RESEARCH DESIGNS

Research is often viewed as the cornerstone of scientific progress. Broadly defined, the
purpose of research is to answer questions and acquire new knowledge. Research is the
primary tool used in virtually all areas of science to expand the frontiers of knowledge. By
conducting research, researchers attempt to reduce the complexity of problems, discover the
relationship between seemingly unrelated events, and ultimately improve the way we live.
Despite the accessibility and prevalence of research in today’s society, many people share
common misperceptions about exactly what research is, how research can be used, what
research can tell us, and the limitations of research. For some people, the term “research”
conjures up images of scientists in laboratories watching rats run through mazes or mixing
chemicals in test tubes. For other people, the term “research” is associated with telemarketer
surveys, or people approaching them at the local shopping mall to “just ask you a few
questions about your shopping habits.” In actuality, these stereotypical examples of research
are only a small part of what research comprises. It is therefore not surprising that many
people are unfamiliar with the various types of research designs, the basics of how research is
conducted, what research can be used for, and the limits of using research to answer questions
and acquire new knowledge.
RESEARCH DESIGNS
Once the researcher has determined the specific question to be answered and has
operationalized the variables and research question into a clear, measurable hypothesis, it is
time to consider a suitable research design. Although there are endless ways of classifying
research designs, they usually fall into one of three general categories: experimental, quasi-
experimental, and nonexperimental. This classification system is based primarily on the
strength of the design’s experimental control.
To determine the classification of a particular research design, it is helpful to ask several key
questions. First, does the design involve random assignment to different conditions? If
random assignment is used, it is considered to be a randomized, or true, experimental
design.If random assignment is not used, then a second question must be asked: Does the
design use either multiple groups or multiple waves of measurement? If the answer is yes, the
design is considered quasi-experimental. If the answer is no, the design would be considered
nonexperimental.
SO IN GENERAL WE CAN SAY THAT A RESEARCH DESIGN IS:
The research design provides the back bone structure of the study, it supports the study and
hold it's together.
The research design refers to the researcher overall plan for answering the research question
or testing the research hypotheses
A research design is:
• a framework for the research plan of action.
 • a master plan that specifies the methods and procedures for collecting and analyzing
the needed information
• a strategy for how the data will be collected

PURPOSES OF A RESEARCH DESIGN

• It provides the scheme for answering research question.
• It maintains control to avoid bias that may affect the outcomes.
• It organize the study in a certain way defending the advantages of doing while being
aware and caution about potential disadvantages
CHARACTERISTICS OF A GOOD DESIGN
• 1- Appropriateness to the research question.
• 2- Lack of bias.
• 3- Precision.
• 4- Power.
• Research design can be either quantitative or qualitative. Both designs complement
each other because they generate different kinds of knowledge that are useful.

