You are here:
Toolbox >>Understanding Development>>Understanding Research

Understanding Research

What is in this guide?

This guide is meant to provide a basic understanding of research and its role in development. It has the following sections:

  1. What is research?
  2. How is research used
  3. Different ways of doing research
  4. How do we do research?
  5. Tools to analyse research
  6. Important research terms

  1. What is research?

Generally, research is the organised and systematic method of finding answers to questions.  It is systematic because it is a process broken up into clear steps that lead to conclusions. Research is organised because there is a planned structure or method used to reach the conclusion.  Research is only successful if we find answers, whether we like these answers or not. Development research is focussed on relevant, useful and important questions. If there are no questions, there can be no research.

If government, business, institutions, labour, organisations and society in general are to function efficiently and effectively, it is important that the decisions they make are based on valid and reliable information and thorough analysis. The search for this information is referred to as the research process. There maybe an existing body of evidence (prior research, studies etc) you can make use of. If there is not, there is a need for research. 

For example, the Department of Health in planning a HIV and AIDS prevention programme may have to ask some of the following questions before agreeing on and rolling out the programme. The Department of Health may have an existing body of evidence that assists in finding the answers to some of these questions whilst others may require research.

Questions that could be researched are:

Once a decision is made that research is required, the Department of Health must decide on the research methods and process that will be used to answer the questions.

  1. How is research used

As indicated above, the primary purpose of research is to find answers to questions. Research allows us to find the right solutions to key issues in our communities by:

Here are some examples of questions that research will help to answer in community development work:

Research can play an important role in winning support for a programme or cause (sometimes called advocacy.) It helps make a case through strengthening arguments, providing information, and outlining cost benefits.

Research can confirm what you were already sure of.

Often people have firm beliefs about particular issues, but when they have to argue their case they lack reliable information to back up their beliefs. Research helps to clarify and strengthen beliefs especially in the face of opposition and doubt from others. Whilst research can confirm your views, it is important that the researcher remains open-minded and impartial even when the results fail to confirm your views.

Research can give your views and arguments substance.

Research produces hard facts that could support your arguments and beliefs.

Research gives you new information.

Research often throws up other facts which you may not have been aware of that helps to strengthen, or even change, your arguments and beliefs. These facts make it easier to plan programmes and ensure that interventions are effective.

Research can show you what is most likely to address your issue successfully.

Research may provide key information that will enable you to develop clear strategies.

Research can provide you with anecdotes and examples to use.

In addition to providing statistics, research provides you with real life experiences that are more convincing than statistics organised into graphs and tables. For example, parts of a research report on poverty in a rural community can deal with actual case studies that will have a great impact on readers.

Research allows you to make cost-benefit arguments.

Often people are convinced that a programme or project justifies high amounts of money being spent. Research can confirm if this is correct or suggest other ways for the money to be spent.

  1. Different types of research

There are different types of research activities than can assist you in undertaking research. In this section we touch on some basic methods:

  1. Desktop research refers to seeking facts, general information on a topic, historical background, study results, etc., that have been published or exist in public documents. This information can be obtained from libraries, newspaper archives, government, university, websites, NGOs and CBOs etc. For example, most research undertaken by government departments is easily accessible on the internet or at government offices.
  1. Interviews and conversations are used when you want to find out the community's past experience with an issue. The best way to find it is usually by talking to individuals. For example, it may be best to talk to local community leaders to collect information on the history of an area. You may use informal conversations, structured interviews, or a combination to get as much information as possible. Sometimes, it is useful to ask questions of a group of people (for example, in a workshop situation), as this can stimulate different views and discussion.
  1. Surveys are used if you want to know what most people in the community think or feel about an issue.  For example, how many people would take advantage of a service if it were available? A survey is a way to reach a lot of people in a short space of time. A survey usually consists of a list of simple questions on a topic, and may include some chance for respondents to express a broader opinion or comment on the issue. You can conduct surveys by post, phone, in person, by e-mail, on a web site, or by making them available in public places (See the CDW Skills Manual, p. 42, for more on doing your own surveys.) 

The information that is collected through these methods is either quantitative or qualitative in nature. Quantitative research depends on numbers and statistical procedures. For example, a household income survey is a quantitative survey that looks at the average household income in an area.

