Practical Research 2  One-shot Case Study Research Design

In this type of experimental study, only one dependent

ESSENTIAL ELEMENTS OF THE group or variable is considered. The study is carried out
after some treatment which was presumed to cause
RESEARCH METHODOLOGY
change, making it a posttest study.
Research Methodology- addresses the questions. “How
 One-group Pretest-posttest Research Design:
does the researcher answer questions stated in Chapter
This research design combines both post-test and
1?"
pretest studies by carrying out a test on a single group
Research Design - This refers to the overall strategy
before the treatment is administered and after the
that you choose in order to integrate the different
treatment is administered. With the former being
components of the study in a coherent and logical way,
administered at the beginning of treatment and later
thereby ensuring you will effectively address the
at the end.
research problem.
-It is a very important aspect of research methodology
 Static-group Comparison:
that describes the research mode (whether it is
In a static-group comparison study, 2 or more groups
qualitative research or quantitative research, or if the
are placed under observation, where only one of the
researcher will use a specific research type e.g.
groups is subjected to some treatment while the other
descriptive, survey, historical, case, or experimental)
groups are held static. All the groups are post-tested,
Respondents of the Study. This describes the target
and the observed differences between the groups are
population and the sample frame.
assumed to be a result of the treatment.
Instrument of the Study. It describes the specific type
 Quasi-experimental Research Design
of research instrument that will be used such as a
The word "quasi" means partial, half, or pseudo.
questionnaire, checklist, Questionnaire-checklist,
Therefore, quasi-experimental research bears a
interview schedule, teacher-made tests, and the like.
resemblance to true experimental research but is not
Establishing and validating reliability. The instrument
the same. In quasi-experiments, the participants are
must pass the validity and reliability tests before it is
not randomly assigned, and as such, they are used in
utilized.
settings where randomization is difficult or impossible.
Statistical Treatment. One of the many ways of
This is very common in educational research, where
establishing the objectivity of research findings is by
administrators are unwilling to allow the random
subjecting the data to different but appropriate
selection of students for experimental samples.
statistical formulas and processes.
Some examples of quasi-experimental research design
THE DEFINING FEATURES OF THE DIFFERENT TYPES
include; the time series, nonequivalent control group
OF QUANTITATIVE RESEARCH
design, and the counterbalanced design.
The Experimental Designs
True Experimental Research Design
Experimental research is a scientific approach to
The true experimental research design relies on
research, where one or more independent variables are
statistical analysis to approve or disprove a hypothesis.
manipulated and applied to one or more dependent
It is the most accurate type of experimental design and
variables to measure their effect on the latter. The
may be carried out with or without a pretest on at
effect of the independent variables on the dependent
least 2 randomly assigned dependent subjects.
variables is usually observed and recorded over some
The true experimental research design must contain a
time, to aid researchers in drawing a reasonable
control group, and a variable that can be manipulated
conclusion regarding the relationship between these 2
by the researcher, and the distribution must be
variable types.
random. The classification of true experimental design
The experimental research method is widely used in
includes:
physical and social sciences, psychology, and education. It
 The posttest-only Control Group Design: In this design,
is based on the comparison between two or more
subjects are randomly selected and assigned to the 2
groups with a straightforward logic, which may,
groups (control and experimental), and only the
however, be difficult to execute.
experimental group is treated. After close observation,
Mostly related to a laboratory test procedure,
both groups are post-tested, and a conclusion is drawn
experimental research designs involve collecting
from the difference between these groups.
quantitative data and performing statistical analysis on
 The pretest-posttest Control Group Design: For this
them during research.
control group design subjects are randomly assigned to
Types of Experimental Research Design
the 2 groups, both are presented, but only the
Pre-experimental Research Design
experimental group is treated. After close observation,
In pre-experimental research design, either a group or
both groups are post-tested to measure the degree of
various dependent groups are observed for the effect of
change in each group.
the application of an independent variable which is
 Solomon four-group Design: This is the combination of
presumed to cause change. It is the simplest form of
the pretest-only and the pretest-posttest control
experimental research design and is treated with no
groups. In this case, the randomly selected subjects are
control group.
placed into 4 groups.
Although very practical, experimental research is lacking
The first two of these groups are tested using the
in several areas of the true-experimental criteria. The
posttest-only method, while the other two are tested
pre-experimental research design is further divided into
using the pretest-posttest method.
three types
 Characteristics of Experimental Research  Internal validity refers specifically to whether an
 Variables experimental treatment/condition makes a difference or
Experimental research contains dependent, independent not, and whether there is sufficient evidence to
and extraneous variables. The dependent variables are support the claim.
the variables being treated or manipulated and are  External validity refers to the generalizability of the
sometimes called the subject of the research. treatment/condition outcomes.
The independent variables are the experimental
treatment being exerted on the dependent variables. Factors that jeopardize internal validity
Extraneous variables, on the other hand, are other  History--the specific events which occur between the
factors affecting the experiment that may also first and second measurement.
contribute to the change.  Maturation--the processes within subjects that act as a
 Setting function of the passage of time. i.e. if the project lasts
The setting is where the experiment is carried out. a few years, most participants may improve their
Many experiments are carried out in the laboratory, performance regardless of treatment.
where control can be exerted on the extraneous  Testing--the effects of taking a test on the outcomes
variables, thereby eliminating them. of taking a second test.
Other experiments are carried out in a less controllable  Instrumentation--the changes in the instrument,