EXPERIMENTAL DESIGNS
KEY ELEMENTS OF EXPERIMENTAL DESIGN
INDEPENDENT VARIABLE: The independent variable is the factor that is manipulated or
controlled by the researcher. In most studies, researchers are interested in examining the
effects of the independent variable.
It is also common for a research study to include multiple independent variables, perhaps
with each of the independent variables consisting of multiple levels. For example, a
researcher may attempt to investigate the effects of both medication and psychotherapy on
symptoms of depression. In this example, there are two independent variables (i.e.,
medication and psychotherapy), and each independent variable could potentially consist of
multiple levels (e.g., low, medium, and high doses of medication; cognitive behavioral
therapy, psychodynamic therapy, and rational emotive therapy). As you can see, things have a
tendency to get complicated fairly quickly when researchers use multiple independent
variables with multiple levels.
DEPENDENT VARIABLE: The dependent variable is a measure of the effect (if any) of the
independent variable. The variable that is measured to see if it changes in response to
manipulations in the independent variable.
CONTROL VARIABLE: Other variables that are kept constant to ensure that any effect on
the DV is due to the IV alone.
CONTROL GROUP: A group that does not receive the experimental treatment, providing a
baseline to compare with the experimental group.
RANDOMIZATION: Randomly assigning participants to different conditions to minimize
bias.
A true experimental design is one in which study participants are randomly assigned to
experimental and control groups.
A random numbers table is nothing more than a random list of numbers displayed or printed
in a series of columns and rows. Typically, computer programs that generate such lists allow
you to request a specific quantity and range of numbers to be generated. To use a random
numbers table to assign study participants to groups, you must first determine the exact
numbers that you will use to determine the assignments.Most studies now rely on the use of
random numbers tables to help them assign their research participants.
For example, if you have three groups or conditions, you may use the numbers 1, 2, and 3.
Alternatively, if you were assigning participants to two groups, you could use the numbers 1
and 2, or simply odd or even numbers, to determine the group assignments.
A random numbers table is nothing more than a random list of numbers displayed or printed
in a series of columns and rows. Using a random numbers table is one effective way to
randomly assign participants to groups within a research study.
No matter what method is used to randomly assign participants to groups in a study, random
assignment helps ensure that any changes in the dependent variable are likely due to the
independent variables, rather than other outside factors. For example, a researcher testing a
treatment's effectiveness wants to ensure the experimental group (receiving the treatment)
and the control group (receiving a placebo or other intervention) start out similar. If the
groups differ from the beginning, any changes observed by the end could be due to those
initial differences, not the treatment itself. While the researcher could try to match groups on
various factors to make them more alike, it's impossible to control for every individual
difference that might impact the study’s results.
For example, a researcher might match two groups on factors like age, gender, race, and
socioeconomic status, assuming these might affect the treatment outcome. While this can
make the groups more alike, they may still differ on other important, unmeasured factors like
intelligence, motivation, or past treatment experiences. These unknown differences make it
harder for the researcher to confidently say that any changes are due to the treatment itself.
Random assignment, however, helps spread individual differences more evenly across groups
so that the main difference between them is the treatment being tested.
This is why, whenever possible, researchers prefer to use a randomized experimental design.
Simply put, it gives the researcher the most control over the study and the strongest ability to
make confident conclusions about cause and effect. Generally, randomized experiments can
be done using one of three main designs: (1) a randomized two-group posttest or pretest-
posttest design, (2) a Solomon four-group design, or (3) a factorial design.
TYPES OF EXPERIMENTAL DESIGN
➢ True-experimental design
➢ Quasi experimental design
➢ Pre-experimental design

TRUE -EXPERIMENTAL DESIGN

1-Pretest-posttest Control-Group Design

➢ A traditional, classical design, this procedure involves random assignment of
participants to two groups. Both groups are administered both a pretest and posttest to
both groups, but the treatment is provided only to experimental Group A.
➢ the experimental and control groups are both randomly assigned from the sample that
was randomly selected.
➢ the researcher observes the two groups to determine effect of manipulation (post test)
➢ The treatment is under control of the researcher, the dependent variables is measured
twice before and after manipulation of the independent variables

Group A R _____ O _______ X________ O

Group B R _____ O _________________ O
Advantages of Pre test – post test control group design
ADVANTAGES
➢ The use of a control group minimizes threats to internal validity by allowing for a
clear comparison between the experimental and control conditions.
➢ Randomization ensures that each participant has an equal chance of being assigned to
any group, helping to reduce selection bias and increase the fairness of the
experiment.
➢ It allows researchers to examine and measure the changes that occur as a result of the
manipulation of the independent variable
DISADVANTAGES
➢ The results may not be easily generalized beyond the specific groups involved in the
experiment, limiting external validity.
➢ Participant attrition (mortality) may impact one group more than the other, potentially
leading to biased results if the drop-out rate differs across groups.

2-Posttest-Only Control-Group Design

➢ This design controls for any confounding effects of a pretest and is a popular
experimental design. The participants are randomly assigned to groups, a treatment is
given only to the experimental group, and both groups are measured on the post-test.
 ➢ The pre-test can not be established, and also subject response to post test can be
altered by the effect of the pre test, so post test only is performed
➢ The Posttest-Only Control Group Design is a type of true experimental design where
participants are randomly assigned to either an experimental group or a control group.
In this design, there is no pretest or baseline measurement before the treatment.
Instead, the dependent variable (DV) is only measured after the manipulation of the
independent variable (IV). This design is commonly used when a pretest might
influence participants' behavior or when measuring the DV before the treatment is not
necessary.
Group A R _____ X________ O
Group B R ________________ O

Advantages:
Random assignment decreases the threats to validity.
Disadvantages:
Without the effect of pre test the researcher become unable to determine the effect of
the treatment received

3-Solomon Four-Group Design

➢ A special case of a 2x2 factorial design, this procedure involves the random
assignment of participants to four groups. Pretests and treatments are varied for the
four groups. All groups receive a posttest.
➢ Four groups can be used one experimental and three are control, some times are
divided equally two control and the other are experimental .
➢ It is a stronger design than the post test only but require so complicated statistical
analysis.