Information can also be qualitative - based on observations of behavior, participants' reports of how they or their lives have changed, etc.  For example, the Department of Transport may want to find out the impact of its Arrive Alive Campaign using a qualitative study in areas that are usually high accident zones. Here the researchers will observe how road users conduct themselves in these areas and in addition speak to a sample of them to find out what impact the television and radio advertisements have had on their behavior on the roads.

Some studies seek to understand cause and effect - what causes something else to happen or the connection between two factors. For example, the Department of Water Affairs may want to find the cause of certain rivers being highly polluted and the effect this has on the lives of people living along these rivers.

Some studies are conducted to find answers to very specific questions.  For example, the Department of Agriculture may want to find out whether maize or pumpkins are the best crops to grow in a particular area as part of a poverty alleviation project.

  1. Community investigation involves going to an area to establish facts about a specific problem or state of affairs. For example, councilors sitting on a Transport and Roads Standing Committee of a local municipality may visit an area to establish the extent of the problem in relation to the occurrence of potholes and sinkholes. This would require the councilors to talk to residents, examine the road conditions and make notes. This would enable the Standing Committee to make a decision based on facts they established first hand.

In some cases, community investigation may require actual detective work. For example, an advice office worker may want to report a factory owner to the Department of Labour for locking his workers overnight in the factory. The advice office worker may have to take photographs of the locked doors or film the workers operating behind locked doors. S/he can also get affidavits from worker to establish the facts and present a case to the Department of Labour.

    1. Case studies that describe the experience of individuals or groups affected by an issue can be very effective for research that aims to change a situation or influence decision-makers. Politicians and the public are often more easily swayed by stories they can identify with than by statistics. Finding people who can provide convincing first hand information is an important part of research. Key people and activists in the target community are good sources for finding people who can provide first-hand information. For example, the Department of Social Services and Population Development may want to find out the impact of drought on a rural village. Local people who have lived in the area for a long time will be able to provide compelling stories and anecdotal information on the impact of drought in the area and how the community has coped with this over the years.


  1. How do we do research?

As discussed previously, research is a systematic and organised process. It is about collecting information that answers a question. Throughout this process the researcher has to ensure that information is gathered in a systematic and accurate manner.

Information gathered must be cross-checked by using other sources and references, even when the researcher is convinced that the information already obtained provides a good answer to the question asked.

Below are guidelines and steps for a general research process, no matter the type or method or research being undertaken.

Step 1: Identify and define the issue or question

This step assists in identifying the problem or issue that requires research. For example, South Africa has a high incidence of road death. Research already done shows that around 10,000 people are killed in road accidents each year. Now we need to find out what are the causes and impact of the high incidence of road deaths. We need to know what other facts and evidence already exist so that we can build on that.

Step 2: Deciding direction by identifying a focus and refining the question

In this step we set out the aims and objectives of the research. For example, the aim of the research may be to “assess the social and economic impact of road accidents on the South African population”. The aim of the research may provide a title for the research, i.e. “The causes of road accidents and the social and economic impact on the South African population”.

A clear aim will make it easier to develop objectives for the research, for example:

The objectives will help you to decide which questions need answers. For example,
“What are the three most common causes of road accidents?”

Step 3: Organising the work plan to answer the questions

This step entails organising the work and choosing the methods that will be used to conduct the research. A terms of reference (ToR) should be drawn up that that spells out the work needed. This is usually given to the researcher who must then prepare a proposal about how they will go about doing the research. A ToR usually has the following sections: background, research objectives, methodology to be used, resources to be used (people, money for travel, etc), and timeframes for completing the project (broken down into phases, e.g. when the fieldwork will be completed, when the report will be written).

Step 4: Collecting information to help answer the question

This step entails the actual collection of information. This may require fieldwork. The research example on “The causes of road accidents and the social and economic impact on the South African population” is a huge and difficult one that will require lots of resources. For example, 80,000 fieldworkers were employed to conduct the 2001 Census. In this case the fieldworkers were called enumerators.

Other research may be conducted on a much smaller scale and may include a team of 5-10 people and the amount of resources required would be less.

