A. E. (2013). A Q-methodological study to explore Muslim girls’ viewpoints around how a secondary school setting can promote and support their inclusion. Research thesis (Educational and Child Psychology), Department of Educational Studies, University of Sheffield, UK.
Hughes, M. (2012) Researching Behaviour : A Q Methodological Exploration Of The Position Of The Young Person As Researcher. Doctor of Education (Educational Psychology) Department of Educational Studies
Hughes, M. (2014) What might adults learn from working with young researchers? Chapter 14 in Westwood, J., Larkins, C., Moxon, D., Perry, Y. & Thomas, N. (2014). Participation, Citizenship and Intergenerational Relations in Children and Young People's Lives: Children and Adults in Conversation. London:Palgrave Macmillan.
Massey, R. (2010) A Q-methodological study to investigate adults’ role in supporting the social and emotional well being of children and young people who are deaf. Research thesis (Educational and Child Psychology), Department of Educational Studies, University of Sheffield, UK.
Morrow, V. & Richards, M. (1996). The Ethics of Social Research with Children: An Overview. Children & Society, 10, 2, 90.
Plummer, C. (2012) Who Cares? An exploration, using Q methodology, of young carer and professionals’ viewpoints. Research thesis (Educational and Child Psychology), Department of Educational Studies, University of Sheffield, UK.
Prochaska, J. O., Norcross, J. C. & DiClemente, C. C. (1994) Changing for Good: The revolutionary program that explains the six stages of change and teaches you how to free yourself from bad habits. New York, W. Morrow.
Stenner, P., Watts, S. & Worrell, M. (2008) Q Methodology, Chapter 13 in Willig, C. & Stainton-Rogers, W., (2008) The Sage Handbook of Qualitative research in Psychology. London, Sage.
Stollery, R. L. (2013) A Q Methodological Study of the Support Valued by Students with English as an Additional Language. Research thesis (Educational and Child Psychology), Department of Educational Studies, University of Sheffield, UK.
Warren, S. (2000) Let’s do it properly : inviting children to be researchers, Chapter 10 in Lewis, A. & Lindsay, G. (Eds) (2000) Researching children’s perspectives Buckingham, Open University.
Watts, S. & Stenner, P. (2005) Doing Q methodology: theory, method and interpretation. Qualitative Research in Psychology 2, 67-91
Watts, S. & Stenner, P. (2012) Doing Q methodological research : Theory, method and interpretation. London, Sage.
Wint, F E (2013) 'Am I bothered?' - Using Q methodology to explore what bothers young people on Facebook. Research thesis (Educational and Child Psychology), Department of Educational Studies, University of Sheffield, UK.
Operant Subjectivity is the official journal of the International Society for the Scientific Study of Subjectivity (ISSSS).
https://qmethod.org/issss/ takes you to a number of different resources, including Q listserv, the official e-mail information and discussion list for ISSSS. It is a moderated forum for the exchange of information related to Q Methodology.
PQMethod is a statistical program tailored to the requirements of Q studies. You can download the (free) programme here
This animation, intended for a general audience, provides a short non-technical introduction to the methodology, outlining the different stages involved in conducting a Q study.
Useful Methodological references
Brown, S. R. (1980). Political subjectivity: Applications of Q methodology in political science. New Haven: Yale University Press.
McKeown, B. and Thomas, D. (2013) Q Methodology, 2nd edition London, Sage Publications.
Stenner, P., Watts, S., & Worrell, M. (2008). Q methodology. In Stainton Rogers, W. & Willig, C. (Eds.), Handbook of qualitative research methods (Chapter 13, pages 215-239). London: Sage.
Watts, S. & Stenner, P. (2005). Doing Q methodology: Theory, method, and interpretation. Qualitative Research in Psychology, 2, 67-91.