Advantages : Allow the examiner to examine the effect of the pre test on the post test.
Disadvantages : Mortality can affect the results
Group A R _____ O _______ X________ O
Group B R _____ O _________________ O
Group C R _______________ X________ O
Group D R _____ O _________________ O

General Advantages of Experimental Design

➢ Are most appropriate for testing the cause and effect relationship between the
variables.
➢ Most appropriate for testing hypothesis.
General Disadvantages of Experimental Design
 ➢ For some ethical reasons manipulation can not be achieved especially on the human
studies.
➢ Pre test can affect the response of post test.

QUASI-EXPERIMENTAL DESIGN
A quasi-experiment is a research design in which the researcher implements an experimental
treatment, but some characteristics of a true experiment, such as control or randomization, are
lacking. It is used to test cause-and-effect relationships, but the primary issue with quasi-
experiments is the reduced confidence in making causal assertions, often due to subject
contamination.
For example, examining differences in the amount of pain experienced by patients during a
painful procedure and the effect of nursing interventions on the experimental group may be a
quasi-experiment. In this case, it might be impossible to pretest the amount of pain.
How and When to use Quasi-experimental research
This design is used when researchers have control over the "when" and "whom" of
measurement, but lack control over the "when" and "to whom" the exposure occurs, meaning
complete control is not possible. It is commonly used in educational research.
1-Non-equivalent (Pretest and Posttest) Control-Group Design
In this design, a popular approach to quasi-experiments, the experimental group A and the
control group B are selected without random assignment. Both groups take a pretest and post-
test. Only the experimental group receives the treatment.
Group A X ________ X_________O
Group B O_____________________O
2-Single-Group interrupted Time-Series Design
In this design, the researcher records measures for a single group both before and after a
treatment.
Group A O – O – O – O - X- O- O – O – O
3-Control-Group interrupted Time-Series Design
A modification of the single-Group interrupted time-series design in which two groups of
participants, not randomly assigned, are observed over time. A treatment is administered to
only one of the groups (e.e.,Group A).
Group A O – O – O – O - X- O- O – O – O
Group B O – O – O – O - X- O- O – O – O

PRE-EXPERIMENTAL DESIGNS
Pre-experimental designs are the simplest form of experimental designs where the researcher
does not have full control over the experimental setup, particularly in terms of random
assignment. These designs are typically used in situations where it's not feasible or ethical to
manipulate the independent variable in a true experimental manner. However, they provide
limited ability to draw causal conclusions due to the lack of control over extraneous
variables.
1-One-Shot Case Study Design
Group A: X → O
o X = Treatment or intervention
o O = Posttest or measurement

In this design, a single group is exposed to a treatment (X) and then measured for the
outcome (O) without any pretest or comparison group. It’s often used for exploratory
purposes but lacks control over extraneous variables, making it difficult to establish causal
relationships.
2. One-Group Pretest-Posttest Design
Group A: O1 → X → O2
o O1 = Pretest measurement
o X = Treatment or intervention
o O2 = Post test measurement
This design includes a pretest measure (O1) followed by a treatment (X) and a posttest
measure (O2) for a single group. It allows researchers to observe changes over time, but the
lack of a control group makes it difficult to determine whether the observed changes are
solely due to the treatment.
3. Static Group Comparison (Posttest-Only with Nonequivalent Groups)

Group A: X → O
Group B: → O
o X = Treatment or intervention
o O = Posttest measurement
After implementing a treatment for Group A, a researcher selects a comparison group (Group
B) that did not receive the treatment. Both groups are measured with a posttest (O). The main
limitation is that the groups are not randomly assigned, so differences between them may be
due to preexisting factors rather than the treatment.
4. Alternative Treatment Posttest-Only with Nonequivalent Groups Design

