sc2145 ch1-3 PDF
sc2145 ch1-3 PDF
sc2145 ch1-3 PDF
A.P. Kelly
SC2145, 2790145
2011
This guide was prepared for the University of London International Programmes by: Aidan P. Kelly, BA, MA, Lecturer in Social Research Methods, Department of Sociology, Goldsmiths, University of London. This is one of a series of subject guides published by the University. We regret that due to pressure of work the author is unable to enter into any correspondence relating to, or arising from, the guide. If you have any comments on this subject guide, favourable or unfavourable, please use the form at the back of this guide.
University of London International Programmes Publications Office Stewart House 32 Russell Square London WC1B 5DN United Kingdom Website: www.londoninternational.ac.uk Published by: University of London University of London 2007 Reprinted with minor revisions 2011 The University of London asserts copyright over all material in this subject guide except where otherwise indicated. All rights reserved. No part of this work may be reproduced in any form, or by any means, without permission in writing from the publisher. We make every effort to contact copyright holders. If you think we have inadvertently used your copyright material, please let us know.
Contents
Contents
Introduction ............................................................................................................ 1 What this course is about .............................................................................................. 1 Aims ............................................................................................................................. 1 Learning outcomes ........................................................................................................ 2 Syllabus......................................................................................................................... 2 The importance of social research methods .................................................................... 2 Structure of the course................................................................................................... 5 Prerequisite ................................................................................................................... 6 Essential reading ........................................................................................................... 6 Further reading.............................................................................................................. 6 Online study resources ................................................................................................... 8 Using the internet.......................................................................................................... 9 How to use the textbook and readings ........................................................................ 10 Hours of study and use of this subject guide................................................................. 12 The examination and examination advice ..................................................................... 13 Part A: Social research strategies ......................................................................... 15 Chapter 1: What is social research?...................................................................... 17 Aims and objectives ..................................................................................................... 17 Learning outcomes ...................................................................................................... 17 Essential reading ......................................................................................................... 17 Further reading............................................................................................................ 17 Websites ..................................................................................................................... 17 Works cited ................................................................................................................. 18 Introduction ................................................................................................................ 18 Research strategies ...................................................................................................... 19 Research design .......................................................................................................... 24 Research methods for data collection and analysis ....................................................... 27 Principles and practice in social research ...................................................................... 30 Social research and academic sociology ....................................................................... 30 The reaction against positivism and the paradigm wars ................................................ 31 Methodological awareness in sociology and social research .......................................... 32 The biographical context .............................................................................................. 34 Conclusions ................................................................................................................. 35 A reminder of your learning outcomes.......................................................................... 36 Sample examination questions ..................................................................................... 36 Chapter 2: Concepts in social research ................................................................ 37 Aims and objectives ..................................................................................................... 37 Learning outcomes ...................................................................................................... 37 Essential reading ......................................................................................................... 37 Further reading............................................................................................................ 37 Websites ..................................................................................................................... 38 Works cited ................................................................................................................. 38 Introduction ................................................................................................................ 38 What is a concept? ...................................................................................................... 38 Theory, concepts and concept formation....................................................................... 40
i
Concept formation in sociology: the case of social exclusion ......................................... 41 Operational definitions in quantitative social research .................................................. 41 Coding and recoding in quantitative social research ..................................................... 42 Linking concepts and the outcomes of social research .................................................. 43 Sensitising concepts................................................................................................... 43 Social exclusion in qualitative social research ............................................................... 44 Retroduction and double-fitting ................................................................................. 44 Conceptual modelling .................................................................................................. 44 Conclusions ................................................................................................................. 45 A reminder of your learning outcomes.......................................................................... 45 Sample examination questions ..................................................................................... 46 Part B: Qualitative social research........................................................................ 47 Chapter 3: Designing qualitative social research ................................................. 49 Aims and objectives ..................................................................................................... 49 Learning outcomes ...................................................................................................... 49 Essential reading ......................................................................................................... 49 Further reading............................................................................................................ 49 Works cited ................................................................................................................. 49 Introduction ................................................................................................................ 50 Sources of qualitative data........................................................................................... 53 Non-probability sampling and theoretical sampling ...................................................... 56 Prestructuring of qualitative research design ................................................................ 57 Conclusions ................................................................................................................. 57 A reminder of your learning outcomes.......................................................................... 58 Sample examination questions ..................................................................................... 58 Chapter 4: Qualitative data analysis .................................................................... 59 Aims and objectives ..................................................................................................... 59 Learning outcomes ...................................................................................................... 59 Essential reading ......................................................................................................... 59 Further reading............................................................................................................ 59 Websites ..................................................................................................................... 59 Works cited ................................................................................................................. 59 Introduction ................................................................................................................ 60 Research questions and qualitative analysis ................................................................. 60 Validity in qualitative data analysis............................................................................... 64 Conclusions ................................................................................................................. 65 A reminder of your learning outcomes.......................................................................... 66 Sample examination questions ..................................................................................... 66 Part C: Quantitative social research ..................................................................... 67 Chapter 5: Designing quantitative social research .............................................. 69 Aims and objectives ..................................................................................................... 69 Learning outcomes ...................................................................................................... 69 Essential reading ......................................................................................................... 69 Further reading............................................................................................................ 69 Works cited ................................................................................................................ 69 Introduction ................................................................................................................ 70 Research questions and quantitative designs................................................................ 72 Descriptive quantitative social research ........................................................................ 72 Hypothesis testing and evaluation research .................................................................. 72 Interpretative model building ....................................................................................... 73 Synthetic model building.............................................................................................. 74
ii
Contents
Causal models in quantitative social research ............................................................... 75 Validity in social survey research .................................................................................. 76 Causal inference in the classical scientific experiment ................................................... 78 The panel study ........................................................................................................... 79 The social survey the random sample one-group, post-test only design .................... 80 Causal modelling and survey design............................................................................. 81 Designing social surveys: interview schedules and questionnaires ................................. 82 Multi-dimensional attitude scales ................................................................................. 83 Sample design, sampling theory and statistical inference .............................................. 84 Secondary analysis of large-scale social surveys ............................................................ 85 Conclusions ................................................................................................................. 86 A reminder of your learning outcomes ......................................................................... 86 Sample examination questions ..................................................................................... 87 Chapter 6: Analysing quantitative data ............................................................... 89 Aims and objectives ..................................................................................................... 89 Learning outcomes ...................................................................................................... 89 Essential reading ......................................................................................................... 89 Further reading............................................................................................................ 89 Websites ..................................................................................................................... 90 Introduction ................................................................................................................ 90 Univariate analysis....................................................................................................... 91 Sampling distributions, standard error and confidence intervals .................................... 91 Exploring causal models .............................................................................................. 93 Bivariate data analysis ................................................................................................. 93 Measures of association and bivariate tables ................................................................ 95 Measures of association based on proportionate reduction in error (PRE) ..................... 95 Measures of association based on Chi-square: Cramers V ........................................... 96 Which measure of association do we use in our crosstabulations? ................................ 97 Multivariate analysis .................................................................................................. 101 The elaboration paradigm .......................................................................................... 105 Causal modelling and path analysis ........................................................................... 110 The assessment of indirect effects .............................................................................. 113 Conclusions ............................................................................................................... 115 A reminder of your learning outcomes........................................................................ 115 Sample examination questions ................................................................................... 116 Part D: The context of social research ................................................................ 119 Chapter 7: The ethics and politics of social research ......................................... 121 Aims and objectives ................................................................................................... 121 Learning outcomes .................................................................................................... 121 Essential reading ....................................................................................................... 121 Further reading.......................................................................................................... 121 Websites ................................................................................................................... 121 Works cited ............................................................................................................... 122 Introduction .............................................................................................................. 122 The ethics of social research....................................................................................... 122 Feminism and the politics of social research ............................................................... 125 The politics of identity ................................................................................................ 126 Conclusions ............................................................................................................... 127 A reminder of your learning outcomes........................................................................ 128 Sample examination questions ................................................................................... 128 Appendix: Sample examination paper............................................................... 129
iii
Notes
iv
Introduction
Introduction
What this course is about
Welcome to this course 145 Social research methods. This subject guide aims to help you learn about social research. In this introduction, I want to point out some features of learning about social research methods and discuss how to use this guide and its associated readings. Social research methods is a technical subject that has its own language and this language must be used in a very precise way. When studying social research methods what may seem to be everyday terms, such as validity have a much more specific meaning. In this course, the precise meaning of technical terms has to be learned and then used carefully, especially in the examination. That is why one of your first tasks will be to start a glossary of your own. As students of social sciences, you will be familiar with the idea of sociology as a multi-paradigm discipline: sociology does not have a single paradigm of interlinked theory and method. Debates and controversies abound! By far the deepest divide, up until recently, has concerned the philosophical basis of sociological research and the choice of appropriate social research methods. This divide has a long history and it widened considerably during the 1970s. Many textbooks introduce sociology students to the debate between the positivists who argued that social science was similar in method to the natural sciences, and the antipositivists who argued that the nature of the subject matter made the idea of a social science a contradiction in terms. The division between positivism and anti-positivism has become synonymous with that between qualitative and quantitative social research strategies. The so-called paradigm-wars between these opposing strategies are thought to be outdated, but the conflicting perspectives live on in the structure of textbooks on social research methods and in this subject guide. It seems impossible to teach social research methods in any other way than to consider quantitative and qualitative approaches separately. As far as the practice of social research is concerned, it is also important to keep in mind three developments: the increased willingness to use both quantitative and qualitative strategies the argument that choice of strategy is based on the research question and what sort of answers are to be pursued the shift of emphasis in the paradigm wars, away from quantity versus quality, to the debate questioning the existence of a social reality that is independent of our means of studying it. This subject guide retains the old distinction between qualitative and quantitative research strategies as it is reflected in the essential textbook.
Aims
This course is designed to: describe the key components of social research develop skills and knowledge about quantitative and qualitative social research identify criteria used to evaluate the quality of social research.
1
Learning outcomes
By the end of the course, and having completed the Essential reading and activities, you should be able to: formulate researchable questions define a research strategy and design a research project to answer a research question discuss the practice and principles of qualitative and quantitative social research use skills and knowledge acquired in the course to evaluate the quality of published research by sociologists and other social scientists.
