ARTICLE 28: Research Methods for Ph. D. and Master’s Degree Studies: Experimental Methods

Written by Dr Hannes Nel


Can lessons from events be gained in a laboratory?

Lessons that will apply to every place and all people?

You be the judge of that after reading this article.

Quantitative experimental methods

The essential feature or experimental methods is that you deliberately control and manipulate the conditions which determine the results in which you are interested.

To achieve this, you will need to introduce an intervention and measure the difference that it makes.

It is mostly used by people doing quantitative research and in natural sciences.

The precondition for research to be experimental is that it should meet the requirements for scientific research.

The requirements for scientific research are that the researcher must have control over the variables, careful measurement, and establishing cause and effect relationships.

An experiment involves making a change in the value of one variable – called the independent variable – and observing the effect of that change on another variable – called the dependent variable.

Using a fixed design, experimental methods can be confirmatory, seeking to prove or disprove a hypothesis, or explanatory, which would be discovering the effect of certain variables.

The variability and validity of experimental methods often depend on the ability to replicate the results from an experiment in similar situations and generalizing the findings to other similar situations.

You can, for example, test the psychological effect of the COVID-19 pandemic on people by asking them relevant questions during an interview.

You can also ask people to complete a questionnaire.

Responses to the questions can be measured in different ways, depending on the type and context of the experiment.

In the experimental methodology, generalization is referred to as “external validity”.

External validity means that the same results will be obtained with different groups and in different places.

You can imagine that external validity will not be easy to achieve.

This is mostly because of the number of different external variables that can influence your experiment.

The less external variables there are, the better will the chances be that you will achieve external validity.

Experiments can be experiments in a laboratory, field experiments or natural experiments.

Regardless of where the experiments are done, they require quite some planning and preparation.

Most, if not all, tests that are done in a laboratory are examples of experimental research.

Experiments in a laboratory

A laboratory experiment would be an experiment that is conducted under highly controlled conditions where accurate measurements are possible.

Laboratory experiments are mostly easy to replicate.

That is because standardized procedures and measurements of variables are used.

There are not many external variables over which you have no or little control.

Laboratory tests, however, do not always replicate real life, with the result that findings will probably not be generalizable outside the laboratory.

Field experiments

Field experiments are done in real-life environments.

The researcher still manipulates the independent variable but in a real-life setting.

Findings in field experiments should be generalizable to real-life settings.

A limitation of field experiments is that there will probably be many external variables that might bias the results.

Consequently, it would be difficult for other researchers to replicate the results.

Natural experiments

Natural experiments are also conducted in real-life environments.

They differ from field experiments in that the researcher cannot manipulate the independent variables – events occur as they would in real life.

Behavior in a natural experiment is more likely to reflect real life because of its natural setting.

It might be necessary for covert observation.

That is, the target group for the research does not know that they are being observed.

The reason for this is that people often behave differently when they know that they are being watched than they would normally behave.

This is even worse if they know that video footage is being taken of them.

Covert research can have ethical consequences which you should consider and remove in advance.

The researcher has no control over external variables that might bias the results.

Experiments in qualitative research

The parallel in qualitative research is when a part or the whole qualitative research study is the subject of attempted replication of a phenomenon or event.

For example, within a single qualitative research study, replication can sometimes be achieved by conducting more than one of the same or at least similar case studies.

The more similar the findings from the cases, the more a replication might be claimed.

However, replication is not always possible because of the dynamic nature of human phenomena.

Paradigmatic approaches that fit well with experimental methods include behaviourism, empiricism, modernism, phenomenology, scientism and symbolic interactionism.


In experimental research:

  1. You, as the researcher, manipulate the conditions for the research.
  2. You must have control over the variables.
  3. Careful measurement of phenomena or behavior will be necessary.
  4. You will need to establish a cause and effect relationship.
  5. External validity will not be easy to achieve.

Experimental methods can be confirmatory or explanatory.

An independent variable is the input variable, whereas the dependent variable is the outcome variable – the result.

It is not easy to achieve external validity with any of the experimental research types.

That is because of differences in variables between the place where the experiment is done and real-life conditions.

Experiments can be experiments in a laboratory, field experiments or natural experiments.

The conclusions of experimental methods can be achieved through qualitative research.


In closing,

I used one additional source of information for this article. That is McLeod, S.A. (2012, Jan 14) Experimental method. Simply psychology: https://www.simplypsychology/org/experimental-method.html. I accessed the source on 10 March 2020.