setting. The choice of setting used in research depends observers, or scorers which may produce changes in
on the nature of the experiment being carried out. outcomes.
 Multivariable  Statistical regression--It is also known as regression to
Experimental research may include multiple independent the mean. This threat is caused by the selection of
variables, e.g. time, skills, test scores, etc. subjects on the basis of extreme scores or
Example: characteristics. Give me forty worst students and I
The effect of positive reinforcement on one's attitude guarantee that they will show immediate improvement
to excel in school. right after my treatment.
Cream formulation of an effective mosquito repellent: a  Selection of subjects--the biases which may result in
topical product from lemongrass oil (Cymbopogon the selection of comparison groups. Randomization
citratus) Stapf. (Adeniran,Oluremi & Fabiyi, E.. (Random assignment) of group membership is a
(2012).Cream formulation of an effective mosquito counter-attack against this threat. However, when the
repellent: a topical product from lemongrass oil sample size is small, randomization may lead to
(Cymbopogon citratus) Stapf.. J. Nat. Prod. Plant Simpson’s Paradox, which has been discussed in an
Resour..2.322-327. Retrieved from earlier lesson.
https://www.researchgate.net/publication/284079450_  Experimental mortality--the loss of subjects. For
Cream_formulation_of_an_effective_mosquito_repellent_a_ example, a Web-based instruction project entitled
topical_product_from_lemongrass_oil_Cymbopogon_citratus Eruditio, started with 161 subjects and only 95 of
_Stapf) them completed the entire module. Those who stayed
ANALYTICAL INVESTIGATION OF THE EXTRACT OF in the project all the way to the end may be more
LEMON GRASS LEAVES IN REPELLING MOSQUITO motivated to learn and thus achieve higher performance.
(Elizabeth, Ojewumi & Banjo, M & Oresegun, M &  Selection-maturation interaction--the selection of
Ogunbiyi, Ajibola & Ayoola, Ayodeji & Awolu, Olugbenga comparison groups and maturation interaction which
& Ojewumi, Emmanuel. (2017). may lead to confounding outcomes, and erroneous
ANALYTICAL INVESTIGATION OF THE EXTRACT OF interpretation that the treatment caused the effect.
LEMON GRASS LEAVES IN REPELLING  John Henry effect--John Henry was a worker who
MOSQUITO.8.10001009.Retrieved from outperformed a machine under an experimental setting
https://www.researchgate.net/publication/315575384 because he was aware that his performance was
ANALYTICAL INVESTIGATION OF THE EXTRACT OF compared with that of a machine.
LEMON GRASS LEAVES IN REPELLING
MOSQUITO/citation/download Threats to external validity
DETERMINATION OF MOSQUITO REPELLENT Sampling bias-The sample is not representative of the
PROPERTY OF THE population
CRUDE ETHANOLIC EXTRACT FROM THE LEAVES History-An unrelated event influences the outcomes
OF(FAMILY BASELLACEAE)AGAINST AEDES AEGYPTI Experimenter effect-The characteristics or behaviors of
(https://www.academia.edu/8415695/DETERMINATION the experimenter(s) unintentionally influence the
_OF_MOSQUITO_REPELLENT_PROPERTY_OF_THE_CRUD outcomes
E_ETHANOLIC_EXTRACT_FROM_THE_LEAVES_OF_BASE Hawthorne effect-The tendency for participants to
LLA_ALBA_FAMILY_BASELLACEAE_AGAINST_AEDES_AE change their behaviors simply because they know they
GYPTI) are being studied.
Factors Jeopardizing Internal and External Validity Testing effect-The administration of a pre-or post-test
Please note that the validity discussed here is in the affects the outcomes.
context of experimental design, not in the context of Aptitude-treatment-Interactions between characteristics
measurement. of the group and individual variables together influence
the dependent variable
Situation effect- Factors like the setting, time of day,
location, researchers' characteristics, etc. limit the
generalizability of the findings.
Differences between Experimental and Non-Experimental
Research Example:
1. In experimental research, the researcher can control Yaşar, H., & Sağ san, M. (2020). The Mediating Effect
and manipulate the environment of the research, of Organizational Stress on Organizational Culture and
including the predictor variable which can be changed. Time Management: A Comparative Study With Two
On the other hand, non-experimental research cannot Universities.SAGE
be controlled or manipulated by the researcher at will. Open.https://doi.org/10.1177/2158244020919507
This is because it takes place in a real-life setting, Adams, R. V., & Blair, E. (2019). Impact of Time
where extraneous variables cannot be eliminated. Management Behaviors on Undergraduate Engineering
Therefore, it is more difficult to conclude non- Students' Performance. SAGE Open.
experimental studies, even though they are much more https://doi.org/10.1177/2158244018824506
flexible and allow for a greater range of study fields. Kalogeropoulos, A., Negredo, S., Picone, I., & Nielsen,
2. The relationship between cause and effect cannot be R. K. (2017). Who Shares and Comments on News?: A
established in non-experimental research, but it can be Cross-National Comparative Analysis of Online and Social
established in experimental research. This may be Media Participation. Social Media Participation. Social
because many extraneous variables also influence the Media + Society.
changes in the research subject, making it difficult to https://doi.org/10.1177/2056305117735754
point at a particular variable as the cause of a Nathan,N.,Zeitzer, J. A survey study of the
particular change association between mobile phone use and daytime
3. Independent variables are not introduced, withdrawn, sleepiness in California high school students. BMC Public
or manipulated in non-experimental designs, but the Health 13, 840(2013).https://doi.org/10.1186/1471-
same may not be said about experimental research. 2458-13-840
Non-experimental designs are research designs that https://bmcpublichealth.biomedcentral.com/articles/10.11
examine social phenomena without direct manipulation 86/1471-2458-13-840#citeas
of the conditions that the subjects experience. There is Olutola, Adekunle & Thomas,& Olatoye, Rafiu &
also no random assignment of subjects to different Ademola, Olatoye. (2016). Assessment of Social Media
groups. As such, evidence that supports the cause-and- Utilization and Study Habit of Students of Tertiary
effect relationships is largely limited. Institutions in Katsina State. JOURNAL OF
There are two main types of non-experimental designs: EDUCATION AND PRACTICE.7.178-188.
comparative design and correlational design. https://www.researchgate.net/publication/309310376
In comparative research, the researcher examines the Assessment of Social Media Utilization and Study Habit
differences between two or more groups on the of Students of Tertiary Institutions in Katsina State
phenomenon that is being studied. For example,
studying gender differences in learning mathematics is DIFFERENT TYPES OF VARIABLES
comparative research.
The correlational design is a study of relationships
between two or more constructs. A positive correlation Variable- An entity that can take on different values.

