Of Methods and Methodology 6 | Practical Research Considerations

Sociologists do research for a wide range of reasons and in this post we’re looking at a range of practical research considerations relating to, firstly, choice of topic and secondly, choice of method.

As luck would have it (you didn’t seriously think I planned this stuff, did you?), this all fits neatly into my “5 Things You Need To Know” patent-not-pending revision technique…

choice of topic

Decisions about what to study may be influenced by a range of personal and institutional factors:

1. The interests of the researcher, for example, are likely to be a key influence on any decision about what to study., one that reflects their areas of expertise and specialism. To take a slightly extreme example perhaps, the Glasgow Media Group – and Greg Philo in particular – have specialised in research into mass media bias for over 40 years – from “Bad News” (1976) to “Bad News for Labour” (2019).

2. A second area of influence involves things like current debates and intellectual fashions: the popularity of different research topics, for example, wax and wane for a range of reasons, not least being the availability or otherwise of research funding – an issue we’ll address in a moment.

One driver of research choices in this respect, at least in the UK, is that institutional, departmental and personal funding in universities is sensitive to the popularity of a piece of research. An important way this is measured is in terms of how many other researchers around the world cite an individual’s research. The more the citations, the greater the measured popularity and the higher the level of future research funding for a university department.

A research topic that is currently popular and / or intellectually fashionable may stand a greater chance of attracting citations because the pool of interest – among other researchers, the media and general public – will be that much larger than for a much smaller, more niche, research topic  (Southerton et al’s (1998) “Research Note on Recreational Caravanning” being a personal favourite).

3.  Funding is one of the more prosaic – but nevertheless hugely important – practical considerations in relation to topic choice. Research, in simple terms, costs money – whether, like Cant et al (2019), you’re emailing a questionnaire to sociology teachers to explore their opinions on the status of sociology in English schools, co-ordinating a group of academics and field researchers in a lengthy study of religion and spirituality (Heelas and Woodhead, 2004) or simply, like Ferrell (2018) or Venkatesh (2009), engaging in extensive and lengthy forms of participant observation.

While finding the money to fund a piece of research that may last anything from a few weeks to, in Venkatesh’s case around 10 years, may dictate the researcher’s choice of topic – if the funding’s not available, it’s unlikely to be studied – a further dimension are the questions:

Who pays?

And why?

Those who commission and pay for sociological research, from universities, through charities and private Think Tanks to government departments, are likely to want – and in some cases demand -an important say in the ultimate choice of topic to be studied. In both the UK and USA, for example, the trend over the past 30 – 40 years has been to commission sociological research designed to help government policymakers make decisions. If your choice of topic (and method…) doesn’t fit with this brief or aid in this process it may be harder to attract funding.

4. What a researcher chooses to study (as well as how they choose to study it) will be affected by the time they have available for research. This covers considerations like:

the level and length of funding available: if funding is only available for a relatively short period this may determine what the researcher can research within the given time.

the method/s used: email questionnaires, as we’ve suggested, can be quicker and easier to administer and decode than something like ethnographic research (participant and non-participant observation, for example). Research that uses existing secondary data may have fewer time constraints than research that involves collecting in-depth primary data.

the scale and scope of the research topic: large-scale longitudinal research, such as the Peterborough Adolescent Development Study (2002 – 2010) designed to understand how families, schools and communities shaped young people’s social development, can be time-consuming to carry-out and manage. The principal investigator Per-Olof Wikstrom had to manage a diverse team of 30+ student investigators, academic collaborators and so forth over a 10-year period.

5. A further set of practical research considerations revolve around two related questions: access and respondent co-operation.

In terms of the former the researcher may not be able to get access to the people they want to study. Conventionally this is a problem associated with various types of ethnographic research involving overt and / or covert observational methods. If a researcher, for example, doesn’t share the characteristics of a group they want to covertly study (such as a female researcher trying to study an all-male group) this makes the research impossible.

In terms of the latter, some groups may simply decline a researcher’s efforts to study them, as Wallis (1976) found when he was refused access to current members when he tried to study Scientology (although he subsequently adapted his research to focus on former members of the Church). While Wallis’ experience was one of out-right rejection, it’s often the case that respondents may not be contactable for more-prosaic reasons.

Cant et al (2019), for example, found that even where contact attempts were electronic rather than face-to-face “not all schools were contactable (after three attempts), and sometimes school policy prohibited the sharing of teacher email addresses, or the request that follow-up emails be sent to elicit the information were sometimes unanswered”. While they felt a “20% response rate” for their questionnaire research was acceptable, not everyone might agree.

Although, for the sake of explanation, we’ve looked at these as discrete (i.e. separate and self-contained) categories, it’s important to remember that in the “real world” of sociological research such categories frequently and necessarily overlap. The personal interests of a researcher may, for example, coincide with current intellectual fashions and trends which, in turn, may attract greater and more-frequent levels of funding.

choice of method

As with topic choice, sociological research involves confronting and resolving a range of practical issues and problems relating to a researcher’s choice of method (or methods), many of which cover the same areas at which we’ve just looked:

1. The time available for the research project is a significant factor, given that some methods are more time-intensive than others in terms of both collecting and analysing data: various forms of observation – from participant through to non-participant – are likely to be far more time-intensive than something like a “postal” (or, as is probably more-likely now, email) questionnaire.

