[None]

Chapter 2: Research, Theory and Method

1. See if Durkheim's theories of suicide still fit data about current suicide rates. For this, you will need a table of suicide rates by country (see http://tinyurl.com/ytvygu), and data on changes in economic performance and on religious affiliations by country (these are produced by national and international statistical offices and can also be found on the world wide web. For example, you can download a table of gross national product per person (GNP per capita) from http://tinyurl.com/26g32f and numerous statistics about countries for the world from the CIA World Factbook, http://tinyurl.com/jor92).

2. This chapter has suggested a particular model of social enquiry, one which proposes that social research involves theories, data, indicators and theory testing. In some ways this model can be regarded as itself a theory-a theory about social research. Like any theory, it ought to be capable of being compared with data. For this project, you should locate in the Library a recent issue of one of the major journals in your field. In sociology, this might be one of the Sociology, Sociological Research Online, Sociological Review, the British Journal of Sociology, the American Sociological Review or the American Journal of Sociology. Find an article in your chosen issue that looks interesting. Read the article closely to see the way in which the author puts forward his or her argument. Write down, in as few words as you can, the theory being advanced in the article. List the concepts that are used in the theory. For each concept, identify the indicators that the author uses. For each concept and indicator, briefly suggest what the implied measurement theory is. For some articles, these steps are easy to carry out. In other cases, you may find the theory, the concepts or the indicators hard to pin down. Is this because there is something amiss with the research being reported in the article, or because the model of social enquiry proposed in this chapter does not fit the research in the article you have been examining?

Chapter 3: Refining the Question

Devise a researchable question and then respond to the following. For each, provide a few sentences of explanation.

1. Does the question deal with a topic or issue that interests me?

2. Is the question one that can, in principle, be answered using data?

3. What types of information is needed to answer the research question?

4. How much data will be needed to answer the question?

5. Given the type and scope of the information, is the question too broad, too narrow, or about right?

6. What sources will have the type of information to answer the research question (e.g. journals, books, Internet resources, government documents, people)?

7. Can I access these sources?

[None]

Chapter 5: Searching for and Reviewing Literature

[None]

Chapter 6: Designing Samples

1. A Catholic magazine included a self-completion questionnaire on the attitudes of Catholics to recent changes in the Catholic Church and received over 10,000 replies from its readers. What could these data tell you about the attitudes of Catholics?

2. The government wishes to understand more about the reasons why juveniles commit offences. They decide to select a probability sample of 300 offenders from boys attending ten Youth Detention Centres. What could you learn about juvenile delinquency from this sample?

3. How would you obtain a sample of the following groups?

a. Illegal drug users in order to study the process of becoming a drug user.
b. Adults who are diabetic in order to study the effect of diabetes on family and social relationships.
c. Second generation Italians, i.e. people who were born in Britain and whose parents were born in Italy, for a study of the maintenance of Italian cultural practices. In each case, first define more precisely the population to be studied, then suggest alternative sampling strategies. How well would your proposed sample represent the population you initially defined?

4. Outline how you would select a probability sample of:

a. 5,000 women aged 16-59 which is nationally representative. (This is the sample selected in Martin and Roberts, 1984.)
b. 400 older people from a medium-sized town for a survey of unmet health needs.

Chapter 7: Grounded theory and inductive research

[None]

Chapter 8: Participatory Action Research

[None]

Chapter 9: Mixed Methods

1. Choose a topic, such as people's experience of watching television or listening to music, which is not too personal or intrusive. Ask 5 or 6 people to keep diaries for a specific length of time, e.g. a week, noting events, thoughts and experiences relevant to your topic. Analyse the diaries, then follow-up the diary-keeping with short, qualitative interviews with each individual. Combine or integrate the diary and interview findings in a short report.

2. Chose a subject for which quantitative data are accessible, for example a large-scale government survey such as the General Household Survey. Examine the data for a specific topic (for example recent use of health services). You may analyse the data yourself, or you may draw on results from the surveys that are published online. Using these results design an interview schedule that could be used to explore questions about your topic that the results bring up (for example the circumstances in which participants choose to consult a doctor or a complementary practitioner, or to rely on self-care). How has the quantitative analysis helped with the design of the interview schedule?

3. Conduct 3 or 4 qualitative interviews with friends about a chosen topic, perhaps asking about their experiences of holidays and their ideas about what makes a good holiday. Analyse these data. Use the emerging themes to design a structured questionnaire that could be used in a survey about your topic. How has the qualitative analysis helped with the design of the survey instrument?