Step 5: organise the information collected and discard what is not needed

This phase entails organising and analysing the information gathered in the previous step. To analyse means to make calculations, such as adding up the different responses so as to get a full picture of the situation. For example, after analysis it might be that 70% of those that were interviewed may have been driving over the speed limit of 120km/hr. The analysis may be in the form of tables, graphs, percentages, etc. Similarities may emerge. For example, the incidence of road deaths may be higher during rainy days. Similarly patterns may start to emerge. For example, the occurrence of drunken driving is higher during weekends and at the end of the month when people get paid.

Step 6: Drawing conclusions

This step entails discussing the findings and drawing conclusions.
Findings are often in table, graph, numeric or percentage form. The discussion involves using words to describe the findings. The discussion section is where the researcher gives opinions based on the findings of the research. The researcher then draws conclusions and may make recommendations based on the findings. The conclusion may be that “Road deaths are mainly caused by drunk drivers, drunken pedestrians, un-roadworthy vehicles and poor driver behaviour. The main economic impact is on the productive workforce due to high death rate and the more than 100 000 economically active people who are disabled annually.  Impact is most severe on individual families affected. “

Step 7: Writing a research report

The writing of a report is important as it leaves a body of evidence that can be used by politicians, planners, community organisations and future researchers. A report generally has six sections: introduction, literature review, methodology, research results, discussion, and conclusions and recommendations (for more information, see section 5 of this chapter).

Step 8: Reflecting on and evaluating the work done

This step entails reflection to decide on what action is needed and what steps should be taken to use the research effectively. This may include a plan for communicating the research results to community members and decision makers. More research may also be needed to answer new questions thrown up by the research done.

  1. Tools for analysing research

Government, community based organisations and other stakeholders work to create a better life for all. These stakeholders are often confronted by issues that require new or further research. With a number of institutions of higher learning and other private and public organisations conducting research, it can be difficult to differentiate between “good” and “bad” research. Research often informs the decisions that are made. Poor research leads to poor decisions.

This section provides some basic tools to help us understand research and the terminology and presentation used.

Research documents are usually organised in a similar way:

Here are some questions that can be used to decide whether research findings are valid, relevant and useful.

Analysing the introduction

As we noted, the introduction should explain the purpose of the research study. The introduction should be factual and not contain the researcher’s opinions. It should also focus on the aims and objectives of the research.

In analysing the introduction the following should serve as a guideline:

Analysing the literature review

Does it give an overview of the existing body of relevant evidence and draw out the lessons we can learn from this evidence?

Analysing the method section

The method section should provide details of how the research was conducted, including descriptions of the participants and setting, the research and intervention procedures, and the data collection and analysis procedures.

The method section must provide clear details of the following:

Analysing the results section

The result section contains detailed statistics and complex terms. Many people tend to avoid the results section and move on to the discussion section for this reason.  This is dangerous as it is meant to be a factual statement of the data whilst the discussion section is the researcher's interpretation of the data. 

Understanding the results section may lead the reader to differ with the conclusions made by the researcher in the discussion section.

The results section must provide:

To understand how research results are organised and presented, you must understand the concepts of tables and graphs. Below we use information from the Department of Education’s June 2003 publication “Education Statistics in South Africa at a Glance in 2001” to illustrate the different ways the information can be organised.


Tables organise the information in rows (horizontal/sideways) and columns (vertical/up-down). In the example below there are two columns, one indicating the learning phase and the other the percentage of students in that learning phase within ordinary schools in 2001.

Learning Phase

Percentage of students











Table: Percentage Distribution of Learners by Phase in Ordinary Schools in 2001

This table indicates that for every 100 learners in ordinary schools in South Africa, a little more than two (2.3) were in pre-primary grades, 63.3 were in primary schools, and 34.3 were in secondary schools.

When you deal with big populations, findings are usually presented as percentages (or how many in a hundred). This makes it much easier to compare between different groups or issues.

Types of graphs

Graphs help provide a visual representation of information, usually statistics. There are three main types of graphs: pie charts, bar graphs and line graphs.

Pie charts

A pie chart is like a big pie or cake that is divided into slices. Each slice represents one of the categories of your findings and it clearly shows the size of each category in relation to the others.

The information in the table above on the “percentage distribution of learners by phase in ordinary schools in 2001” can be organised in a pie chart that provides a visual description of the information (see next page). Pie charts show the relationship of each part (learning phase) to the whole (total percentage of students)

Bar graph

Bar graphs also compare values in a category or between categories. A bar graph allows you to compare more than one category of information. The bar graph below makes a visual comparison of male and female learner in ordinary schools and compares them for each grade in 2001. The categories in this case are male and female learners, and the different grades.

