research methods

current trends in development research

Research methods are conventionally divided into quantitative, qualitative and participatory each with differing underlying approaches, tools and techniques. Quantitative, qualitative and participatory approaches have different disciplinary origins, developed distinctive tools and each has developed its critique of the other approaches.

In development research quantitative methods have typically been the main focus, with qualitative and participatory methods often relegated to desirable 'frills'. This is partly because of the overwhelming emphasis in many development agencies with economic economic growth and economic dimensions of poverty. In many development agencies the concern with quantification intensified over the 1990s with requirements for performance assessment and targets in logical frameworks. Pressures for quantification have further intensified to demonstrate progress on Millenium Development Goals, 'scaling up' of impacts and macro-level change.

Traditional disciplinary divides are however becoming increasingly breached. Moreover new tools and new solutions to shortcomings of old tools are continually being developed. Increasingly the emphasis is on developing an appropriate integrated mix of research methods.

Quantitative Methods

Quantitative methods as they are commonly conceived derive from experimental and statistical methods in natural science. The main concern is with rigorous objective measurement in order to determine the truth or falsehood of particular pre-determined hypotheses.

• the main focus is on measuring 'how much is happening to how many people'.
• the main tools are large scale surveys analysed using statistical techniques. Quantitative measurable indicators relevant to the pre-determined hypotheses are identified and combined into questionnaires.
• questionnaires are then conducted for a random sample or stratified random sample of individuals, often including a control group.
• causality is assessed through comparison of the incidence of the variables under consideration between main sample and control group and/or the degree to which they co-occur.
• in large-scale research projects teams are composed of a number of skilled research designers and analysts assisted by teams of local enumerators.

For easily accessible overviews of the strengths and pitfalls of different statistical techniques see the website for Statsoft

For access to many further resources see the quantitative methods, statistics and quantitative database sections on the University of Amsterdam 's SocioSite

Return to top

Qualitative methods have their origins in the humanities: sociology, anthropology, geography and history. They differ from quantitative methods in aiming, not primarily at precise measurement of pre-determined hypotheses, but holistic understanding of complex realities and processes where even the questions and hypotheses emerge cumulatively as the investigation progresses.

• typically focuses on compiling a selection of microlevel Case Studies investigated using a combination of informal interviews, participant observation and more recently visual media like photography and video.
• questions are broad and open-ended and change and develop over time to fill in a 'jigsaw' of differing accounts of 'reality', unravelling which may be said to be generally 'true' and which are specific and subjective and why.
• different sampling methods are combined: different purposive sampling techniques, identification of key informants and also 'random encounters'.
• Causality and attribution are directly investigated through questionning as well as qualitative analysis of data. Computer programmes are used to deal systematically with large amounts of data.
• typically requires long-term immersion of a skilled researcher in the field who engages in a reflexive process of data collection and analsysis.

For access to many further resources see the qualitative methods sections on the University of Amsterdam 's SocioSite

The Forum for Qualitative Research website brings together resources and debates in English and other European languages.

For overviews of computer analysis software see Lewins, Ann and Silver, Christina (2004) Choosing a CAQDAS (Computer-Aided Quatlitative Data Analysis) Package: A Working Paper

Return to top

Participatory Methods

Participatory methods have their origins in development activism: NGOs and social movements. Here the main aim is not so much knowledge per se, but social change and empowerment - and this wherever possible as a direct result of the research process itself. In particular it seeks to investigate and give voice to those groups in society who are most vulnerable and marginalised in development decision-making and implementation.

The participatory process may involve small focus groups, larger participatory workshops or individual diaries and diagrams which are then collated into a plenary discussion. Participation (and hence sampling) may be open or carefully targeted to particular social groups. Larger meetings may be subdivided into what are assumed to be more 'homogeneous groups' or groups with complementary information.

Participatory research typically uses and adapts diagram tools from farmer-led research, systems analysis and also oral and visual tools from anthropology, though many commonly used tools have also been developed by NGOs and participants in the field. Use of diagram and oral and diagram tools makes both discussion and analysis accessible to non-literate participants and across language groups. Through sharing their different sources of information participants themselves may increase their understanding of development issues and the problems they face and develop solutions, as well as giving more reliable and representative information to researchers. In some cases local people themselves conduct research following initial design of specific tools and training. Some recent NGO innovations propose doing this on a large scale.

For more discussion and links see participation and participatory action learning pages on this website

towards an integrated methodology

Wherever possible, most research will use an integrated methodology which builds on the complementarities between methods in order to build on strengths, crosscheck and triangulate the information which is most crucial for addressing the particular research questions concerned and also try to disseminate information in different ways for different audiences in order to ensure, as far as possible, benefits for outcomes for participants.

Participatory methods should play a central role at all stages from conception, through piloting and refinement to the research proper and then finally dissemination. Using participatory methods as the 'first port of call', has many advantages in terms of rapidity and reliability of collecting many types of qualitative as well as quantitative information, manageability in terms of time and resources and also its potential for contributing to the development process. Participatory methods are not necessarily a substitute for other methods, but enable much more cost-effective targeting of everyone's time and energy on those areas of the research for which participatory methods are rather more problematic. They are also generally an essential component of research dissemination to those participating in the research, a stage which is commonly ignored and omitted.