4. Chose ten newspaper articles addressing the same subject. Analyse the material using different methods. (A) Content analysis: choose themes and develop codes to categorise these themes, based on the usage of specific words or the column inches devoted to specific themes. You may also wish to develop codes for any visual material that accompanies the article. After you have coded your sources, provide tables or graphs to display your findings. (B) Interpretive analysis: instead of choosing themes and coding them, you will develop an understanding of the ways the topic is presented in the articles, demonstrating these with illustrative quotes from your sources. Or you may wish to conduct a semiotic analysis of the text or visual images. How do the content and interpretive analyses compare?

Chapter 10: Ethics

[None]

Chapter 11: Questionnaires

Put these guidelines into practice by developing a self-completion questionnaire. Imagine that you have been asked by a local authority to carry out an investigation into attitudes towards the recycling of waste. You have been asked to select a representative sample of the population (see chapter 6) and to find out whether respondents recycle their waste, if so to what degree, if not, why not, etc. For example, what have respondents recycled during the past month? You will need to devise a working definition of what is meant by recycling activity to guide the design of the questions. The authority is particularly interested in how people might be encouraged to recycle more, so questioning along these lines will be necessary.

As this will be a self-completion questionnaire, try to include as many closed questions as possible, with pre-coded categories. So that correlations can be made when you analyse the data, choose a number of variables that may be relevant, for example: age, sex, employment status, income, marital status, housing status and so on.

Prepare a draft, test it with a friend, and then try the final questionnaire out on a small sample (e.g. five people).

Chapter 12: Measuring Attitudes

To do something worthwhile with your own data probably means a class project. Decide on a researchable area, probably one that other students will be interested in, since you will need willing research subjects. You could, for instance, examine attitudes to participating in higher education. This may well be multidimensional: for instance, some people are more concerned with the "pure" benefits of education: general intellectual development, self-actualisation, and so forth; whereas others will emphasise the practical advantages to the individual and to the nation. Do a bit of reading around, to get an idea of the necessary breadth of coverage. Start collecting items: keep a note of statements made in the press, especially student journals; record group discussions and extract coherent opinions. Edit your collection, bearing in mind the rules discussed in chapter 11. Assemble between ten and twenty items into a questionnaire, adding a few factual questions-gender, age, course of study, parents' occupations etc. Ideally, design and attempt to achieve a proper sample-though that is a project in itself. Set everyone in the class a quota: you will need at least two hundred completed questionnaires. Put the data onto a computer and (if you know enough about factor analysis, or can get more detailed advice) use my example as a model. For a relatively small data set you may be able to manage without a specialised package like SPSS. If you can get advice on a good spreadsheet program like Excel you won't be able to do factor analysis, but you can weed out items that don't seem to correlate with anything else, and then add together the remaining scores. Finally, look at the relationship between your scales and students' background characteristics. Which disciplines seem to be related to which justifications for education? Are there class background differences?

If you cannot collect your own data, you will find that many data sets available for secondary analysis (see chapter 21) are particularly suitable for attitude research: as well as the NORC GSS from which my illustration was taken (see Davis and Smith, 1992), there are the annual British Social Attitudes Survey and the Euro Barometer series sponsored by the European Commission. If you have access to suitable computing facilities you can learn a lot by obtaining one of these data sets from your national Data Archive and hunting through it for groups of connected attitude items. The British Social Attitudes Survey, in particular, is the subject of an annual report (for example, Jowell et al, 1999) that might give you some ideas on where to start .

Recently some data sets have become available for public analysis and downloading via the Internet. To get your own copy of the NORC General Social Survey used in the Factor Analysis example, point your browser at http://tinyurl.com/yrh2zs, and follow the instructions on screen. The subset I used took only a few seconds to download from the archive in California.

Chapter 13: Focus Groups

The government wishes to establish the views of people regarding the possible legalisation of drugs. You have been commissioned to study this issue by means of focus group discussions. The budget for the research would pay for up to 15 groups, but expenditure of that order would need to be justified. Briefly consider the following:

• How many groups should be convened?
• What should their composition be?
• How should they be organised?
• Where should they be held?

• Attitudes towards drugs and drug taking.
• Experience of using drugs.
• Attitudes towards the legalisation of drugs.
• Consequences of legalisation.

Chapter 14: Interviewing

Everyone has been to school. In this exercise you will pair with a partner for an interview. Each acts as interviewer and respondent in turn. The exercise takes about an hour. Begin by thinking over your school experiences and choose some aspect of school about which to ask questions. It could be relations with teachers, how people prepare for exams, the value of religious education, or many other things.