The main source of information that I use for all my articles and videos on Ph. D. and master’s degree research is my book entitled Preparing Qualitative Research Reports.

Continue Reading

ARTICLE 22: Research Methods for Ph. D. and Master’s Degree Studies

Written by Dr. Hannes Nel


For my first doctoral study, I used a quantitative method, specifically CHAID supported with log-linear analysis and the CATMOD procedure. I chose this statistical method because I intended to send out many questionnaires. I had to ask for the assistance of the statisticians of the South African Council for Human Research to process my completed questionnaires.

Of course, I had to analyze the printouts of the processed data myself, else I would not have been able to come to any conclusions, findings or suggested solutions to my research problem.

In retrospect, I was walking on thin ice by making use of a statistical process about which I knew little.

Most of the methods that I will discuss should be used with qualitative research, although some should preferably be used with quantitative research.

All research methods can be used with a mixed approach.

Quantitative research methods are often dedicated to a specific scientific field of study rather than to a research method.

They sometimes also deviate from the generally accepted structure of a dissertation or thesis.

Research methods overlap with research paradigms and data collection methods. I will discuss them separately in future posts.

Mixed research methods

Research on the doctoral and master’s degree levels is becoming increasingly diverse and inclusive of both qualitative and quantitative research methods.

That is one of the reasons why a mixed-method approach is often followed.

Another reason why students decide to follow a mixed-method approach is that they feel that a combination provides a better understanding of a research problem than just focusing on qualitative or quantitative research methods.

Some students argue that a mixed-method enhances the validity and accuracy of data and the relevance and value of research findings.

The overarching premise is that the integration of two or more methods should facilitate the achievement of research objectives.

The decision of whether to integrate multiple methods depends on a combination of the research objectives, the resources and time available, your philosophical assumptions, your knowledge of research methods and your style or personal preferences. The people who will read your dissertation or thesis, especially your sponsor, but also your study leader, will also have an influence on which research method or methods you will use.

Different methods can be used as a form of triangulation in the hope that the findings from the different methods will corroborate one another.

Where will you discuss the research method that you will use?

You will probably discuss the research method that you will use in the second chapter of your dissertation or thesis.

You should explain:

  1. The method that you will use without trying to “teach” your study leader or other readers the intricacies of research. They probably know more about this than you.
  2. Why you chose the method.
  3. The reason why you chose the method.

You should also provide a detailed description of the components of the methodology of your choice. I will discuss the components of different research methods in future posts, seeing that they differ for different research methods.

The method that you choose must enable you to progress from a problem situation to a solution.

As far as possible you should do as much of the work as possible on your own.

For my second doctoral study, I used Atlas.Ti to code and analyze the data that I collected, mostly through interviewing. I used grounded theory, which is a qualitative research method. This time around I attended a short course on Atlas.Ti at the University of Guildford in Surrey before embarking on my studies, so that I could do all the work on my own.

Your method or methods can consist of a combination of literature study, models, arguments, mathematical proofs, surveys, case studies, experimentation, etc.

Did you notice that I am integrating research methods with data collection methods here?

The chapter on your research methods is mostly not difficult to write.

After all, you will have all the information from a book or books that you consulted.

Even better, your study leader can suggest a method that will work well with the nature of your research.

In addition to the purpose of your research, you should also consider your knowledge, experience, strong points, weak points and the data that you will have access to.

Don’t choose a quantitative method if you feel uncomfortable with calculations and statistics or if you do not have access to laboratories or other scientific equipment that you will need.

Don’t choose a qualitative method if you will not have access to or feel comfortable with people as your target group.


You will probably use quantitative research methods if you feel comfortable with figures and statistics, if you have access to the laboratories or other scientific equipment that you will need for your research and if the topic of your research is such that it can be investigated quantitatively.

You will probably use qualitative research methods if you feel comfortable when working with people when the topic of your research is suitable for qualitative research and if you need to solve immediate social or community problems.

You will probably use mixed research methods if there is a need to collect and analyze quantitative and qualitative data.

You can use more than one research method, and some of them might be quantitative while others are qualitative in nature. Different methods can be used as a form of triangulation, which should give you a measure of corroboration of your data and findings.

Keep in mind that research methods, paradigms, data collection methods and data analysis methods are all parts of the research process.

You will probably discuss the research method that you will follow in the second chapter of your dissertation or thesis. Although it is an important chapter, you should not go into too much detail.

Continue Reading