Chapter 2: Apply scientific research methods

Scientific research is fundamental for advancing agriculture and extension services in Kenya, enabling professionals to develop evidence-based solutions to challenges such as pest control, soil fertility, and farmer advisory methods. Applying appropriate research methods equips extension officers and agricultural scientists with tools to collect reliable data, analyze trends, and recommend interventions that improve productivity and sustainability. This chapter explores key scientific research methods and sampling techniques relevant to Kenya's agricultural sector, ensuring that research findings accurately represent diverse farming communities and ecosystems.

2.1 Research methods

A research method is the way a researcher gathers evidence to answer a research question. The choice of method decides what kind of data are collected, how reliable they are and what can be concluded from them. An extension officer who wants to know why farmers do not use a recommended practice needs a different method from a researcher who wants to know which of three feed rations gives the highest milk yield. This topic gives the terms used in research and then describes the five methods an agriculture and extension technician applies.

2.1.1 Definition of terms

A shared vocabulary lets researchers, farmers and reviewers understand each other. The terms below are used with the methods that follow.

Research

Research is a systematic process of collecting and analysing information to increase knowledge or solve a problem. In agriculture it might be a study of how crop yield responds to different irrigation methods.

Scientific method

The scientific method is a structured approach to inquiry that moves through observation, a question, a hypothesis, testing, analysis and a conclusion. Extension officers use it when they test a new farming technique under controlled conditions before recommending it.

Research method

A research method is a specific procedure for collecting data, such as a questionnaire, an interview, an experiment, a survey or observation.

Data

Data are the facts, measurements or statements collected in a study. Primary data are collected by the researcher directly from the source, such as milk yields recorded on a farm. Secondary data were collected earlier by someone else, such as rainfall records or published yield figures.

Qualitative research

Qualitative research explores a subject through non-numerical data, such as words from interviews and discussions or notes from observation. It is used, for instance, to understand why livestock keepers are slow to adopt a new breed.

Quantitative research

Quantitative research gathers numerical data that can be analysed statistically to find patterns or test hypotheses. Measuring the yield difference between fertilised and unfertilised maize plots is an example.

Variables

Variables are measurable factors that can change within a study. The independent variable is the one the researcher changes or compares, such as the type of fertiliser, and the dependent variable is the one that responds, such as plant growth.

Population

The population is the whole group of interest from which data may be collected, for example all farmers growing a particular crop in a defined area.

Sample

A sample is a smaller group selected from the population so that the study can be done within the time and money available, and its results are used to say something about the whole group.

2.1.2 Types of research methods

The method must suit the question. Questionnaires and surveys reach many people, interviews go deep with a few, experiments test cause and effect, and observation records what people and crops actually do. Researchers often combine methods so that one makes up for the weakness of another. The design of each method when a proposal is written is covered in Designing data collection method; here each is looked at as a method to be applied in the field.

2.1.2.1 Questionnaires

A questionnaire is a written set of questions that respondents answer themselves or that an enumerator reads out and fills in. It is suited to collecting the same information from many farmers cheaply, for example the area under a crop, the inputs used and the problems faced.

1. Decide exactly what information is needed and write only the questions that supply it.
2. Use simple, clear words in a language the respondents understand, with one idea per question.
3. Test the questionnaire on a few farmers who are not in the study and correct any unclear question.
4. Train the people who will administer it so that each asks the questions in the same way.
5. Give each respondent the purpose of the study and ask for consent before starting.
6. Check each completed form on the same day for gaps and unreadable entries.

Questionnaires are quick and allow comparison between respondents, but they depend on the respondent's honesty and memory and give little room for explanation.

2.1.2.2 Interviews

An interview is a conversation in which the researcher asks questions and records the answers. It may be structured, with fixed questions, or semi-structured, with a guide and room for follow-up, and it may be one to one or a focus group discussion with a small group of farmers. Interviews are chosen when reasons and opinions matter, for example why farmers have stopped using a certain seed variety.

1. Prepare an interview guide that lists the topics and key questions.
2. Choose a quiet place and a time that suits the interviewee, and avoid busy field hours.
3. Introduce yourself, explain the study and ask permission to take notes or to record.
4. Ask open questions, listen without interrupting and use short follow-up prompts to get detail.
5. Keep the interview focused and within the time agreed.
6. Write up or transcribe the interview soon afterwards while it is fresh.

Interviews give rich detail, but they take time, reach few people and can be affected by the way the interviewer asks.

2.1.2.3 Experiment

An experiment is a method in which the researcher deliberately changes one factor and measures the effect on another while keeping everything else the same. It is the method for showing cause and effect. A field trial that compares different seed rates of one crop, or a feeding trial that compares two rations, are typical agricultural experiments.

1. State the hypothesis and decide the treatments to be compared, including a control that receives no treatment.
2. Select uniform plots or animals so that differences come from the treatments and not from the starting condition.
3. Assign treatments to plots or animals at random and repeat each treatment on several units.
4. Apply the treatments in the same way and at the same time on every unit.
5. Measure the response with the same instrument and method on every unit, and write down the results straight away.
6. Keep all other management, such as weeding and watering, the same across the units.

An experiment gives strong evidence, but it needs land, time and careful control, and it may not be possible when the factor cannot be changed.

2.1.2.4 Survey

A survey is a method for collecting information from a sample of a population to describe it, using questionnaires, interviews or direct measurements. It covers many farms in a short time, such as a survey of how many farmers in an area use certified seed.

1. Define the population and the information needed.
2. Decide the size of the sample and the way it will be chosen.
3. Prepare the tools and train the team that will collect the data.
4. Visit the selected farms or farmers and collect the data in the same way each time.
5. Check the forms, follow up on missing answers and record those who refused or could not be found.
6. Enter the data into a table for analysis.

A survey describes the situation across many farms, though it cannot show cause and effect and the quality of its results depends on the sample.

2.1.2.5 Observation

Observation is the method of watching and recording what happens in its natural setting, without asking questions or changing anything. It is used to record what people actually do, which can differ from what they say. A technician might watch how a farmer applies fertiliser or sprays crops, or record the growth stages of a crop through the season.

1. Decide what is to be observed, where and for how long.
2. Prepare a recording sheet with the items to be checked.
3. Stay as unobtrusive as possible so that the people being observed behave normally.
4. Record the events as they happen and note the date, time and place.
5. Separate what is seen from the observer's own opinion in the notes.
6. Review and tidy the notes at the end of each session.

Observation gives direct evidence of practice, but it is time consuming and the presence of an observer may change behaviour.

Chapter Summary

This chapter explored various scientific research methods essential for conducting effective investigations. It began by defining key research terms and proceeded to examine different research techniques such as questionnaires, interviews, experiments, surveys, and observation, highlighting their unique applications and strengths. The discussion then shifted to sampling techniques, distinguishing between probability methods that rely on random selection and non-probability methods that use purposive approaches. Further, the chapter addressed the identification and use of diverse research materials necessary for gathering information, including encyclopedias, academic papers, books, articles, journals, and reputable websites. Understanding these components equips researchers with the tools to design, implement, and support their studies rigorously. The integration of suitable methods, appropriate sampling strategies, and reliable sources forms the foundation of credible scientific inquiry. This comprehensive approach ensures research outcomes are valid, replicable, and valuable across various fields.