Since you will only be interviewing for 10 minutes, you only need to prepare for the beginning stage of the interview. Write down several questions addressing the research issue, and a standard 'project explanation', a general statement of the research issue which you can say to your respondent to get the interview going.

Take ten minutes to design your questions.

Now choose who will be interviewer first. Do your interview in 10 minutes. There is no need to take notes on the respondent's replies.

The next stage is debriefing. The interviewer should write down skills they managed well and those needing work. Discuss it with your respondent. This should take 5 minutes.

Now swap roles, do another 10 minute interview and another 5 minute debriefing.

When you have finished, discuss with your partner your experience of interviewing and the accuracy of your debriefing notes. Allow 15 minutes.

Chapter 15: Ethnography

This project is an exercise in collecting observational data. The emphasis is on developing your skills of observation and gaining experience of the techniques of recording social events. First, think about a research design before your fieldwork. Decide what research topic is appropriate to study through observation. For instance, you might be interested in the unstated rules that govern queuing behaviour, and so watch people lining up for buses. Or you might be interested in how people interact when they are engrossed in video games, and find your way to the Student Union to watch them. Another idea is to watch how people order drinks in the pub, noting gender differences. Several students might work on the same topic so that they can (literally) compare notes.
Second, carry out field observation. Record your observations by writing fieldnotes. Now write a description of the research procedure and the sort of data you gathered. Mention any problems in using the method and evaluate how it went.
The exercise works best if there is someone with whom you can discuss it and who has also tried it out. When you are thinking over the experience, or discussing it, here are the sorts of questions you need to ask to assess the adequacy of your observations:

1. How accurate an observer am I of sequences of action? of dialogue?
2. Have I the ability to write ‘concretely’ or do my notes contain generalisations and summaries?
3. Was my research aim realistic? Was it adequately specified at the outset? Was it interesting? Was it sociological?
4. Was I comfortable doing the observation? Did I tell anyone I was researching? If I did not tell anyone I was researching, do I think anyone guessed?
5. What other methods could I have used to get this data? Now that I have tried observation, was it the best available method to get this data?

A photo that appeared in more than one British newspaper showed a group of people in an anteroom in Number 10 Downing Street, waiting for the Prime Minister to summon them for the crucial meeting at which the decision was taken to invade Iraq in 2002.  Most of the people in this photograph are members of the British Cabinet.  Two of them - a man  and a woman - are not; they are high ranking civil servants.  The man is in close conversation with the government ministers: the woman is outside the circle, looking at their backs.
This photo was accompanied by no reference to the gender of those depicted, nor to the placing of persons in relation to one another, but to a student of social interaction it would appear that a group of powerful men – cabinet ministers – conversing informally are excluding one of the two people of a lower status and including the other.  The first is a woman, the second is a man.

The project is either to

(a) Explore ways in which possible evidence of patriarchy is displayed pictorially in documents (in relation to a particular area of life such as politics or in a particular medium such as advertisements) and how such visual evidence is referred to in accompanying text.

(b) Investigate this kind of data in a particular organization.  Professional bodies might well be fruitful cases, because professional journals, newsletters and even the media print the pass lists, including prize winners, of professional qualifying exams.  Their names usually enable one to determine gender and by following up members over time in membership records etc one could determine the relative importance of ability and gender.  Reports and photos of conferences and similar occasions can provide further evidence.

Chapter 17: Narrative Analysis

You have been asked to carry out narrative interviews on the subject of childhood with 3 people of different ages. Ideally, at least one of the interviewees should be aged 70+ and no interviewee should be younger than 25. The interviews should last approximately one hour. You need to develop an interview guide consisting of 6 or 7 main questions together with subsidiary prompts. One of the interviewees should be interviewed on 2 occasions so that you can explore the advantages (and any disadvantages) of multiple interviews with the same person.

Chapter 18: The internet and research methods

Imagine that you have been commissioned to carry out a study exploring attitudes to alternative medicine amongst university students. You can decide whether the study will be a qualitative piece of research using interviews, or a quantitative survey using a questionnaire. Develop a detailed design for the research project, addressing the following issues:

• How will you identify a suitable sample of students?
• What kinds of bias might this sample involve, and how might this affect the generalisability of results?
• How will you identify students and recruit them for interviews or questionnaires? Is there anyone whose permission you should ask (note: it may be important to clear use of institutional mailing lists with someone in authority first)?
• Would you use email interviews or would you conduct interviews in real-time using instant messaging? Would you send surveys in the body of an email or as an attachment, or would you place the survey on a web site?
• How will you frame your initial contact with potential research subjects? What would your approach be to informed consent? How, practically speaking, would you expect people to signal their consent?
• What response rates might you expect? What means could you use to encourage response?

