BASIC RESEARCH METHODOLGY

Research Designs
Introduction

The research design is the overarching plan you adopt to combine the many components of the
study in a logical and cohesive manner, guaranteeing that you will effectively address the research
problem; it is the blueprint for data collection, measurement, and analysis. The framework of
research methodologies and techniques adopted by a researcher is referred to as research
design. The researchers' design allows them to use the most appropriate methodologies for
the study and to set up their investigations effectively in the future as well. A research might
be qualitative, quantitative, or combined in design. The type of research (experimental
studies, survey research, correlational, semi-experimental, quasi-experimental review) and its
sub-types (experimental design, defining research problems, descriptive case-study) are
explained by the design of a research subject.

The fourth stage of the eight-stage research process is research design. Before deciding on the type
of design to adopt, the researcher must first identify the research problem, then evaluate relevant
information (review of literatures), and finally specify the objectives and hypotheses. A research
design should constitute at least the following elements to be effective:

A statement of objectives of the study or the research output.

A statement of the data inputs required on the basis of which the research problem is
to be solved.

The methods of analysis which shall be used to treat and analyze the data inputs.

Qualitative research uses quantitative computations to find links between acquired data and
observations. Statistical approaches can be used to verify or invalidate theories about
naturally occurring phenomena. Qualitative research methods are used by researchers to
determine "why" a specific theory exists as well as "what" respondents have to say about it.

Quantitative research is used in situations when statistical findings are required to get
practical information. Numbers provide you a greater perspective when it comes to making
1|Page
important business decisions. Quantitative research methodologies are required for every
organization's progress. When it comes to making business choices, insights derived from
hard numerical data and analysis have shown to be quite useful.

To enhance both the internal and external reliability of the research, quality research typically
assures minimal degrees of bias in the process of collecting data.

Correlational Research Design

Correlational research involves observing the values of two or more variables and determining
the relationships that exist between them. It is the association or link between two or more
entities. A correlational study investigates how one entity influences the other and the changes
that occur when one of them changes. This technique of study is used to investigate naturally
existing correlations between variables. As a result, at least two groups are necessary to
undertake a successful correlational quantitative study. The researcher has no influence over
the variables in this study; the researcher is just attempting to determine whether o r not there
is a link between two variables and makes no effort to manipulate variable rather observe
them ‘as it is’. Correlational research can only tell us whether there's an association between
two groups; they can't actually prove causation.

A correlation coefficient is a numeric value that varies from -1 to +1 and determines the strength and/or
direction of a link between two (or more) variables. There are three types of correlation: positive,
negative and no/zero correlation. If the correlation coefficient is -1, the two variables have a complete
negative correlation, meaning that when one variable (age) increases, the other variable drops (purchase
of sports products). If the correlation coefficient is +1, it means that there is a complete positive
correlation between the two variables, meaning that if one variable (age) increases, the other goes up as
well (purchasing beauty enhancing products). If the correlation coefficient is 0, the two variables have
no/zero correlation, meaning if one variable increases/decreases there will be no change/impact on the
other variable.

For example, assume that a researcher is investigating the correlation between cancer and marriage.
There are two variables in this study: illness and marriage. Let's suppose there's a negative correlation

2|Page
between marriage and cancer. Married people are less likely to acquire cancer as a result of this. This
does not, however, imply that marriage straightaway dodges cancer.

Experimental Research Design

Experimental research, often known as experimentation, is investigation carried out with two
or more variables using a scientific approach. It is where one or more independent variables
are manipulated and applied to one or more dependent variables to measure their effect on the
latter. The first variable is a constant that you could manipulate to see how the second variable
is affected.

Variable is a trait, attribute or a dimension of behaviour, that is capable of taking one or more
than one value or magnitude. It is any stimulus that can be measured.

I. Independent variable – The variable that supposedly produces a change in the other
variable. When manipulating the magnitude of the independent variable, the researcher
expects a change in the phenomenon being investigated.
II. Dependant variable – The variable that is dependant or influenced by the independent
variable.
III. Extraneous variables – Variables that are not part of the research study but may have an
influence/effect on the variables being investigated. For instance – culture, gender,
educational status, etc. can may affect familiarity status & confounding errors.

The impact of the independent variables on the dependent variables is usually observed and
recorded over some time, to aid researchers in drawing a reasonable conclusion regarding the
relationship between these two variable types.

The majority of quantitative research investigations are experimental. Experimental research

aids in the collection of evidence needed to make more informed conclusions regarding your
suggested hypothesis. The success of experimental research typically demonstrates that the
observed change in the dependent variable is completely due to the modification of the
independent variable. The most practical and accurate type of research approach for

3|Page
establishing causality is experimental study design. In the social sciences, this study design
is used to understand and monitor human behavior.