On graphs there are always two lines – one on the left and one at the bottom of the graph. They form an L and the left vertical line is called the x – axis and the horizontal line the y – axis.

In the example, below the x – axis indicates the number of students whilst the y-axis indicates the different learner grades.

Line Graphs

Line graphs are another way to show the relationship between two or more different categories. The same information from the bar graph can now be illustrated differently to show other information that is not easily visible from the bar graph. For example, the line graph at first glance immediately says to the reader that there are more female learners than male learners from Grades 8 to Grade 12 whilst there are more male learners than female learners from Grades 1 to 5. 

Parts of a bar and line graph

Understanding the discussion section

This section should provide an interpretation of the results. The reader must ensure that the researcher has not misrepresented or misinterpreted the results or applied them to a different setting from which the research was undertaken. The reader has to look out for the limitations of the study, implications for practice, and future research needs. 

Understanding the conclusions and recommendations section

In this section we need to ensure that the conclusions and recommendations arising from the results and discussion sections provide a clear framework to address issues. In addition we need to check on what follow up action is required and other research areas that have been thrown up
  1. Research terminology

Much of the terminology that researchers use is unfamiliar to others. In this section we explain the terms most commonly used in research.

Mean and median

For each of these terms, we will use the following set of nine numbers to explain the basis of our calculations:
1, 2, 3, 4, 5, 6, 7, 8, 54

Mean -The arithmetic mean is a commonly used term and is also referred to as the average. The mean is worked out by adding the numbers in a series and dividing the total by the number of items in the series. Adding our nine numbers and dividing by nine results in a mean (or average) of 10.

1+ 2 + 3 + 4 + 5 + 6 + 7 + 8 + 54 = 90
90 / 9 = 10
Mean = 10

Median -The median is the value which lies at the middle of a distribution: that is, 50% of the values are above (7, 8, 9, 54) and 50% (1, 2, 3, 4) below.
1 2 3 4 5 7 8 9 54
Median = 5


The group of subjects (people) from whom the data are collected.
Sampling Error

Maximum Sampling Error (MSE) is the highest possible percentage that the findings could be out by. For example, you might see the following statement in a research report: "Results are subject to a maximum sampling error (MSE) of + 5% at the 95% confidence level." This MSE tells you that the chances are 95 in 100 that the results are within 5 percentage points, higher or lower, of the true percentage for the entire population. The bigger your sample the smaller the MSE will be.

Statistically significant

This means that the research results are not likely to have occurred by chance. For example, research reports show that more people are applying for child support grants in ward 12 Phalaborwa after an education and awareness program run by the local councillor, and that these findings are "statistically significant." This means that the researcher is reasonably sure that increase in the number of people seeking child support grants was influenced by the education and awareness programme. If research findings are not statistically significant, any increase reported may be due to chance, rather than a result of the intervention.


This means to take a general set of facts and break them into smaller, more meaningful pieces. For example you can find that 40% of people are unemployed in an area. When you break it down into gender you will find that only 20% of men are unemployed, but 60% of women are unemployed. Your approach to dealing with unemployment will change. You can also disaggregate for age, class, educational level, etc.


This means to take some proven facts and to make a prediction based on them. You could take the track record of Pirates and Chiefs and make a prediction that they will end up in the top half of the results table next year. This is extrapolation.


This is a prediction of what the outcome of the research may be.

References and further reading

For each section we provide details of reference sources used in compiling this document.

  1. What is research?
    Unpublished paper “Research for a changing South Africa” PhD Student Noel Chellan – University of Kwazulu Natal - 2004
  2. How research is used?
  3. Different ways of doing research
  4. How do we do research?
  5. Tools to analyse research– section on Parts of a bar and line bar
  6. Important research terms


Understanding Poverty and Development  |   The Millennium Development Goals   |  Understanding Basic Economics   |  Sustainable Development   |  Understanding Globalisation  |  Understanding Research  |  Project Management Guide PDF  |  How to achieve Black Economic Empowerment (BEE) PDF  |  Government spending and income [PDF]

This material may not be used for profit without permission from ETU