You have been asked to investigate people’s beliefs in the paranormal through administering a questionnaire in the street to the general public. From such a study, you can begin to investigate people’s understanding of scientific knowledge.

Decide whether you are going to administer a survey questionnaire or carry out a structured interview. Design your interview schedule or your interviewer-administered questionnaire to include the following questions. Try to pre-code the questions where possible and mark your questionnaire with response codes and column numbers in order to simplify data entry:

1. What kinds of things come to mind when you think of the paranormal?
This is an open question to which people may offer more than one answer.
Allow several lines on the questionnaire in which to write the answers.

2. Do you think that any of the following are true:

a. It is possible to make someone turn round just by looking at them.
b. Prayers can sometimes be answered.
c. It is possible to know what someone else is thinking or feeling even if they are hundreds of miles away and out of touch by ordinary means.
d. Some houses are haunted.
e. The Earth has been visited by beings from outer space.
f. Dreams can sometimes foretell the future.
g. Some people can remember past lives (i.e. reincarnations they may have lived).
h. It is possible to get messages from the dead.
You should expect people to respond with either a positive or a negative answer, although it is likely that there will also need to be a ‘Don’t know’ category.

3. Are you a religious person?
Decide whether to use a yes/no answer or an ordinal scale with responses ranging from very religious to not at all religious.

4. Age. Decide whether to use age ranges or to ask people their actual age.

5. Sex

6. Social class

7. Ethnic group. Remember to include columns for an identification number. After designing your questionnaire, select a small sample to interview and administer it to them. Create a codebook and then code the open question. Finally, you could create the data file.

If you have access to a computer with almost any version of SPSS you will almost certainly be able to use the small subset of the General Social Survey which is included with the package.  Ask your local SPSS coordinator how to do this, and how to get a frequency count for all the variables.  Having got this far, browse through the variables until you find one that looks interesting as a dependent variable:  you could start with satjob, hapmar or life (roughly satisfaction with job, marriage and life generally).  Try to find out what sort of people are happiest with their lot.  What marital status is most conducive to general life satisfaction?  What truth is there in the assertion that men get more out of marriage than women?  All these questions can be addressed by cross-tabulating a satisfaction variable by one or more background factors, possibly after recoding.

1. Compare the advantages and disadvantages of conducting secondary analysis of a large government survey with carrying out some focus groups for a research study of the reasons for gender differences in earnings.

2. Think of a research problem that can be examined using secondary analysis. For example, what is the extent of racial prejudice in Britain and what kinds of people are more or less likely to hold such attitudes?

Then undertake the following:

a) Find out what surveys could be used to address your research problem by using the ESDS Data Catalogue.
b) Having identified a relevant survey, select questions to measure the concepts needed to address your research problem. Provide a critical appraisal of the adequacy of these questions as measures of the concepts to be studied.
c) Draw a conceptual model of the inter-relationships between each of your concepts, including the direction of any expected causal relations.

3. Conduct secondary analysis using a subset of the 2002 General Household Survey, which can be downloaded from http://www.soc.surrey.ac.uk/uss/. For example, elaborate a two-variable relationship by controlling for one or more theoretically relevant control variables and draw conclusions about the direction of relationship between these variables. Your analysis could be conducted using cross-tabulation or graphical techniques. Your analysis will be easier if you first recode all the variables to be analysed into either two or three categories.

Chapter 22: Computer Assisted Qualitative Data Analysis

The sooner you start using CAQDAS program, the sooner you will start to feel a more secure about what the program expects of you and its style and you will gain the confidence to try out new things.

• Download a free trial version of the software (see resources section)
• Create a new project in the software
• Find out what formats the software accepts data in – see online help menus and information in Suggestions and preparation checklist above.
• Experiment by importing or assigning documents (data files) or literature abstracts 0 or anything textual
• Experiment in the creation of memos to map out elements of future work
• Explore main menus –and try out right button context menus with various things selected.
• Try out various tasks such as code creation, apply codes to selections within the textual documents you have assigned or imported
• Nothing you do at this stage is dangerous – make mistakes and learn from them
• Delete codes or rename them
• Find out what you can’t do!
• Find out how to Save your work, close the project and try to re-open it.

1. Analysing devices or structures in talk.

Collect a video or tape recording of a discussion between representatives of the major political parties, or an in-depth interview with one politician.  Examine the ways in which specific positions and policies are described.  What are the devices used to make specific policies seem reasonable and sensible?  How do these devices work? (Atkinson’s 1984 study of political rhetoric will be helpful for this project.)