means that high values of a variable are associated with An aspect of a theory that can vary or change as part

high values of another variable. For instance, the of the interaction within the theory. Anything that can

academic performance of students is positively related affect or change the results of a study. Anything that

to their self-esteem. On the contrary, a negative can vary can be considered a variable; these are needed

correlation means that high values of a variable are to understand differences. May take different forms:

associated with low values of the other variable. For age, gender, IQ level, temperature, lifestyle, and medical

example, teacher-student conflicts are negatively related treatment used. Anything can be a variable. Something

to the student’s sense of belonging to the school. that a researcher is interested in. Classifications of

(Frey, B. (2018). The SAGE encyclopedia of educational Variable:

research, measurement, and evaluation (Vols. 1-4).  Numeric- variables with values that describe a

Thousand Oaks,, CA: SAGE Publications, Inc. measurable numerical quantity. Answers the question

doi:10.4135/9781506326139) how many? and how much? For Quantitative data.

In Survey Research. A survey is intended to acquire Continuous/ Interval Variables- assume any value

information from people concerning the predominance, between a certain set of real numbers depending on the

distribution, and interrelations of variables within the scale used. A value between the whole numbers. E.g.,

identified group of people. In simple terms, surveys are time, age, height, and weight. Discrete Variables- can

done to gather evidence of people's knowledge, opinions, only assume any whole value within the limits of the

attitudes, and values on various issues and concerns. given variables. Whole numbers. E.g., Class attendance,

In Descriptive Research. It aims to define the existing no. of the house in an area, and no. of children in the

condition of a classified variable. This type of research is family.

intended to offer methodical data or systematic  Categorical- variables with values that describe a quality

information about a phenomenon. It does not usually or characteristic of a data unit. Anseres the question

start with a hypothesis but rather after systematic what type? And which category? For Qualitative data.

data gathering since the test of the hypothesis comes Ordinal Variables- can take a value that can be logically

from data analysis. Gathering of data necessitates ordered or ranked. Organized/ranked. Nominal Variables-

thorough selection not only of the unit studies but the values that cannot be organized in a logical sequence.

measurement of each variable as well. Classification. Dichotomous Variables- variables that

 represent only two categories. Polychotomous Variables-
variables that have many categories.

 Experimental- Variables that determine a causal

relationship. Independent Variables- Causal variable.
Presume to cause changes in another variable. Usually
manipulated in an experiment. Dependent Variables-
Effect Variable. Variables that change because of
another variable. Usually affected by the manipulation
of the independent variable. Variables that are
monitored in an experiment. Control Variables- variables
that are held constant. Help to identify the possible
differences in the outcomes as a result of controlling
certain variables. Moderator Variables- a variable that
delineates how a relationship of interest changes under
different conditions or circumstances. Maybe
Quantitative or Qualitative in nature. Extraneous
Variables- variables that are already existing during an
experiment. These variables could influence the results
of the study. As much as possible, must be controlled
because they can offer an alternative result.
 Non-experimental- variables which cannot be
manipulated by the researcher. Predictor Variables-
variables that change /affect other variables in a non-
experimental study. Criterion Variables- Variables that
are influenced by the predictor variable in a non-
experimental study.