Syllabus
The contexts of Social Research: Social research as a professional activity. The market for social research outputs. The contexts of data collection. Social research and academic sociology. Philosophy and the practice of social research. The cultural context and ethnocentrism. Developing research proposals in context. Models and Modelling in Social Research: Ontological and epistemological status of models. Models and researchable questions. Exploratory and Confirmatory approaches to model building. Flexibility in Research Design. The use of models in quantitative and qualitative social research. Concepts in Social Research: Approaches to concept formation in Sociology. Theory and concepts. Induction and deduction. The use of concepts in qualitative and quantitative social research. Coding qualitative data. Operationalising concepts. Complex concepts: property spaces. Qualitative Social Research: Sources of qualitative data: the interview, focus groups, participant observation and field work methods, documents. Sampling in qualitative social research: grounded theory and theoretical sampling. Analytic induction and qualitative classification analysis. Small N research and case studies: thick description. Quantitative Social Research: Sources of quantitative data: the social survey, administrative and official statistics. Secondary analysis of survey data. Designing social surveys. Quasi-experimental designs. Units and levels of analysis. Graphics and visualising data. Explanation and causal inference. Deciphering and Evaluating Social Research Outputs: The components of a social research publication. Identifying components. Reconstructing the research project. Validity in quantitative and qualitative research.
Introduction
Political parties use opinion surveys and focus groups to monitor the likely reaction to new policy developments or to indicate the points of advantage the party has over its political rivals. Governments and large businesses are increasingly evidence-based in their approach to decision making. It is a feature of contemporary societies that people make, or claim to make, decisions based on data: data about impact of medicines, data about impact of marketing campaigns, data about sports performance. If it moves, that movement is being recorded and stored on a database somewhere. Nowhere is private and protected from data collectors: mobile phones record calls made and their duration and spy software invades our personal computers to record keystrokes and websites visited often without our knowledge or permission. Search engines such as Google also collect the same kind of data about the websites we visit. Apart from our worries about the invasion of privacy represented by these developments, the very existence of these data mines should sensitise us to the need to understand how data is constructed, collected and analysed. Knowledge of research methods will lead us to analyse the data collection techniques used and the potential impact of the methods on the analysis of the data collected. Conclusions that others may draw from the analysis of such data can be scrutinised using principles and techniques of social research methods. In everyday life everyone will be affected by the results of analysed data, it is therefore important that we learn how to scrutinise the strategies, methods and techniques of those that produced these findings.
the ability to deploy an informed scepticism when making judgments about your own and others social research the ability to place research projects, including your own, within a specific social context. This includes an ethical context, a political context and an academic context as well as the immediate context of data collection that may influence or may have influenced the construction of the data collected. Knowledge of social research methods will provide you with a checklist of issues surrounding the conduct of social research. Thus when confronted with a journal article reporting on a research project, you will have a new way of reading that article. Your reading will be informed by your ability to identify the key components of a project and a knowledge of the potential weaknesses of the research. It will then deploy a properly informed skepticism an unwillingness to accept the validity of arguments and evidence until a set of criteria has been applied. One of the significant advantages of Brymans book Social Research Methods is his continual reference to actual empirical studies. You may also wish to consult the book by Devine and Heath entitled Sociological Research Methods in Context (Basingstoke: Palgrave, 1999), which summarises and comments on eight recent sociological studies. Without referring to these and other empirical studies, learning about social research methods can become a very artificial, even abstract experience. Since there are limits to the amount of research you can conduct on your own while doing this course, I urge you to develop knowledge of actual examples of the way in which research methods are used and with what results. In this way, you can judge research by the outcomes of social research processes: what findings do the methods produce and how useful are they in advancing your understanding of the social world? This seems far more sensible than evaluating a research method by reference to some pre-existing philosophical scheme such as positivism and interpretivism. Although it is important to understand the philosophical underpinnings of research strategies, these, in my view, have become far too dominant in the practice of teaching social research methods. What I want you to be able to say after taking this course is that you have: studied a range of social research methods seen what kind of research results from a choice or choices of method acquired an awareness of the contribution research can make both to our understanding of the social world and to the way it is changing. Thus, whatever political views you may have about current social arrangements you should be able to conduct empirical social research that can inform debates about these and you can evaluate the research of other sociologists and assess its relevance for current social debates. Not all students have the ambition to become professional social researchers. However, all social science students should recognise the valuable intellectual and practical skills that are acquired through knowledge of social research methods. My own approach to social research may be described as a pragmatic realism. I assume that there is a social world out there, independent of our private conceptions, which we can explore and attempt to model. This assumption allows me to conduct research that may or may not lead me to find out things we do not already know. If the results enhance understanding of the social world then this justifies spending time and resources on the conduct of social research. What other justification can there be for the conduct of social research?
4
Introduction
Parts A and D: The nature of social research and The context of social research
In Part A of the guide, you will be learning about the nature of social research and its key components. This involves exploring the process of social research and the specific issues of deduction and induction in social research especially in relation to concepts and concept formation. As well as knowing how to do social research, it is important to understand the social and intellectual factors that influence the nature of social research. In Part A, these include: philosophical positions and methodological awareness practical matters such as the people and places involved. And in Part D, the following influences on research practice are reviewed: ethical constraints and values the politics of social research, including the influence of feminist politics on the conduct of social research.
Prerequisite
If you are studying this course as part of a BSc degree, you must already have studied 21 Principles of Sociology. You may wish to refresh your understanding of social research you gained from 21 Principles of Sociology, especially Chapters 2 and 3.
Essential reading
The Essential reading for this course is from one textbook:
Bryman, A. Social Research Methods. (Oxford: Oxford University Press, 2004) second edition [ISBN 9780199264469].
Detailed reading references in this subject guide refer to the edition of the set textbook listed above. New editions of this textbook may have been published by the time you study this course. You can use a more recent edition of any of the books; use the detailed chapter and section headings and the index to identify relevant readings. Also check the virtual learning environment (VLE) regularly for updated guidance on readings.
Further reading
Please note that as long as you read the Essential reading you are then free to read around the subject area in any text, paper or online resource. You will need to support your learning by reading as widely as possible and by thinking about how these principles apply in the real world. To help you read extensively, you have free access to the VLE and University of London Online Library (see below). * = edited collection which contains many supplementary readings. ** = particularly recommended reading.
** Ali, S. et al. Politics, Identities and Research in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. Aneshensel, C.S. Theory-Based Data Analysis for the Social Sciences. (London: Pine Forges Press, 2002) [ISBN 978-0761987369]. ** Arbor, S. Secondary Analysis of Survey Data in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Blalock, H. Are there really any constructive alternatives to causal modeling, Sociological Methodology 21 (1991), pp.32535. Blalock, H. The Operationalism Controversy in Blalock, H. and A.B. Blalock (eds) Methodology in Social Research. (London, New York: McGraw-Hill, 1971) [ISBN 0070941815]. Blumer, H. The Ethics of Social Research in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Bryman, A. Quantity and Quality in Social Research. (London: Routledge, 1988) [ISBN 9780415078986]. Bryman, A. Paradigm Peace and the Implications for Quality, International Journal of Social Research Methodology, 9(2) 2006, pp.11126. Byrne, B. Qualitative Interviewing in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. Dale A., S. Arber and M. Procter Doing Secondary Analysis. (London: Allen and Unwin, 1988) [ISBN 9780043120422] Chapters 811. * Devine, F. and S. Heath Sociological Research Methods in Context. (Basingstoke: Palgrave, 1999) [ISBN 9780333666326] Chapters 1, 7 and 10. 6
Introduction Devine, F. and S. Heath Sociological Research Methods in Context. (Basingstoke: Palgrave, 1999) [ISBN 9780333666326] Chapter 9 Disarming Patriarchy. De Vaus, D. Analyzing Social Science Data. (London: Sage Publications 2002) [ISBN 978-0761959380]. Especially Chapter 32, How and When to Use Cross Tabulations and Chapter 35, How to Interpret a Correlation Coefficient. De Vaus, D. Surveys in Social Research. (London: Routledge, 2002) fifth edition [ISBN 9780415268585]. De Vaus, D. Research Design in Social Research. (London: Sage Publications, 2001) [ISBN 9780761953470] Chapters 1, 2, 3 and 12. ** Fielding, J. Coding and Managing Data in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Fielding, N. Ethnography in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Gilbert, N. Research Theory and Method in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. * Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Hammersley, M. The Politics of Social Research. (London: Sage Publications, 1995) [ISBN 9780803977198]. Healey, J. Statistics: a Tool for Social Research. (Belmont: Thomson Wadsworth, 2005) seventh edition [ISBN 9780534627942]. Kirk, J. and J. Miller Reliability and Validity in Qualitative Social Research. (London: Sage Publications, 1986) [ISBN 9780803924703]. Layder, D. Sociological Practice: Linking Theory and Social Research. (London: Sage Publications Ltd., 1998) [ISBN 9780761954293] Chapters 3, 4 and 5. Lewins, A. Computer Assisted Qualitative Data Analysis in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Marsh, C. The Survey Method: the Contribution of Surveys to Sociological Explanation. (London: Unwin Hyman, 1982) [ISBN 9780043100158]. Procter, M. Analysing Survey Data in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Ritchie, J. and J. Lewis Qualitative Research Practice. (London: Sage Publications, 2003) [ISBN 9780761971108]. Robson, C. Real World Research. (Oxford: Blackwell Publishers Ltd, 2002) second edition [ISBN 9780631213055] Chapter 2 Approaches to Social Research. * Seale, C (ed.) Researching Society and Culture. (London: Sage Publications, 2004) second edition [ISBN 9780761941965]. ** Seale, C. Coding and Analysing Data in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. ** Seale, C. Statistical Reasoning: Causal Arguments and Multivariate Analysis in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. Seale, C. et al. The Quality of Qualitative Research. (London: Sage Publications, 1999) [ISBN 9780761955986] Chapters: Trust Truth and Philosophy, Guiding Ideals and Grounding Theory. Tonkiss, F. Analysing Text and Speech in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. Tonkiss, F. Focus Groups in Seale, C (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. Wooffitt, R. Conversation Analysis and Discourse Analysis. (London: Sage Publications, 2005) [ISBN 9780761974260].
The VLE
The VLE, which complements this subject guide, has been designed to enhance your learning experience, providing additional support and a sense of community. It forms an important part of your study experience with the University of London and you should access it regularly. The VLE provides a range of resources for EMFSS courses: Self-testing activities: Doing these allows you to test your own understanding of subject material. Electronic study materials: The printed materials that you receive from the University of London are available to download, including updated reading lists and references. Past examination papers and Examiners commentaries: These provide advice on how each examination question might best be answered. A student discussion forum: This is an open space for you to discuss interests and experiences, seek support from your peers, work collaboratively to solve problems and discuss subject material. Videos: There are recorded academic introductions to the subject, interviews and debates and, for some courses, audio-visual tutorials and conclusions. Recorded lectures: For some courses, where appropriate, the sessions from previous years Study Weekends have been recorded and made available. Study skills: Expert advice on preparing for examinations and developing your digital literacy skills. Feedback forms. Some of these resources are available for certain courses only, but we are expanding our provision all the time and you should check the VLE regularly for updates.