To give you some idea of time, Whyte (1943) spent around four years gathering extensive information about the behaviour of one youth gang in Boston, America while Venkatesh (2009) took around 10 years documenting the lives of the black residents, gang members and non-members alike, in a small area of Chicago.

Time is not, however, simply a factor related to particular methods. Longitudinal studies that may or may not involve a mix of methods (some forms of longitudinal study simply involve tracking and interpreting official statistics, for example) take place, by definition, over a long period and while a researcher may not devote all their time exclusively to a single study covering 20 – 40 years, they will necessarily revisit it from time to time.

Similarly, where a mix of methods (triangulation) is used to satisfy different types of research question within the same topic, time may be an important factor around what can or cannot be researched. Heelas and Woodhead’s (2004) ambitious study of spirituality in Kendal, for example, used a wide range of methods – from interviews to questionnaires to personal observations – in, at times, a limited way. Rather than interview / question a representative sample of individual Kendal residents about their spirituality, for example, they conducted a “street survey” that involved selecting particular streets that were broadly representative of different areas in the town and then surveying the views of individual residents within those areas.

2. The Kendal Project raises another important issue in relation to choice of method in the sense that this may be influenced or determined by the range of resources – particularly physical resources – available to the researcher. Heelas and Woodward, for example, needed 25 people, one for each Church, in order to do a physical head count of everyone who attended on a selected day and they were able to recruit students from their nearby University (Lancaster) to achieve this. Without this resource this aspect of the research could not have been carried-out.

3. Some topics lend themselves more easily to one type of method than another:

On a general level, where the objective is to create reliable data, establish statistical relationships and the like, quantitative methods are clearly more-useful and appropriate. If, on the other hand, the objective is to carry-out some form of ethnographic study that digs-into people’s feelings, emotions and perceptions, such as Diken and Laustsen’s (2004) analysis of tourist behaviour in Ibiza and Faliraki, a qualitative approach is more appropriate, given the descriptive nature of the research.

More-specifically, who (or what) is being studied – in terms of things like the size and composition of a group – may be a factor in method choice. Quantitative methods, such as social surveys or questionnaires lend themselves more-easily to the study of large, widely-dispersed, groups, while participant observation may be more appropriate for the study of small, geographically-localised, groups.

4. The level of at which a piece of research is funded may influence the methods used: email (or even postal) questionnaires are generally cheaper than in-depth interviews which, in turn, may be less-expensive than participant observation – although additional factors, such as the size and scope of the study, will come into play here. There are, in this respect, no hard-and-fast rules of funding and its relationship to choice of method: a large-scale questionnaire-based study covering the whole of the UK, for example, may be more-expensive to fund than a small-scale, single-researcher, participant observation study. The important point, however, is that funding will necessarily have some influence on a researcher’s choice of method.

For a more-specific example, funding levels may influence the size of any research team and this, as the Kendal Project research shows, can influence the choice of research methods. Without the ability to call on a large pool of student volunteers, for example, Heelas and Woodhead would not have been able to carry-out their observational count of Church attendees.

Boaz and Ashby (2003) also identify a further dimension to funding as a practical research consideration when they note that “Sensitivity to the sponsor’s (i.e. those funding the research) requirements can…introduce biases that conflict with the aim of producing objective, good quality evidence”. In other words, while Dunican (2005) argues “It seems logical that the selection of any research method should be based on the nature of the research question” this may not always be a choice in the gift of the researcher: a sponsor, such as a government department, may specify the type of research they want carried-out and the type of methods it would be appropriate to use.

5. The personal preferences and skills of the researcher can also be an influence on their choice of method. In terms of the former, for example, how the researcher sees the world – from a broadly positivist or broadly interpretivist position for example – will influence their choice of methods when studying that world. A researcher who values data reliability, for example, is unlikely to lean towards the use of observational methods.

In terms of the latter, different research methods require different skills: a researcher who, for whatever reason, lacks the listening skills or ability to establish the strong rapport required of an unstructured interviewer is unlikely to use such a method. Similarly, a male researcher will lack the necessary physical attributes to engage in covert participant observation in an all-female group.

References

Boaz, Annette and  Ashby, Deborah (2003) “Fit for purpose? Assessing research quality for evidence based policy and practice”: ESRC UK Centre for Evidence Based Policy and Practice: Working Paper 11

Cant, Savage and Chatterjee (2019) “Popular but Peripheral: The Ambivalent Status of Sociology Education in Schools in England”

Diken, Bülent and Laustsen, Carsten Bagge (2004) “Sea, Sun, Sex…and Biopolitics”: Department of Sociology, University of Lancaster.

Dunican, Enda (2005) “A Framework for Evaluating Qualitative Research Methods in Computer Programming Education”: Psychology of Programming Interest Group

Ferrell, Jeff (2018) “Drift: Illicit Mobility and Uncertain Knowledge”

Heelas, Paul and Woodhead, Linda (2004) “The Spiritual Revolution: Why Religion Is Giving Way to Spirituality

Wallis, Roy (1976) The Road to Total Freedom: A Sociological Analysis of Scientology

Venkatesh, Sudhir (2009) ” Gang Leader for a Day”: Penguin

Whyte, William (1943) “Street Corner Society”: University of Chicago