Group A: X1 → O
Group B: X2 → O
o X1 = Treatment 1 (for Group A)
o X2 = Treatment 2 (for Group B)
o O = Posttest measurement

Similar to the Static Group Comparison design, but in this case, the two groups receive
different treatments. Group A gets Treatment 1 (X1), while Group B gets Treatment 2 (X2),
and both are measured using the same posttest (O). Since the groups are non-equivalent, the
results may not be directly comparable, limiting the ability to make causal claims about the
treatments.
MIXED DESIGN
With the development and perceived legitimacy of both qualitative and quantitative research
in social and human sciences, mixed methods research, employing the data collection
associated with both forms of data, is expanding.
Mixed methods research is relatively new in the social and human sciences as a distinct
research approach, it is useful to convey, in a proposal, a basic definition and description of
the approach.
Mixed methods strategies are less well known than either the quantitative or qualitative
approaches. The concept of mixing different methods originated in 1959 when Campbell and
Fisk used multimethods to study validity of psychological traits. They encouraged others to
employ their multimethod matrix to examine multiple approaches to data collection. This
prompted others to mix methods, and soon approaches associated with field methods, such as
observations and interviews (qualitative data), were combined with traditional surveys
(quantitative data; Sieber, 1973). Recognizing that all methods have limitations, researchers
felt that biases inherent in any single method could neutralize or cancel the biases of other
methods. Triangulating data sources, a concept introduced by Jick (1979), began as a way to
seek convergence between qualitative and quantitative methods. By the 1990s, this evolved
into integrating these data types more directly. For example, results from one method can
inform participant selection or questions in the other method (Tashakkori & Teddlie, 1998).
Qualitative and quantitative data can also be combined into a single dataset or used together
to strengthen findings (Creswell & Plano Clark, 2007). Additionally, mixed methods can
serve a transformative purpose, advocating for marginalized groups (Mertens, 2003).

In particular, three general strategies and several variations within them are illustrated in this
book:
• Sequential mixed methods procedures are those in which the researcher seeks to elaborate
on or expand on the findings of one method with another method. This may involve
beginning with a qualitative interview for exploratory purposes and following up with a
quantitative, survey method with a large sample so that the researcher can generalize results
to a population. Alternatively, the study may begin with a quantitative method in which a
theory or concept is tested, followed by a qualitative method involving detailed exploration
with a few cases or individuals.
• Concurrent mixed methods procedures are those in which the researcher converges or
merges quantitative and qualitative data in order to provide a comprehensive analysis of the
research problem. In this design, the investigator collects both forms of data at the same time
and then integrates the information in the interpretation of the overall results. Also, in this
design, the researcher may embed one smaller form of data within another larger data
collection in order to analyze different types of questions (the qualitative addresses the
process while the quantitative, the outcomes).
• Transformative mixed methods procedures are those in which the researcher uses a
theoretical lens (see Chapter 3) as an overarching perspective within a design that contains
both quantitative and qualitative data. This lens provides a framework for topics of interest,
methods for collecting data, and outcomes or changes anticipated by the study. Within this
lens could be a data collection method that involves a sequential or a concurrent approach.
Advantages of Mixed Methods Approach:
1. Comprehensive Data: Combines numerical data with detailed, contextual insights,
offering a fuller understanding of the research problem.
2. Cross-Validation: Allows for triangulation, where findings from one method can
validate or enhance those from another.
3. Flexibility: Adapts to complex research questions by using varied approaches,
providing depth and breadth in analysis.
4. Rich Data: Enhances results by capturing both statistical trends and personal
experiences, making it especially useful for studies involving diverse populations.
5. Transformative Potential: Can address social issues by combining rigorous evidence
with the voices of marginalized groups, supporting advocacy goals.
Disadvantages of Mixed Methods Approach:
1. Complexity: Designing and implementing two methods requires more resources,
time, and planning than single-method approaches.
2. Resource Intensive: Demands substantial time, funding, and expertise in both
qualitative and quantitative methods.
3. Data Integration Challenges: Combining and interpreting data from different
methods can be difficult and may introduce inconsistencies or bias.
4. Potential for Overwhelming Data: Produces a large volume of data, which can
complicate analysis and require careful organization.
5. Skill Requirements: Researchers must be skilled in both qualitative and quantitative
methods, which may require specialized training.