Introduction
To access the majority of resources via the Online Library you will either need to use your University of London Student Portal login details, or you will be required to register and use an Athens login: http://tinyurl.com/ ollathens The easiest way to locate relevant content and journal articles in the Online Library is to use the Summon search engine. If you are having trouble finding an article listed in a reading list, try removing any punctuation from the title, such as single quotation marks, question marks and colons. For further advice, please see the online help pages: www.external.shl.lon. ac.uk/summon/about.php
documents use Google scholar as available on the free browser from Firefox (www.mozilla.com/en-US/). Many online resources explain concepts and techniques used in social research. You can also consult the website, (www.oup.com), which provides a list of selected web links made available to those who have purchased the Bryman text.
10
Introduction
The kinds of research questions addressed by qualitative social research allows this to happen more spontaneously. In addition, taking a slightly different approach provides more of a test of your understanding of what you have read in the textbook. This subject guide requires you to use the Bryman textbook and readings in a creative way. There is no substitute for reading Bryman and supplementing this with some of the further reading I suggest. You must not rely on this subject guide alone. This guide alone does not provide a basis for the successful completion of the course. I would begin this course by reading and making notes on whole chapters of Bryman in the following sequence: Part A of the subject guide Chapter 1 Social research strategies Chapter 2 Research designs Chapter 3 The nature of quantitative social research Chapter 13 The nature of qualitative social research Part B of the subject guide Chapter 14 Ethnography and participant observation Chapter 15 Interviewing or Chapter 16 Focus groups Chapter 19 Qualitative data analysis Chapters 21 and 22 on combining qualitative and quantitative social research Part C of the subject guide Chapter 7 Asking questions Chapter 5 Structured interviewing Chapter 4 Sampling Chapter 10 Secondary Analysis Chapter 11 Quantitative Data Analysis Part D of the subject guide Chapter 25 The ethics and politics of social research
11
process can be mechanistic or highly structured processes, others are not. In the case of qualitative social research, rules of conduct have not been fully developed and the consequence is that rules are constructed by the researcher during the process of data collection and analysis. In qualitative research, the absence of rules generates the need for creativity and flexibility that are seen as an advantage rather than a drawback. There must be a sequence to learning about the various topics covered in a social research methods course. However, it is often apparent to students and tutors that the first topics covered would be easier to understand if the students had already completed the topics that follow! It is quite normal to teach students how to design research before covering matters of data analysis, even though it is considerably easier to understand how to ask research questions and design social research when the problems of data analysis and the shaping of research outcomes have been addressed. Learning social research methods is iterative, not sequential, and this means that you will need more than one run through of the complete materials to understand properly each of the topics that make up social research.
Learning by doing
This subject guide has to convey knowledge of research methods by means of addressing required readings and assessing students ability to reproduce descriptions, evidence and arguments under examination conditions. This has to be the case in the nature of the degree programme offered by the University of London. It is not possible to assess students ability to apply what they have learnt through projects or coursework exercises requiring the collection and analysing of data. There is no reason why you should not practise qualitative observation and perhaps some restricted form of simple counting in your everyday life. We will be requiring you to undertake social research practice yourself at some stage. However, I would encourage you to begin to address your everyday life as an apprentice social researcher. There is no reason why you should not become a trained observer of social life, developing a systematic approach to the observation of selected social settings. For example: Social interaction in bus queues and on bus journeys. How do people start up conversations with strangers? What is the content of the conversations you overhear? Collect your local newspapers for several weeks and list the content of stories and the priority given to stories by their placement in the paper. Classification is something you can practice in a wide range of social settings. At a sports event, can you identify and classify different types of sports fans by their demeanour at the fixture concerned? This will be taken up again in the chapter on qualitative research methods.
12
Introduction
13
Practical Read the introduction to the subject guide. Buy Bryman, Social Research Methods
October
Complete reading and note Prepare outline examination taking on Chapters 1, 2, 13 and answers on the nature of 3 of the textbook. social research strategies. Complete reading and note taking on Chapters 14, 15 or 16, 19 of the textbook. Complete reading and note taking on Chapters 21 and 22 of the textbook. Begin observing and making field notes on everyday social situations. Practise answering one question from Part A of the example paper. Prepare outline examination answers on Qualitative Design
November
December
January
Complete reading and note taking on Chapters 7, 5, 4 and 10 of the textbook. Complete reading and note taking on Chapters 11 and 12 of the textbook.
Prepare outline examination answers on Qualitative Analysis. Practise answering one question from Part B of the example paper. Prepare outline examination answers on Quantitative Design.
February
Practise answering two Complete reading and note taking on Chapters 25, 26 and questions from Part B of the example paper 24 of the textbook. Prepare outline examination answers on Quantitative Analysis March Practise answering two questions from Part C of the example paper. Prepare outline examination answers on the Ethics and Politics of Social Research. April Practise full examination paper answers under untimed and then timed conditions.
14
15
Notes
16
Learning outcomes
By the end of this chapter and having completed the Essential reading and activities, you should be able to: explain the difference between research strategies, research designs and research methods discuss the nature of quantitative and qualitative research strategies demonstrate an understanding of the relevance of philosophy (epistemology and ontology) to the practice of social research distinguish between types of methodological awareness and assess their implications for the practice of social research describe the intellectual and social factors that influence the conduct of social research.
Essential reading
Bryman, A. Social Research Methods, Chapter 1, pp.7578 and pp.27984.
Further reading
Bryman, A. Paradigm Peace and the Implications for Quality, International Journal of Social Research Methodology 9(2) 2006, pp.11126. Bryman, A. Quantity and Quality in Social Research. (London: Routledge, 1988) [ISBN 9780415078986]. Devine, F. and S. Heath Sociological Research Methods in Context. (Basingstoke: Palgrave, 1999) [ISBN 0333666321] Chapter 1 and Chapter 10. Robson, C. Real World Research. (Oxford: Blackwell Publishers Ltd, 2002) second edition [ISBN 9780631213055] Chapter Approaches to Social Research. Seale, C. et al. The Quality of Qualitative Research. (London: Sage Publications, 1999) [ISBN 9780761955986] Chapter 3 on Trust Truth and Philosophy and Chapter 4 on Guiding Ideals.
Websites
www.intute.ac.uk/socialsciences/researchtools/ SRA, government social research, BSA
17
Works cited
Bechhofer, F. (1996) Quantitative Research in British Sociology: Has it Changed since 1986? Sociology, 30, and 3, pp.58391. Becker, H.S., B. Geer and E.C. Hughes Boys in White: Student Culture in Medical School. (Chicago; London: University of Chicago, 1961). Berger, P . and Luckmann, T. The Social Construction of Reality. (London: Allen Lane, 1966). Blalock, H.M. (1989) The Real and Unrealised Contributions of Quantitative Sociology, American Sociological Review, 54, June, pp.44760. Blumer, H. What is Wrong with Social Theory, American Sociological Review, 19 1954, pp.310. Cicourel, A.V . Method and Measurement in Sociology. (New York Free Press, 1966). Douglas, J. The Social Meanings of Suicide. (Princeton: Princeton University Press, 1967). Durkheim, E. Suicide: a Study in Sociology. Translated by J.A. Spaulding and G. Gibson (London: Routledge & Kegan Paul, 1952). Edwards, R. Contested Terrain. (New York: Basic Books, 1979). Glaser, B. and Strauss, A. The Discovery of Grounded Theory: Strategies for Qualitative Research. (Chicago: Aldine, 1967). Goldthorpe, J.H. Progress in Sociology: the Case of Social Mobility Research, Sociology Working Papers 200308 at: www.sociology.ox.ac.uk/swp.html Goldthorpe, J.H. The Affluent Worker in the Class Structure. (London: Cambridge University Press, 1969). Hammersley, M. The Paradigm Wars: Reports from the Front, British Journal of the Sociology of Education, 13(1) 1992, pp.13143. Law, J. and Singleton, V . Object Lessons, Centre for Science Studies, Lancaster University (2003) at: www.comp.lancs.ac.uk/sociology/papers/LawSingleton-Object-Lessons.pdf Law, J. After Method: Mess in Social Science Research. (London: Routledge, 2006). May, C. Methodological Pluralism, British Sociology and the Evidence Based State: a Reply to Payne et al., Sociology, 39(3) 2005, pp.51928. Oakley, A. Gender Methodology and Peoples Ways of Knowing: Some Problems with Feminism and the Paradigm Debate in Social Science Sociology, 32(4) 1998, pp.70731. Oakley, A. Paradigm Wars: Some Thoughts on a Personal and Public Trajectory, International Journal of Social Research Methodology 2(3) 1999, pp.24754. Payne, G., M. Williams and M. Chamberlain Methodological Pluralism in British Sociology, Sociology, 38(1) 2004, pp.15363. Selvin, H. Durkheims Suicide: Further Thoughts on a Methodological Classic in Nisbet, R.A. (ed.) Emile Durkheim. (Englewood Cliffs, NJ: Prentice Hall, 1965). Williams, M. Social Research the Emergence of a Discipline?, International Journal and Social Research Methodology, 3(2) 2000, pp.15766. Zimmerman, D. Record-keeping and the Intake Process in a Public Welfare Agency in S. Wheeler (ed.) On Record: Files and Dossiers in American Life. (New York: Russell Sage, 1969) pp.31954.
Introduction
Social research is a practical activity aimed at answering a research question by means of a research strategy, a research design and a method (or methods) of data collection and analysis. Any research strategy has implications for research design and the use of methods of data collection and analysis (research methods). In this chapter, we will be very much concerned with the three components of social research:
18
strategy design method. In learning about social research in this chapter, we will be examining two major research strategies: quantitative and qualitative. When we talk of quantitative research strategies, we mean counting the numbers of people (or other social units such as groups, households, or communities) in defined categories. For example, Durkheim used counts of the number of people (per thousand) in a community who had committed suicide. Qualitative research strategies record and analyse talk and actions, or analyse text that someone has written. This distinction between quantitative and qualitative social research serves as a vehicle to learn how to describe strategy, design and method in social research. Social research is a technical subject and has a precise language. Although not every term used has been defined in the literature in exactly the same way, it is important for the purposes of this course that you are both clear and consistent in the use of technical language. Accordingly, this chapter includes a large number of definitions that you will need to: remember understand apply. Activity: starting a glossary You will find it useful to construct your own glossary that will force you to define key concepts and will be a reminder of their precise meanings. Social research has a technical language that you must use with precision. You may keep your glossary on cards with a card index system or you may order them in a Microsoft Word file. To understand what social research is about you must clearly understand what is meant by each of the following: a research strategy a research design a research method. We will now examine each of these in turn.