For example, you're curious about the impact of phone use before night on sleep patterns. You
specifically inquire as to how the amount of minutes spent on a person's phone before to going
to bed influences the number of hours they sleep. Therefore, the number of minutes spent on
a person’s phone becomes the independent variable and the number of hours they sleep
becomes the dependent variable.

Difference between Correlational and Experimental

 Correlational research merely demonstrates connections between variables, whereas

controlled experiments prove causation.
 In experimental research, you manipulate an independent variable and quantify its
influence on a dependent variable in an experimental design. Other factors are kept in
check so that they don't have an influence on the outcome.
 You measure variables in a correlational research without changing any of them. You
can see whether your variables change simultaneously, but you can't be certain that
one caused the other to change.
 External validity is strong in correlational research, while internal validity is high in
experimental research.

Quasi-Experimental Research Design

The prefix ‘quasi’ means “resembling”. Except for the fact that they both show the cause-effect
connection between the independent and dependent variables, a quasi-experimental design is quite
similar to an experimental design.

Although quasi-experimental research resembles experimental research, it is not the same. A

quasi-experiment, unlike a real experiment, does not rely on random assignment. Subjects are
instead divided into groups depending on non-random factors. In circumstances where genuine

4|Page
trials are not possible due to ethical or practical considerations, quasi-experimental design is a
helpful technique.

The following are examples of quasi-experimental designs:

1. Nonequivalent groups design: the researcher selects existing groups that appear to be
comparable, but only one of the groups receives treatment.
2. Individuals just below the threshold (minimum differences between two groups) are used
as a control group, while those just over the threshold are used as a treatment group.
3. Natural experiments: In most cases, the researchers have complete control over which
group the individuals are allocated to. A random or random-like assignment of individuals
to the treatment group is caused by an external event or occurrence ("nature").

In field settings, when random assignment is difficult or impossible, quasi experiments are most
likely to be carried out. They're frequently used to assess the efficacy of a treatment, such as
psychotherapy or a learning intervention. Quasi experiments, in terms of internal validity, fall in
between correlational research and true experiments.

For example, you learn that a number of the clinic’s psychotherapists have opted to try out the new
therapy, while others who serve comparable patients have chosen to continue with the old one.
These pre-existing groups can be used to compare the symptom development of individuals who
received the novel therapy vs those who received the normal course of treatment. Although the
groups were not allocated at random, if any systematic differences between them are correctly
accounted for, you may be relatively certain that any differences must be due to the therapy and
no other confounding variables.

Ex-Post Facto Research Design

Ex-post facto research is derived from the Latin term which means "after the event". An
approach in which groups with traits that already exist are compared on a dependent variable is
known as an ex post facto research design. Ex post facto research is considered quasi-experimental

5|Page
since the subjects are not randomly allocated; instead, they are grouped based on a certain attribute/
trait.

It is a quasi-experimental study examining how an independent variable, present prior to the study,
affects a dependent variable. Although independent and dependent variables are studied and
compared across groups, it is not a true experiment because random assignment is not used. The
researchers divide the participants into groups based on the variable that interests them the most.

Ex post facto and true experiments are both seeking to say: this independent variable is driving
changes in a dependent variable. Any experiment starts with a hypothesis that one variable is
impacting another. An experimental group and a control group are used to accomplish this. If
you're testing a new drug, the experimental group will receive the new medication while the control
group will receive the old medication.

A researcher, for example, is interested in how adult weight affects self-esteem. As a result, the
participants would be divided into three groups (underweight, normal weight, and overweight) and
their self-esteem levels would be assessed. Because the groupings were formed using a pre-
existing attribute (weight), this is an ex post facto design.

Difference between Ex post facto and true experiments

Ex post facto designs are distinct from true experiments in that they do not employ random
assignment. Because you're looking at something else, true experiments contain random
assignment. You're looking at a previous variable that was present in the participant ex post facto.

You do not allocate people to an experimental or control group at random in an ex post facto
design. You're putting individuals in a certain category on purpose, based on something they
already have. I use the word 'thing' since it might mean anything from 'must wear spectacles' to
'must be overweight'. You may split the population in any number of ways you like.

Conclusion

The general procedure of the flow of the research for the provided study is specified by the research
technique and design. Using an accurate research design will help your study be successful. Research
studies that are successful and include the least amount of error provide important insights that are free
of bias. Researchers create surveys in order to meet all the characteristics of a research design.

6|Page
References

Research Design - Choose a framework for your study with Voxco's Research Design
Research Guides: Organizing Your Social Sciences Research Paper: Types of Research Designs
Ex Post Facto Research Design definition | Psychology Glossary | alleydog.com, 2022

7|Page