2. Analysing the factual orientation of discourse

Tape record interviews with friends who believe they have had a supernatural or paranormal experience. Ask them to describe what happened. Then transcribe their account, and try to identify how the account is organised to establish that the event actually happened, and was not, for example, an hallucination, dream or misidentification. (Wooffitt, 1992, describes some of the descriptive practices through which speakers build robust accounts of extraordinary experiences.)

3. Analysing the discourses in household objects

Collect everyday bathroom products that come in packaging: shower products, shampoos, soap dispensers, bathroom cleaning materials, and so on. Concentrate primarily on the written words. Following Parker’s method outlined above, look first for proper nouns, and then try to identify the categories of person referred to or invoked on the object. What kind of consumer is being constructed? What kind of person is this? What kind of ‘proper’ or normatively acceptable behaviour is being reinforced? On the basis of this analysis then try to describe the broader discourses that informs these texts. What do these discourses do? Who benefits?

Chapter 24: Analysing Visual Materials

(1) Media analysis: Choose a research question, for instance, how a social group (either a broad one, like mothers or young people, or a narrower one, like motorcyclists or police officers) is depicted.  Choose a type of existing visual data - movies, newspaper photographs, advertisements, illustrations in children's books, television programmes, historical paintings or something else - which might portray your group.  Analyse your data using one of the methodologies mentioned.  In your write-up, defend your choice of method and sampling.  Think about how to present the visual component in your essay.

(2) Photo elicitation:  Choose one of these projects - or create your own.  (A) Ask your grandparents, or a few elderly neighbours, to show you photographs from their youth.  Interview them about consumption patterns (what did they buy, make, or do without?) or social rituals (how did they celebrate birthdays, what did they do and where did they go with their friends?).  (B) Pick a controversial topic, for instance, 'Fashion models encourage girls to be anorexic'.  Gather a number of images (e.g. ten advertisements or pictures of famous people from newspapers, showing different body composition), or find a movie or TV show, on video, that captures the issue.  Organise a focus group of your friends and lead a discussion using your visual material as prompts.  Consider how best to convey the content of your images to your readers.

(3) Envision data: Use your camera or camcorder to gather data. (A) Show how gravestones and cemeteries preserve memory, prestige, and power.  (B) Study what the buildings that house different kinds of organisations say about their occupants.  (C) Examine behaviour in shopping districts, asking how shoppers browse, circulate and queue, or what people communicate by their dress and speed of walking.  In all of these, be scrupulous about research ethics.  It is acceptable to photograph cemeteries and the outsides of buildings, but you will need permission to shoot inside offices and stores.  You may find this permission difficult to obtain, but do not be tempted to cheat.  You will be more unobtrusive, and safe, if you choose a busy high street in tourist areas.  'Quote' your data, or describe them, in your report.

Chapter 25: Writing about Social Research

Every academic paper is reviewed before publication by at least two referees, chosen from amongst the academic community by journal editors for their knowledge of the paper’s topic.  Referees (who are rarely paid for their labours) receive a copy of the paper in typescript and a letter (or sometimes a form) from the editor that requests them to comment on the paper’s suitability for publication in terms of its clarity, originality and the adequacy of its argument.  The referee has between two and six weeks to respond with a verdict (one of: accept, accept with minor revisions such as spelling or stylistic errors, accept but require major revisions, or reject) and a report.  The report, which is always anonymous and will often be forwarded to the author, explains the referee’s verdict, commenting on the overall strengths and weaknesses of the paper and making suggestions for improvements. 

The first step in this project is to identify a research area in which you have an interest (the sociology of health, or deviance, for example).  Look through the Library Current Journals shelves and find a recent issue of a journal publishing in your area.  Choose one paper from that journal.  If possible, make a photocopy of it or download a copy from the internet.

You should imagine that you have received a typescript copy of this paper from a journal editor to referee.  Examine the paper closely.  Is it clear?  Is it well organised?  Can one quickly identify the main conclusions?  Are the justifications for those conclusions soundly based?  Are the data appropriate to the topic of the paper?  Are the methods of data collection and analysis described in sufficient detail that someone else could repeat the study?  Are there plausible alternative interpretations of the data or of the results that the author has not noticed?  Does the abstract adequately describe the contents of the paper?  Are there passages that could be rewritten to make them clearer?  Has the author referenced all the relevant literature? 

It is surprising how often even published papers will wilt under careful scrutiny of this kind.   In about 300 words, write a report about the article.  Your criticisms should be phrased so that they are constructive, polite and encouraging—the aim of a referee’s report is to encourage the author towards a better article, not to damage the author’s ego.