Research strategies
A research strategy is a collection of philosophical and theoretical commitments that may influence decisions made about the research design and the choice of specific methods of data collection and analysis. Strategy also relates closely to the questions the research can address and determines the type of findings that can result from the research. In this chapter, I will be using the distinction between qualitative and quantitative research strategies to introduce you to the defining characteristics of social research. Qualitative research strategies are meaning-centred and informed by the interpretivist tradition in social theory. In this tradition, the description and explanation of the social world must refer to subjective meanings. Interpretivist research aims to reveal the meaningful aspects of cultures as in the case of social anthropology and Chicago School of sociology. This tradition has used ethnographic methods to reveal the meaning of other cultures, including sub-cultures in urban and rural areas. Ethnography involves observing communities, interviewing individuals, reading the
19
text found in documents, and collecting this data directly in the natural context shared by those being studied. This use of the natural context is essential if the researcher is to learn to see the social world as it is seen by those being studied. The aim of ethnographic methods is learning to see the social world as it is seen by those being studied. Qualitative research often aims to explain social action in terms of the subjective meanings of the actor and the constructions placed upon their actions by themselves and other actors. Ethnography is a way of accessing those meanings. Thus if we attempt to understand the reason why some young women decide to have children (the problem of teenage pregnancy), we must refer to the frame of reference that informs such decisions. Although constructed as a problem by policy makers, teenage pregnancy may represent an attractive change of lifestyle for young women since it can lead to greater autonomy and status. An interpretivist sociological explanation provides a meaningful account of the action concerned. For quantitative social research, the social world consists of contingent relationships between well-defined concepts. Quantitative research strategies emphasise the measurement of concepts and can be described as variable-centred in contrast to meaning-centred. Examples of variables include the annual income of a household or the number of children in a household. A sample of data collected for these variables will tell us what the averages are and how typical they are. If we are using systematic randomly sampled data, we can estimate the accuracy of our descriptions for the population as a whole. There may be contingent relations between variables, for example, income may influence the number of children in a household. There is no certainty or determinism, only defined odds, that an income level will be associated with the number of children. These contingencies are often relatively stable, but may vary between regions and may change over time. In sophisticated data analysis, we would want to assess: the explanatory power of each variable how much confidence we can have in the estimates of explanatory power. Subjective meanings are not necessary for these contingencies to operate effectively. After all, the consequences of smoking were experienced by those who died of lung cancer whether or not they subjectively attached that risk to their smoking behaviour. So far, we have seen that qualitative and quantitative social research have very different core themes: one is meaning-centred, and one is variablecentred. These distinctions are not mutually exclusive in research practice, but here they are used to help us identify the differing aims of each of the two major research strategies. Qualitative research seeks to reveal subjective meanings in cultures and sub-cultures, while quantitative social research seeks to establish contingent relationships between variables. Activity: identifying types of research projects Using the short summaries of research projects in the Bryman textbook place the research projects described in Box 2.17 p.44 and Box 2.24 p.49 in the cells in the following table: Qualitative research Descriptive research Explanatory research Table 1.1 Summaries of research projects
20
Quantitative research
correlates these summary scales with other social variables such as gender, income, social status, etc. that explain variations in attitudes to political parties. Attitudes are assumed stable social objects that can be used to explain why people support and vote for political parties. For interpretivists, multi-dimensional attitude scaling imposes the researchers construction of meaning on the subjects own perspective. Surveys pre-structure the subjective in order to develop standardised measures. Using the qualitative research strategy one could first explore the assumed coherence stability of attitudes by analysing their expression in specific contexts such as a focus group. We will be exploring the validity of attitude measurement in a subsequent chapter, but it is worth considering why most political parties now use focus groups rather that relying solely on survey methods to establish public opinion on political issues. Focus group methods collect data on the expression of attitudes in a more natural setting of a group discussion. The attitudes expressed to a survey interviewer may be influenced by the specific context of the interview. In survey research, the attitude is expressed in the researchers words as defined in the questionnaire, and the research cannot measure how stable or sincerely held the attitude or opinion is. Focus group methods access attitudes largely in the form of the subjects own words, and the method allows for the possibility that attitudes can be measured by the developed position on an issue taken by a group after some discussion. The more natural the research situation, the more valid the respondents measured attitudes are likely to be. Qualitative research strategy Quantitative research strategy Defining characteristics Descriptive aims Meaning-centred Detailed descriptions of core meanings of a culture. Interpretation of the meaning of action. Theory generation by means of revealing or reporting the constructions of reality in talk and text including official documents. Interpretivism. The human nature of social objects requires reference to the subjective meaning when explaining social action. Theory is generated by analysis of the subjective. The social world is constructed in our attempts to understand it and act upon it. Symbolic Interactionism to post-modernism. Social structure is formed by institutionalised meanings and constructs. Variable-centred Summary descriptions (averages and variation) of social variables and the degree of confidence in these estimates. Theory testing via the establishment within a specified context of contingent relationships between variables taken as representing theoretical constructs. Positivist or post-positivistic. Theoretical models are developed by deduction and tested by measurement of concepts and applying criteria of causal inference. There is a knowable social world independent of our constructions of it. Structural behaviourism. Meanings and action derive from occupancy of structured social locations.
Epistemology
Ontology
Theoretical perspectives
22
It is possible that survey methods can measure some well-defined attitudes. For example, peoples attitudes to the British monarchy may be regarded as unproblematic to measure. However, survey methods may be less suited to establish attitudes to topics that are more remote from everyday experience and less well defined, such as the use of stem cell research in medicine. This example flags up the possibility that relevance of ontological commitments to research practice is itself an empirical question. In their study of alcoholic liver disease, Law and Singleton describe how the patient experience and organisational processes that are supposed to structure pathways through diagnosis, treatment and outcome are in fact ill-defined with unstable and problematic connections between activity and outcomes. In these circumstances, they argue, orthodox approaches to social research may not capture what they call a mess (Law and Singleton, 2003). However, it is incorrect to assert that mess characterises every social context. Some social contexts are highly structured with well-defined social processes that secure a fairly stable and regular set of outcomes. For these contexts what he terms standard research methods remain important (Law, 2006, 4). We can think of gender socialisation and education as involving relatively well-defined institutionalised social processes that are productive of stable and regular outcomes. Goldthorpe (2003), for example, has used a case of social mobility research, to react against the abandonment of traditional (quantitative) research strategies. Some ontological assumptions suit some but not all research topics. We have established that the two major strategies of social research have contrasting stances on both epistemology and ontology. The third component of the research strategy concerns the relationship between theory and research.
qualitative research that aims to test theories using a realist ontology and deductive logic. Activity: clarifying key concepts in social research Having read and made notes on Brymans Chapter 1, what do you understand by the following? Epistemology Ontology Induction Deduction. Try to remember definitions then check with your own glossary or notes taken from Bryman.
Research design
A research design provides a framework for the collection and analysis of data. This framework references three dimensions: the scale of data collection (includes defining units and spatial limits) the use of time order the use of inter-group comparisons or the comparative dimension.
Scale
Research designs can be large scale or small scale depending on what the key requirement is: an in-depth study of a small number of people (usually to generate ideas) or a large-scale study of systematically selected subjects that can be taken as representative of a wider population. Most research designs are a compromise between these equally desirable requirements. In an ideal world, it would be desirable to have many detailed observations on a large number of subjects, but this would be very expensive and impractical to organise. As we shall see throughout this course, practical matters compromise the pursuit of ideals in social research. Since the scale of research is measured in units, a research design must define what the unit of analysis is. A unit might be a person or a household, a school or an area of a country in quantitative research. In qualitative research, the unit of analysis might be a person to be interviewed, but it could also be a wide range of other units such as a fragment of talk or text or one of a number of documents or narratives. Scale determines the actual or potential number of units to be included in the research. Units are selected within a specific spatial context that forms an important part of a research design. The research design will thus include a definition of the space within which the data is collected. This might be a nation-state, a region, a town, city, or a smaller location such as an organisation or part of an organisation such as medical clinic or classroom.
Time order
Time order refers to measurement of change or collection of data over time. We can study classroom behaviour of children over a six-month period or we can survey opinions from one year to the next to see if people change their attitudes to political issues. The changes recorded in the data can be used to establish the sequence of events, for example, first, this happens, then that happens, or if this happens, then that follows, in most cases. In qualitative research, we might want to identify stages in professional socialisation. For example, are professional ideals in medicine developed, reinforced or abandoned during professional training (Geer, et al., 1961)?
24
In quantitative social research, it would be necessary to collect data over time if we wanted to know whether good teaching practices produced high achievement or whether the presence of high-achieving students stimulated the adoption of good teaching practices.
Comparison
A basic feature of most social research is comparison or implied comparison. Systematic comparison is a basic feature of much social research. The term comparative method is often restricted to studies comparing two or more national cultures, but in its broader sense it is a feature common to all social research. Even when there is no direct comparison involved, implicit comparison is used. For example, when studying the informal sub-culture of male schoolchildren we may be comparing them with girls even though girls are not included in the research design. Making comparisons between groups often begs the question: why are the differences observed? If two groups have different ways of organising marriage partnerships, the question arises as to why this is. The answer may lie in another characteristic of the group or the way it has evolved in its own specific context. Comparison group research is a characteristic of the classical experiment. In a randomised clinical trial (RCT) for a new drug, volunteers are randomly allocated to two groups. First, the health status of each group would be assessed and then over time both groups take a unit of treatment (e.g. a pill). The treatment group are given the drug under trial; the other group, known as the control group, receive a placebo. After the course of treatment is finished, both groups are assessed for health status. If the treatment has been effective, the health status of the treatment group should have increased significantly more than that of the control group. Because we have selected the subjects at random, the only difference between the groups was whether they took the treatment or the placebo, all other rival explanations for differences in health outcomes having been eliminated by random allocation of volunteer subjects to each group. Comparison group research is also undertaken in non-experimental designs. Comparison groups allow the researcher to assess the impact of variables in context or show how qualitative sequences may vary across contexts. For example, in the 1980s, the dramatic increase in levels of unemployment in the UK prompted a number of qualitative studies of the impact of high unemployment. One such study by Avtar Brah reports interesting insights into the experience of Asian Youth (Brah, 2005). It was not part of her research aims, but the use of a comparison group of white youths might have shown the Asian experience to be distinctive. Comparison groups are useful to establish whether findings are common to all groups or are unique features of specific groups. Activity: choosing a research design Part 1 Why would a researcher adopt each of the following dimensions of research design? a small-scale study a large-scale study a comparative study a non-comparative study a study that collected data over time a one-shot or cross-sectional study.
25
Part 2 The Goldthorpe et al. study of The Afuent Worker is described by Bryman (2004, 56) as a case study design since it was based on a single town Luton. They concluded that the privatised community experience in Luton was associated with instrumental collectivism or the use of trade union power for monetary gain rather than pursuit of wider political goals. How might the addition of comparison groups of Liverpool and East London car workers have compromised or enhanced the findings of the Luton study? Similarly, if they had returned to the same research sites three years later, what would a time order dimension have added to the possible conclusions? Research design involves the following three components: Scale that determines the number of units to be selected; a sampling component that defines both the unit(s) of analysis and the ways in which these are to be included in the study; a spatial element that defines where the research is to take place. A time component that defines the beginning and end of data collection and any stages in between. The use of comparison or comparison groups. As we shall see later on in this chapter, it is unrealistic to view social research designs as flowing directly from a set of theoretical and philosophical commitments. It is a useful learning device to see the link between research strategy and research design as this is reflected in the separate components of research design. Table 1.3 shows how the components of research design vary according to the research strategy deployed to answer the research question. Research design components Research question Qualitative research strategy e.g. how do disciplinary incidents get raised and resolved in the classroom? Quantitative research strategy e.g. what social factors explain variations in access to the internet?
Spatial elements: where Small number of school sites: National sample of research activity is located better few but in-depth households: better many with minimum detail Sampling: Non-random sample of Random sample of incidents guided by need for respondents guided by the theory construction need to be representative Discipline incidents in classroom lessons Enough to generate theory Informal theory-guided selection and additions to sample Process-oriented study that may seek to establish stages in a process or a typical sequence of events Not typical Households and the members within the household Enough to be confident in findings Choice of multi-stage, randomly selected sample of households Time order required to establish causality. Annual revisit to panel of households to measure changes over time In experimental designs and comparative studies
a. what units are to be selected? b. how many units are to be selected? c. how are units to be included in the study? Time elements
Comparison groups
status passage. Whether it is concerned with becoming a marijuana user, a qualified medical practitioner, or a Moonie, qualitative research is keen to establish stages in the process of transition from one social status to another. Qualitative research designs are characterised by flexibility and limited structure: the idea is to let analysis and findings emerge from the data over the time-period of the study. Research questions, concepts and theory may not be defined or may only be defined as working framework that is subject to change during the unit of the project. This approach is crystallised in Blumers famous call for sensitising concepts in social research (Blumer, 1954). Concepts and theory are grounded in data because any premature closure of concept formation and theory development is to be avoided. Quantitative social research is most often associated with social surveys. Survey analysis is used most frequently to collect information from large samples of individuals. These samples are selected to be representative of whole populations. Quantitative social research can also take other forms. Government administrative systems routinely collect data and this can be analysed using social research methods. Government statistics are compiled on an area basis and researchers can attempt to establish descriptive and causal findings from an analysis of the data. Quantitative techniques can also be used in the structured observation of social settings and in the content analysis of large text files such as newspaper archives. Even qualitative social researchers can advocate some limited use of counting. In using these data sources, quantitative social researchers are often seen to exemplify the following preoccupations: measurement of concepts to test and refine theories causal inference and model building generalisation of findings replication of studies. Measurement is important in quantitative social research because it is the only way of demonstrating the contingent relationships between concepts. To establish that social class influences life expectancy, it is necessary to measure class by placing individuals within a social class classification scheme. If we then use data measuring the age at which people die, it is possible to see if there are social class differences in the mean age at death. Without measurement, it is impossible to establish contingency. Where there is a lack of measurement, we are unable to test the related theoretical ideas. A practical example concerns car insurance. If insurance companies could get accurate estimates of how well people drive, then they would use this data to predict accident rates and thus determine insurance premiums. Instead they use a number of proxy measures of driving ability such as age, number of previous claims, number of driving convictions etc. In social research we measure education by the qualifications gained but this is not necessarily a precise guide to knowledge and skills. Measurement and the establishment of contingency are not sufficient since such contingencies may reflect chance or spurious relationships between concepts. To establish the validity of a contingency relationship, quantitative social research must establish that the findings meet the criteria of causal inference. A valid model is a set of contingent relationships between concepts that has been established by both measurement and causal inference. What are these criteria of causal inference? They are association, time order and lack of spuriousness.
28
Association: for the units being studied the values of one variable are dependent on the values of another (i.e. there is a contingent relationship between two variables). Time order: or a change in the values of one (so-called independent) variable at t1 results in a change in the value of the other (or dependent variable) at t2. Lack of spuriousness: establishing that the contingent relationship between the dependent and the independent variable is not the product of a third common or antecedent variable. Quantitative social research requires quite complex multivariate analysis to establish that the contingent relationship between two variables is true or to establish in what contexts it is true. Thus we might want to conclude that in most advanced industrial societies (context) there is a close relationship, controlling for other influences (such as age, gender and parental social class), between educational achievement and occupational status. There are two further preoccupations to consider. Quantitative social research often uses research designs based on systematic sampling techniques that ensure that the findings relating to a sample can be generalised to the wider population from which the sample was drawn. Using techniques of statistical inference and significance tests, the degree of confidence in the generalisation can be established. Test results show whether confidence is established at the 95 per cent, 99 per cent or 99.9 per cent levels. Such tests can only be used with systematic random sampling and this is usually found only in quantitative social research. The transparency of data collection techniques and methods of analysis in quantitative social research means that the replication of studies is relatively straightforward. Although replication remains rare in practice, the possibility of confirming or rejecting the findings of previous studies is possible. In qualitative social research, improvements are being made to increase the degree of transparency and the archiving of qualitative data does facilitate replication of the data analysis stage. Quantitative social research is variable-centred. It seeks to test and develop theoretical models of concepts by measurement and causal inference. In measuring concepts such as education, these social objects reflect real underlying causal forces. If they did not then there would be no contingent relationships between measured social objects. In measurement, the research is reifying social constructs, making them object-like and the validity of measurement is confirmed where contingent relationships can be established and thus useful research findings result since our knowledge of the forces of the acting is enhanced. Research method components Methods/instruments of data collection Qualitative research strategy Aims to maximise openness of response in interviews/ unstructured interview/ focus groups. Revealing implicit meanings. Coding text to categories and modelling connections and/or sequences to convey plausibility. Quantitative research strategy Aims to standardise meaning of response in survey questionnaires and records of official activities. Estimating descriptions and (causal) associations between variables to meet criteria for inference.
The research strategies we have outlined in section 1.1 are linked to research designs and to the criteria that are applied to evaluate the validity of social research findings. Our consideration of validity criteria is left to later chapters, as are some of the more technical aspects of research design. Activity: the main components of research design What are the main components of research designs?
take part in research (Oakley, 1999). Social research is a social product shaped by the context in which it is produced. It is influenced by cultural (including intellectual), material conditions and practical constraints and facilities. Although all researchers should strive to implement the ideal research design, it is impossible to do so in practice because the conduct of research involves compromises with circumstances in which it is conducted. It would also be wrong to suggest that all research complied in every way possible with the prescribed ways of conducting research. Type of research Doctoral research The audience and its expectations Funded by individuals and government grants. The need to meet the expectations of available supervisors and internal and external Examiners. Funded by Government Research Councils run by senior academics but also by charities, industries etc. The need to meet the requirements of journal and book publishing referees. The need to meet general research quality standards as applied by academic judges. The need to meet expectations defined by research funded under specific funding programmes such as the e-society programme Government social research Government issues invitations to tender stating requirements. Guidelines govern what projects and methodologies are likely to be acceptable. In market research and public opinion polling strictly commercial considerations will tend to define what research is done and the costs of carrying it out.
Academic research
If sociological research has a qualitative bias this is not the case with social research in general. Much of the research conducted outside academic departments is quantitative. Quantitative sociology is also conducted by sociologists in academic departments other than sociology, such as social policy, health, education, urban studies and human geography. Much empirical research has been marginalised by the sociological mainstream only to flourish in these other institutional contexts (May, 2005). The increasingly technical nature of all social research has given rise to an increasingly large number of Masters degrees that offer social research training. Students taking these degrees are not pursuing a doctorate but preparing to enter the increasingly large number of opportunities to put social research skills in practice. In Britain, the growing importance of social research has been signified by the Labour governments espousal of the doctrine of evidence-based policy. In the employment context, it is not surprising that Williams (2000) has argued that social research has become professionalised. Certainly, there are a number of professional bodies for social research that are involved in developing a body of knowledge and ethical standards. Such bodies are unlikely to pursue one or other research strategy on epistemological or ontological grounds. More recently, there has been some concern within sociology and attempts to re-address quantitative methods, perhaps in response to demands of research funding agencies. It highlights a retreat from the view that research design and methods are inextricably linked to intellectual and philosophical matters alone. These links can be clarified by use of the concept of methodological awareness.
I am not covering two other forms of methodological awareness: scepticism (whereby all methods are treated as equally suspect), and standpoint awareness that views oppressed social groups, such as women (for feminists), or the proletariat (for Marxists), as having privileged insights into the nature of the social.
Principled awareness
Should the practice of social research be strictly informed by answers to philosophical questions about ontology and epistemology? If so then research strategy, by which we mean epistemology, ontology and the theory-research relationship is the foundation of research practice. Research practice informed by such principled methodological awareness must be guided by a specific concept of the nature of the social (ontology) and a specification of the means by which knowledge of the social is obtained and verified (epistemology).
32
Philosophy is therefore given the position of providing the foundations upon which the project of social research must be based. In most sociology methods textbooks, the implied methodological awareness is of this principled type. Bryman makes an important statement when he argues that there is no necessary connection between philosophical positions and the practice of social research. He argues (Bryman, 2004, pp.2125) that practical considerations prevent such a close relationship between philosophy and research practice. Bryman recognises that even the most ardent subscriber to a philosophical position may have to compromise on the ideal research design in order to work around the practical difficulties encountered during the conduct of research. Like Bryman, Seale (1999) is an experienced social research practitioner, and drawing on his own experience of research, he rejects the traditional view that philosophy lays the foundations for research practices. Rather, he argues we should treat philosophical stances as resources to be drawn upon and evaluated during the research process. When preparing to embark on a project and during the conduct of social research, philosophical positions can be assessed for their relevance to the research question and to research practices.
Pragmaticist awareness
One significant constraint on the nature of the research strategy to be adopted, is the research question to be addressed. The form of the research question is not always under the control of the research practitioner, and in these circumstances social researchers may be constrained to choose a strategy that will lead findings that bear directly on the question. Even if the researcher is able to control the specification of the research question, they must keep one eye on the funding for their next project and may see the audience for research findings as a significant influence (if the principle funding agencies are included in that audience). A pragmatic methodological awareness is perhaps becoming more common (Robson, 2002, p.43). It offers the researcher the prospect of choosing the research strategy and research design which offer the most promising outcomes given the nature of the research question and the audience for the research.
Pluralist awareness
Another methodological awareness that avoids the principled adoption of a research strategy is methodological pluralism. One example of this pluralist tendency is the rise and growth of the multi-methods movement in social research (Bryman, 2002, pp.45165). This position insists that both quantitative and qualitative strategies are equally valid since their different instruments of data collection and analysis capture different aspects of the social world. The pluralist awareness requires the use of multi-methods in social research whereas the pragmatic awareness insists that this is an empirical question and depends upon the circumstances. We started this chapter by using the distinction between quantitative and qualitative research as a learning device. We are now in a position to see how such different research strategies relate to significant philosophical and theoretical debates within academic sociology. However, the textbook relationship between philosophy and method is the threatened emergence of forms of methodological awareness that challenge the dominance of the principled awareness. There are a number of influences on the conduct of social research that are encouraging social researchers to adopt a pluralist or pragmatist approach
33
to social research. Rather than be overly concerned with philosophical and theoretical inputs, the modern social researcher may be more concerned with the outputs (the findings and the publications arising from social research) and the outcomes of their research. What is the best strategy to use if we aim to answer the question: What will be the outcome of the research? Is it likely to have positive consequences for the researchers, the funding agencies and those subjects who might gain from the implementation of research findings? These non-philosophical questions shape the choice of research strategy in social research.
34
Academic sociology Theoretical paradigms (structuralisms, interactionisms, marxisms, Foucault and post-isms,)
Philosophies of social science Epistemology and ontology (positivism, empiricism idealism, constructivism, interpretativism, critical realism)
Biographical context Abilities acquisition of craft skills career trajectories, apprenticeship & sponsorship
Formal research training provision MA/MPhil/PhD quantitative qualitative IT provision specialist training initiatives (ESRC)
Professional social research associations (BSA, SRA, MRA,) values, ethics, norms of good practice
Practical constraints and facilities. Contexts of data collection, researchersubject interaction, Access, exchange relationships.
Corporate, private sector and government funding agencies. Definitions of aims, methods, preferred outcomes
Conclusions
In this chapter, we have sought to clarify the nature of social research. We have distinguished between three aspects of social research: research strategy, research design and research method. There are two main research strategies, quantitative and qualitative, and these are associated with different positions on the philosophy of social science: qualitative social research is for the most part interpretivist, constructionist, and inductive; quantitative social research is often depicted as positivistic, objectivist and deductive. Research design is concerned with the scale of data collection, time order and comparison groups. As Bryman (2004, Table 2.1, p.56) describes, qualitative research is usually small-scale, over time and without comparison groups whereas quantitative social research is often large-scale, cross-sectional, and frequently with comparison groups. Recently, quantitative social research has introduced more time-ordered data collection (e.g. panel studies). Research methods are the forms of data collection and analysis that reflect the major preoccupations of the quantitative and qualitative research strategies. Qualitative social research depicts the social as seen through the eyes of subjects, description in context, emphasis on process, flexibility and limited structure, concepts and theory grounded in data. These preoccupations lend it to use methods such as ethnography and in-depth interviews using analytical techniques such as grounded theory. In quantitative social research, the pre-occupations are the measurement of concepts to test and refine theories; causal inference and model building; the generalisation of findings, and the replication of studies. These preoccupations are reflected in the use of large-scale standardised sample surveys and multivariate analysis.
35
We sought to understand the factors that influence the choice of research strategy and associated research design and research methods components of social research. We considered the influence of academic sociology with its tendency to adopt specific philosophical and theoretical commitments. During the paradigm wars, quantitative and qualitative research strategies were depicted as polar opposites and as inherently incompatible strategies. The debate over paradigms assumed the form of a principled methodological awareness defined as the argument that research strategies, designs and methods should follow the implications of theoretical and philosophical positions. Practising social researchers are much more likely to have a pluralist or pragmatic methodological awareness. The multi-methods or pluralist version of methodological awareness argues that all methods capture some aspect of the social and are equally valid as research strategies. Pragmaticist awareness views the choice of method as influenced by the need to secure an appropriate outcome such as the answer to a research question or findings that might improve the social world in some desirable way. Social research is thus more accurately seen as a contested terrain (Edwards, 1979): a workplace for social research practitioners where the influence of dominant paradigms, agencies and audiences for social research, practical considerations and the researchers own biographical context are worked out.
36
Learning outcomes
By the end of this chapter and having completed the Essential reading and activities, you should be able to: distinguish sociological concepts from those used in everyday life discuss the differences between the typical processes of concept formation found in qualitative social research compared with quantitative social research describe the process of coding as a way of classifying data using a theoretical scheme, and as a way of generating general categories by identifying similar contents in the data discuss the differences between induction, deduction and retroduction as descriptions of the process of social research.
Essential reading
Bryman, A. Social Research Methods, pp.6570 and pp.27172 plus indexed references to coding
Further reading
** particularly recommended
Blalock, H. The Operationalism Controversy in Blalock, H. and A.B. Blalock (eds) Methodology in Social Research. (London, New York: McGraw-Hill, 1971) [ISBN 0070941815]. ** Fielding, J. Coding and Managing Data in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Gilbert, N. Research Theory and Method in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Layder, D. Sociological Practice: Linking Theory and Social Research. (London: Sage Publications Ltd., 1998) [ISBN 9780761954293] Chapters 3, 4 and 5. ** Seale, C. Coding and Analysing Data in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986].
37
Websites
Rose, D. and K. OReilly Final Report of the ESRC review of Government Social Classifications. (Swindon: ESRC, 1998) at http:/Governmentstatistics.co.uk
Works cited
Dale, A., S. Arber and M. Procter Doing Secondary Analysis. (London: Allen and Unwin, 1988) Chapters 811. Blumer, H. What is wrong with social theory?, American Sociological Review, 19 1954, pp.310. Burchardt, T. et al. Social Exclusion in Britain 19911995, Social Policy and Administration, 33(3) 1999, pp.22744 Crompton, R. Measuring the Class Structure in Class and Stratification: an Introduction to Current Debates. (Cambridge: Polity Press, 1993). Crompton, R. and J. Scott Class analysis: Beyond the Cultural Turn in Devine, F. et al. Rethinking Class: Culture Identities and Lifestyle. (London: Palgrave, 2005). Jackson, P ., N. Stevenson and K. Brooks Making Sense of Mens Magazines. (Cambridge: Polity Press, 2001).
Introduction
In this chapter, we deal with the basic building blocks of social research: concepts. Concepts are crucial to social research because they are the key to description and explanation: Concepts are linked empirically in various ways to data and these linkages aim to provide descriptions or accounts of the social world. Concepts are linked theoretically to form sequences of defined stages in a social process, or to classify descriptions or typical forms, or linked to form explanations. As in the case of research strategies, approaches to concept formation reflect contrasting ontological and epistemological themes in the philosophy of social sciences. First, we will be looking at what concepts are and how they function in everyday life, and in sociology and social research. This will involve considering how sociological concepts are formed, defined and changed over time. The process of concept formation is rather different in quantitative and qualitative social research and this difference is summed up in comparing the use of two terms: operational definitions sensitising concepts. Of particular importance is the relationship between abstract concepts used in theoretical thinking and empirical data. This relationship between concepts and data is further explored by examining one of the most basic functions of social research: the process of coding data in quantitative and qualitative social research.
What is a concept?
Concepts are so fundamental to our way of thinking about the world that it is all too easy to take them for granted. Here are two different ways of describing concepts: Labels: concepts allow us to label experience by applying the concept to an experienced event, observation and meaning.
38
Categories: concepts divide the social world into fragments that are differentiated. The concept of crime labels a wide range of events that have been observed and have a shared meaning, albeit one that can be contested and refined or expanded. The concept of father is a category that separately defines a person as different from non-fathers; again the concept of father, and what it might involve, are not precisely defined and may change over time. We inherit a vast range of concepts embodied in our language and we use these to participate in the social world. In everyday life, we do not have to use precisely defined concepts. For example, when young people use the word cool to describe actions, it has only a very vague definition or perhaps it is simply a way of affirming that actions are approved to some degree. In social research, we must go beyond everyday concepts and be more systematic in the way in which we define concepts, relate them to empirical materials and deploy them in interpretations of the social world. This does not mean that at all times, in social research, concepts are precisely defined. As we shall see, sometimes it is important to have only working definitions of concepts that are refined in relation to data and only precisely defined at the conclusion of the research process.
Abstraction
An abstract concept can be defined with reference to the specific instances that it labels or categorises. In practice, using empirical referents as examples will help to clarify and convey the meaning of a concept. It may be hard to define sport as an abstract concept other than by using examples such as netball, basketball, soccer, cricket and rugby to convey the meaning. However, my copy of the Pocket Oxford Dictionary (Oxford University Press, seventh edition, 1984) offers a formal, more precise definition:
outdoor competitive activity usually involving physical exertion
The concept of sport specifies an abstract content that can be seen in all the various sporting activities. Even so, it is often possible to think of exceptions to such definitions such as, in this case, indoor athletics. On the other hand, we may quibble as to whether a sport can avoid physical exertion, in which case archery might be excluded from our definition. The abstract version of the concept of sport may be precisely defined as a construct. The concepts of everyday life are directly linked to the empirical things they describe, the constructs of social science require a specification of how they can be measured. For example, the abstract concept of institutionalised racism must have a set of specified measures before it can be recognised and analysed in social research. Concepts transcend, but are related to the world of experience. Dictionaries are continually revised to refine definitions as well as to include new concepts, and the same is true in social research. Activity: defining the concept of household My Pocket Oxford Dictionary (Oxford University Press, seventh edition, 1984) defines household as occupants of a house regarded as a unit. Draw up a list of everyday words that mean the same or have a similar meaning to the word household. What other words describe those living together in households? Then attempt to define the concept of household as it might be used in sociological research. Complete the right-hand column of the table below by inserting a new definition of household, including each of the following possible components of a definition:
39
Definition
Do households have to be defined using all these words? Do we have to allow for variation by using phrases like usually? Will just one or more of the above possibilities provide for a better definition that that provided by my dictionary?
40
define social science as only concerned with the empirical process of measurement has been termed the operationalism controversy (Blalock, 1969). In contemporary social research there is a more developed interrelationship between theoretical constructs and the empirical indicators that measure them. The process of forming operational definitions may also involve deciding on categories after data is collected. A good example of this is the concept of age. Age is a concept that may appear to be unproblematic as it can be measured in years by collecting the date of birth. Complexity arises when deciding how to group people of similar ages to simplify analysis and presentation of findings. The process of conceptualisation involves defining age boundaries of the groups and here there are choices: age as life-cycle where the cut off points that define groups are taken to define childhood, adolescence, adult, middle age, elderly age as maturity is more simply defined as passage of time under 20s, 30s, etc. age as generation defines groups in terms of historical periodisation of cultural and social changes such as in terms of those who reach maturity before the Second World War, after the Second World War, during the Sixties, Seventies, and Eighties. Here the link between theory and measurement is apparent since the choice, in quantitative research strategies, will be based on the causal content that age has for the concept it is being linked with: Variations in health are best explained by maturity. Variations in life cycle might be closely linked to household formation. Variations in political interest and party identification by generational definitions of age. The process of concept formation should not be seen in isolation from such theoretical themes and the need for theoretical explanations. Rather than see causality as the impact of one measured concept on another, sociology is committed to identifying mechanisms that translate the causal components of a concept into their effects on another concept. For instance, if we find that people who do not own cars are more likely to experience ill health, we may not necessarily conclude lack of car ownership is a cause of ill health until we can identify a mechanism that links car ownership to ill health. We can more readily accept the link between social class and ill health because there are numerous risk factors and lifestyle factors linking class position to experience of good or poor health. In the case of housing deprivation, we have mechanisms that allow us to develop a meaningful account of how such deprivation leads to ill health.
avoid unnecessary complexity in theory and interpretation of findings by reducing the number of categories down to a number which is sufficient to make the point being argued combine those categories that are very similar reduce sample size problems when using crosstabulations.
Sensitising concepts
In qualitative social research, there is often a commitment to viewing the social world through the eyes of the people being studied. This implies that as far as possible the social world is to be understood largely in terms of the concepts used by the people being observed. There is the commitment to be appreciative of the culture being studied and reporting back on its conceptual schemes to the wider academic audiences. In this case, the concepts developed would have to be defined broadly in order to accommodate an anticipated range of usages in the data collected and so that the data collected throughout the fieldwork stage of research can influence both the formation and deployment of concepts. Concepts formed in the initial stages of data collection and analysis would only be working concepts or what Blumer (1954) came to describe as sensitising concepts. In opposition to survey research practice, where concept formation processes lead to a well-defined and fixed set of categories, Blumer suggested that social science should not prematurely close down definitions of concepts. Concepts and their definitions should remain open to revision and refinement in the process of data collection and analysis. The idea of sensitising concepts is closely associated with the idea that in qualitative social research, concepts should emerge from the data collected. Bryman uses the example of emotional labour from Hochschild (1983). Activity: understanding an emergent concept Look at Box 13.2 in Brymans textbook: 1. How would you define emotional labour? 2. What examples are given of people who are required to do emotional labour in their occupations? 3. Why is emotional labour more common in gendered occupations? 4. Can you think of other contexts where emotional labour is performed?
43
5. How did the concept emerge from the process of doing qualitative research? A further example is taken from Brymans own research on appraisal schemes in higher education where the concept of procedural compliance emerged to interpret his findings (Bryman, 2004, Box 13.9, p.283).
Conceptual modelling
In the social sciences, the formation of concepts is not a well-structured process with a well-defined end point. For this reason, the idea of conceptual modelling is a preferred way of thinking about the process rather than the closed version of operationalisation and the open version of sensitising concepts. In the practice of social research, we may come to undertake a project with a concept of social exclusion pre-selected in very general terms but sufficiently open for our observations of the social to allow it to be reconfigured. Concepts are likely to remain highly contested.
44
Concepts are abstractions, mental objects/linguistic codes, that describe empirical referents or traces. They can act to select how we see the world (constructionism), but concepts do not encompass the whole of the knowable social world (realism). Concepts are useful descriptions embodied in interpretations/theories/explanations of the social that work for the purposes of those who adopt them until there is a crisis in the conceptual paradigm accepted by a research community (pragmatism). The idea of conceptual modelling recognises that the data may be pre-conceptualised, but not necessarily closed to modification or reformulation. A concept is not taken as reproducing reality but is intended (for some purpose) to be an insightful version of the whole and thus can be evaluated in terms of its purposes (pragmaticism). Concepts that prove not to be useful in some theoretical or practical way are likely to abandoned and replaced.
Conclusions
In this chapter of the subject guide, we have seen that concepts are categories or labels that refer to empirical referents. We have distinguished between theoretical constructs (defined abstractly and can be related to empirical traces or instances) and constructs that are invented to describe empirical traces. These are processes of deductive and inductive concept formation. As we described in Chapter 1, the quantitative research strategy is typically deductive in the approach to concept formation. We can use the example of social survey research to help justify and describe the processes of deductive concept formation. However, we have also shown how in the increasingly popular technique of secondary analysis quantitative social research may involve a process of inductive concept formation. Similarly, the qualitative research strategy is typically viewed and justified as an inductive approach attempting to reveal the social as it is seen by the people who inhabit it. However, we have also seen how in some instances qualitative can be deductive and how it necessarily involves some degree of deduction. An important lesson of this chapter is that although useful as learning devices, the description of the two research strategies by Bryman and others is a necessary oversimplification and thus we must be careful not to apply it to reproduce stereotypes of the research process.
45
46
47
Notes
48
Learning outcomes
By the end of this chapter and having completed the Essential reading and activities, you should be able to: explain and assess the degree to which qualitative research involves the specification of a research design prior to the conduct of data collection and analysis describe the sources of qualitative data such as ethnographic, in-depth interviews, documentary evidence, and participant observation discuss the distinctive approach to sampling that is characteristic of qualitative social research explain and assess the relative importance of concepts such as induction and flexibility in the design of qualitative social research.
Essential reading
Bryman, A. Social Research Methods, Part 3, Chapters 14, 15, 18 and 19. Also Chapters 21 and 22.
Further reading
Byrne, B. Qualitative Interviewing in Seale, C. (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986]. Devine, F. and S. Heath Sociological Research Methods in Context. (Basingstoke: Palgrave, 1999) [ISBN 9780333666326] Chapter 7. Fielding, N. Ethnography in Gilbert, N. (ed.) Researching Social Life. (London: Sage Publications, 2002) second edition [ISBN 9780761972457]. Kirk, J. and J. Miller Reliability and Validity in Qualitative Social Research. (London: Sage Publications, 1986) [ISBN 9780803924703]. Miles, M. and M. Huberman Qualitative Data Analysis. (London: Sage Publications, 1994) [ISBN 0803955405] Chapters 2, 3. Ritchie, J. and J. Lewis Qualitative Research Practice. (London: Sage Publications, 2003) [ISBN 9780761971108]. Tonkiss, F. Focus Groups in Seale, C (ed.) Researching Society and Culture. (London: Sage Publications, 2004) [ISBN 9780761955986].
Works cited
Garfinkel, H. Studies in Ethnomethodology. (Englewood Cliffs, NJ: Prentice Hall 1967). Marshall, C., and G.B. Rossman Designing Qualitative Research. (London: Sage Publications, 1995).
49
145 Social research methods Scott, J. A Matter of Record. (Cambridge: Polity, 1994). Seale, C. et al. Introduction: Inside Qualitative Research in Qualitative Research Practice. (London: Sage Publications, 2004) Silverman, D. Doing Qualitative Research. (London: Sage Publications, 2005) second edition. Skeggs, B. The Formation of Class and Gender: Becoming Respectable. (London: Sage Publications Ltd., 1997)
Introduction
The chapters that make up this second part of the subject guide are designed to assist your learning about qualitative social research. In Chapter 1 we learned that understanding social research involved being very precise about the meaning of each of the following: a research strategy a research design a research method. If you are still not sure about the precise differences between these components of social research you should read the chapter again. In Chapter 1, we also identified the characteristics of the qualitative research strategy, and these are reproduced in Table 3.1. Defining characteristics Descriptive aims Meaning-centred Detailed descriptions of core meanings of a culture. Interpretation of the meaning of action. Theory generation by means of revealing or reporting the constructions of reality in talk and text including official documents. Interpretivism. The human nature of social objects requires reference to the subjective meaning when explaining social action. Theory is generated by analysis of the subjective. The social world is constructed in our attempts to understand it and act upon it. Symbolic interactionism to postmodernism. Social structure is formed by institutionalised meanings and constructs.
As described in Chapter 1, the choice between qualitative and quantitative research strategy has been at the centre of the debate over the methodology of the social sciences. However, in this subject guide, we shall be paying more attention to research questions and the outcomes of social research. In this chapter, we are specifically concerned with one research method, ethnography and the methods of data collection used in ethnography (participant observation, interviews and analysis of documents). Bryman (2004) in his discussion of the preoccupations of qualitative social research, states that ethnography is a method of data collection and analysis that has the following preoccupations:
50
depicting the social world as it is seen through the eyes of subjects providing description in context emphasising process (rather than structure) flexibility and limited structure in the research process developing concepts and theory grounded in data. Any attempt to describe the process of research design in qualitative research is faced with the problem that many advocates of qualitative approaches are committed to the process of induction. A commitment to induction implies that the defining components of the research design and the methods of data collection and analysis, including the research questions and outcomes, are determined during the conduct of the data analysis. In qualitative research research design is a continuous process that is present during most phases of the research process. Thus, it is misleading to think of the conduct of qualitative social research as one that follows a logical three-stage sequence of: 1. research question 2. design of data collection 3. data analysis. Activity: models of the research process Bryman, on p.269 of the course textbook, presents a more complex model of the stages of qualitative research. Examine this diagram now and list the differences between Brymans models of the quantitative and qualitative research process. Whatever the researchers degree of commitment to inductive research, some decisions must be taken before embarking on data collection and analysis. Academic social researchers embark on research projects that are informed by their prior intellectual commitments to study the social world with certain issues or problems or questions in mind. These issues, problems or questions are likely to be linked to general theoretical frameworks and, in some instances, to more specific theoretical themes. All these have implications for the research design decisions and methods of data collection and analysis. Activity: prior intellectual commitments in the research process Examine the summarised study of East End detectives by Dick Hobbs in Devine and Heath Sociological Research in Context and look up the indexed references to Hobbs work in Bryman. What prompted Hobbs to decide on his approach to his study of detectives? All research is likely to be informed by prior experience that will include theoretical thinking and informal analysis of data collected by informal means. Data is routinely collected in everyday life, and some researchers have pointed out the value of investigating topics that arise from personal experience. These make it difficult to determine when the research process actually begins. David Silverman (2005, 150) advises those thinking about doing a social research project to search for data that might inform the conceptualisation and theoretical thinking about a topic, rather than doing data collection and analysis only after a research proposal has been drafted. Before data collection can take place, a topic at least, and perhaps a research question should be defined. Additionally, decisions should be made about which social settings to observe, how these are to be selected or sampled and how data is to be recorded or otherwise assembled.
51
In the case of applying for funding for qualitative social research, the researcher will be constrained to make more explicit the design decisions they will make or at least are likely to make during the unit of the proposed research. Agencies and organisations that fund social research will want to be convinced of the likely success of the research they are funding and they will employ academics and others to review applications or to establish the likely value of conducting the research. According to Marshall and Bensman (1995), qualitative research proposals typically include the following three sections: an introduction containing an overview of the proposal, a discussion of the topic or focus of the inquiry and the general research questions, the studys purpose and potential significance, and its limitations a discussion of related literature that situates the study in the ongoing discourse about the topic and develops the specific intellectual traditions to which the study is linked the research design and methods, which detail the overall design, the site or population of interest, the specific methods for gathering data, a preliminary discussion of strategies for analysing the data and for ensuring the trustworthiness of the study, a biography of the researcher, and ethical and political issues that may arise in the conduct of the study. Marshall and Bensman (2005) also provide a detailed description of a research proposal that is summarised in the following table. Introduction: including an overview, and description of the topic and purposes of the research. The potential significance of the research should be identified. Framework and general research questions: including a review of literature, and theoretical traditions, previous research. This section should conclude with an assessment of theory and its relationship to the proposed research. Design and methodology: including overall approach and rationale; site and method of data selection (sampling), research methods chosen and data-analysis procedures to be adopted. Research staff and resources: including senior researcher experience and performance in qualitative research and a review of ethical and political considerations.
Table 3.2 Sections of a qualitative research proposal
In order to demonstrate the possibility of achieving satisfactory research outcomes, applications for the funding of qualitative research must outline research questions, settings, timings and locations of data collection, and proposed methods of data analysis. The quality of such outlines, along with other factors (including reputation, record of accomplishment of previous research, organisational support, originality or contribution to knowledge etc.) will largely determine the success of the application. However, the proposal should reserve some essential flexibility and openness in research questions and design because these are likely to change during the unit of the research. To reinforce the point as to how much thinking needs to be done prior to the conduct of qualitative research, here are the criteria for evaluating qualitative research, as proposed by Seale et al. (2001), that are of particular relevance to the design stage of qualitative research:
52
The aim and purpose of the proposed research should be described and set in its historical, political, and disciplinary contexts? Does the proposal show how the research will provide new insights in relation to existing knowledge? The rationale for the design of the inquiry should be explained. What arguments are made for a qualitative rather than a quantitative study? The particular contributions made by different methods for collecting and recording evidence should be understood. How good is the rationale provided to justify the choice of methods? A sampling method should be described and justified with reference to ways of gaining access to sources of evidence and the implications of these for the evidence gathered. The researcher should demonstrate openness to emergent issues. The rationale for the choice of analytic strategy should be clear, but an awareness of the potential of other analytic strategies should be demonstrated? The implications of the investigation for broader areas of knowledge and practice (for example, theory, policy, practice) should be explored, and be significant. This list implies that a considerable amount of pre-structuring takes place before the beginning of data collection, and a certain amount of preconceptualisation of issues and problems will be brought to the site of the data collection and analysis. Activity: defining plagiarism How would you design research to reveal that what the university authorities might define as plagiarism is very difficult to distinguish from what students might see as the normal routines of academic work? To explore ambiguity in concepts of plagiarism how would you embark on a qualitative study of an institution of higher education? Would the study be entirely inductive? Would you have a research question to answer? What are the contexts e.g. sites in which you would collect data? What time frame would you adopt to complete the study? Would data be collected in the form of official documents, interviews with student union representatives, interviews with students? Could you use participant observation or observation of meetings? Would focus groups help to explore concepts of plagiarism? How would you record or collect and store the data to be analysed? Finally, how would you argue that your research design was the most appropriate given the topic of the research?
on female offenders, as summarised in Box 15.6, justifies the choice of semi-structured interviews by the nature of the research question and her commitment to feminist research. Apart from choosing between the various types of questions that are available, it is necessary to design the order of the questions that in the case of Charmazs typology (Bryman, 2004, 237) follows a time order. In using a qualitative data collection technique, such as in-depth interviews, there are then a number of design elements that have to be decided upon prior to the conduct of the fieldwork. The data collection may not be fully inductive, although it can be flexible in allowing changes to be made during the course of the investigation. The in-depth interview aims to get close to the subjects view of the social world. It constructs an artificial research situation however skilled the interviewer is in making the subject feel that they are in a natural context. A key influence on the subject in this context will be the character and actions of the interviewer. One way of lessening the possible effect that researchers may have on the subjects prepared and the answers they will give is to use the group interview or, as it is now known, the focus group. This has become a very fashionable technique, but one that has a long history. It involves an unstructured group interview conducted by a moderator. The design issues in the use of focus groups are: How many groups to use? How to recruit members? How to establish rapport among the group? How and when to introduce prompts into the discussion? What degree of moderator involvement? Again, in the conduct of qualitative research practice it is important to recognise that decisions made at the design stage that act to structure the experience of the subjects being studied may impact the behaviour, perceptions and thus the data that will be collected. Activity: designing focus group research Read Chapter 16 of Bryman and retrieve one element of an answer for each of the above questions on the design of focus group research. Focus groups encourage participants to react to each other and to be aware of what others might think and might say. Each subject is sensitised to the social context of the group and usually the selection process ensures that each member is similar in social respect. Minimising the role of the observer or facilitator brings the social interaction within the group to the foreground in something that approximates to a research social situation. The school of naturalism in qualitative research is closely associated with ethnography which involves overt or covert participant observation. Covert observation is used when access would not otherwise have been possible. With overt participant observation, access is more likely to be regulated by a key person or gatekeeper. Activity: types of data in ethnographic studies Read Bryman Box14.1 on p.292 and answer the following questions: What does Taylors study of female drug users reveal about ethnography? What data is collected by those using the ethnographic method?
54
Those about to embark on designing qualitative social research are forced to decide on the balance between naturalistic and artificial forms of data collection. The school of naturalism in qualitative research is closely associated with ethnography and this involves either overt or covert participant observation. With overt participant observation, access to people and places is more likely to be regulated by a key person or gatekeeper who represents the interests of those granting access and possibly restricting the quality of access. Covert observation is often used when access would not otherwise have been possible. Covert observation may interfere with data collection by means of fieldnotes. For Fielding (2001), the construction of fieldnotes is the raison dtre for personal observation since data that cannot be recorded is not collected or analysed. The quality of ethnographical observation is limited by the difficulties of gaining access and the final quality of access. These are described in detail on pp.294300 in Bryman. Activity: alternatives to covert observation in ethnography What are the alternatives to covert observation when designing a study of a closed setting? Give examples from the study of football hooligans described in Bryman, Box 14.4 on p.298.
Using documents
When conducting an ethnographic study a vast range of documentary evidence may be assembled: Personal documents such as diaries and photographs. Official documents from public and private organisational archives. Mass media documents such as newspapers and magazines. In design terms, these have the great advantage of being non-reactive. Documentary evidence has been produced without the researcher being involved in the construction of the data. Researcher involvement may have a part to play where access to the full range of documentation is restricted and in the selection of those documents to include in the study. Confronted with a vast amount of documentary evidence, the selection may proceed by applying Scotts criteria (Scott, 1990): Authenticity or is the evidence genuine? Credibility or is the evidence free from distortion or error? Representativeness or is it typical and can we assess how typical it is? Meaning or can we understand the materials without undue uncertainty? Activity: theoretical sampling and a study of family photographs Look through your collection of family photographs to assess which ones you would select as reflecting a true picture of your family life. Having done this read Bryman pp.38486 and apply Scotts typology of: idealisation natural portrayal demystification. Count how many of each you have selected. Now, go back to your collection and select more photographs to ensure you have equal numbers of photographs in each of Scotts typology.
55
Then answer the following questions: 1. What research question would an analysis of family photographs allow you to answer? 2. How would you go about selecting a sample of family photographs from a series of households in a small community? 3. How might you try to ensure that each of Scotts criteria for selection is covered?
56
Conclusions
In Chapter 1, we learnt that induction is often held to be the defining characteristic of qualitative social research. Its prime commitment was to represent reality as far as possible in the concepts and meanings of those directly involved and to minimise the contribution of the observer to the construction of accounts of the social world. It follows that if research is to be inductive, the degree of openness and flexibility is essential. In practice, openness and flexibility means that the research can operate at the stage of data collection and analysis relatively unconstrained by a pre-determined research design. Design is more of an ongoing process in qualitative studies, one that may continue until nearly the end of the project. Given the commitment to induction and flexibility, qualitative research does not just happen; it is shaped by a context and the choices that have to be made. We have seen that decisions to use a range of data sources, or to make judgments on the use of naturalistic and more artificial devices for data collection, and on the election of data for inclusion in the study, flowing as it does from choices of time-periods of observation and sites of data collection. Where funded research is involved, research proposals are required and these typically include specified research questions and at least an outline of the research design. A less well-specified account of the proposed research will only be allowed when the prior research record of the proposer is thought to be adequate to predict the quality of the research outcomes. Amongst the decisions to be made are: Are multi-sources of qualitative data to be used, as in many ethnographic studies? What balance between naturalistic and artificial research contexts is desirable?
57
How are individuals, items, contexts to be selected for inclusion in the study? Is theoretical sampling to be practiced? Questions remain about the time period of investigation and the sites of data collection. Qualitative researchers often have predispositions to use certain analytical techniques such as grounded theory, conversation analysis, discourse analysis and narrative analysis. Once selected these choices effectively rule out the pursuit of others. An inductive strategy in qualitative research aims to maximise the openness of theory, data sources and data analysis strategies. Openness and flexibility maximise the possibility for emergence. The point about qualitative design is not that it is inductive so much as that it is flexible to allow induction to take place. Even where some initial decisions overly structure data collection, they may be suspended or modified to allow for a line of inquiry stimulated by data collected and/or inductive theory development.
58