Understanding Research Design Elements
Understanding Research Design Elements
How has the Delhi housing market changed over the past 20 years?
Do customers of company A prefer product C or product D?
What are the main genetic, behavioural and morphological differences between Indian
wildcows and hybrid-cows?
How prevalent is disease 1 in population Z?
scientific approach and two sets of variables. The first set serves as a constant against which the
variations in the second set are measured. Experimentation is used in quantitative research
methodologies, for example. If you lack sufficient evidence to back your conclusions, you must
first establish the facts. Experimental research collects data to assist you in making better
that a variable change is due only to modification of the constant variable. The study should
identify a noticeable cause and effect. The traditional definition of experimental design is “the
strategies employed to collect data in experimental investigations.” There are three types of
experimental designs:
Literature research
In-depth interview
Focus groups
Case studies
Historical research is a type of qualitative research that looks at past events to draw conclusions
as well as make predictions about the future. This lesson explores historical research design and
unpacks the strengths and limitations of this type of research. Updated: 09/28/2021
Qualitative Research
Stan's parents survived the Holocaust and immigrated to the United States, where he was born
and raised. He grew up hearing stories about the concentration camps and the horrible things
done to people who were not accepted by the Nazi party. More than once, when telling stories
about the camps, Stan's mom would tear up and ask, 'Why? Why did they do that?'
It's a question that has haunted Stan for most of his life. Why did the Nazis take millions of
people out of their homes, torture them, and then kill them? Stan wonders if the answer to that
question could help prevent genocide in the future. He's passionate about finding the answer.
Stan is a psychologist, and he has always done research that involves numbers. He looks at
averages and percentages and tries to figure out how people act in a lab. But he's starting to
wonder if that's the best way to attack his mother's question. Instead, he thinks maybe he should
focus on qualitative research, which involves examining non-numerical data. There are many
ways to gather qualitative data. Let's look at one type of qualitative research closer, that of
historical design, and its strengths and limitations.
Historical Design
So, Stan decides that he wants to figure out why the Nazis acted the way they did. He wants to
do historical research, which involves interpreting past events to predict future ones. In Stan's
case, he's interested in examining the reasons behind the Holocaust to try to prevent it from
happening again.
Historical research design involves synthesizing data from many different sources. Stan could
interview former Nazis or read diaries from Nazi soldiers to try to figure out what motivated
them. He could look at public records and archives, examine Nazi propaganda, or look at
testimony in the trials of Nazi officers. There are several steps that someone like Stan has to go
through to do historical research:
1. Formulate an idea: This is the first step of any research, to find the idea and figure out the
research question. For Stan, this came from his mother, but it could come from anywhere. Many
researchers find that ideas and questions arise when they read other people's research.
2. Formulate a plan: This step involves figuring out where to find sources and how to approach
them. Stan could make a list of all the places he could find information (libraries, court archives,
private collections) and then figure out where to start.
3. Gather data: This is when Stan will actually go to the library or courthouse or prison to read or
interview or otherwise gather data. In this step, he's not making any decisions or trying to answer
his question directly; he's just trying to get everything he can that relates to the question.
4. Analyze data: This step is when Stan goes through the data he just collected and tries more
directly to answer his question. He'll look for patterns in the data. Perhaps he reads in the diary
of the daughter of a Nazi that her father didn't believe in the Nazi party beliefs but was scared to
stand up for his values. Then he hears the same thing from a Nazi soldier he interviews. A
pattern is starting to emerge.
5. Analyze the sources of data: Another thing that Stan has to do when he is analyzing data is to
also analyze the veracity of his data. The daughter's diary is a secondary source, so it might not
be as true as a primary source, like the diary of her father. Likewise, people have biases and
motivations that might cloud their account of things; perhaps the Nazi soldier Stan interviews is
up for parole, and he thinks that if he says he was scared and not a true Nazi believer, he might
get out of jail.
Once Stan has gone through all of these steps, he should have a good view of what he wants to
know about his question. If he doesn't, then he goes back to step two (formulating a plan) and
starts again. He will keep doing steps two through five until he finds something that he can use.
Cohort study
A cohort study is a sort of longitudinal research that takes a cross-section of a cohort (a group of
people who have a common trait) at predetermined time intervals. It’s a form of panel research in
which all of the people in the group have something in common.
Cross-sectional study
In social science, medical research, and biology, a cross-sectional study is prevalent. This study
approach examines data from a population or a representative sample of the population at a
specific point in time.
Longitudinal study
A longitudinal study is a type of study in which the same variables are observed repeatedly over
a short or long period of time. It’s usually observational research, although it can also take the
form of a long-term randomized experiment.
Cross-sequential study
Cross-sequential research design combines longitudinal and cross-sectional research methods,
with the goal of compensating for some of the flaws inherent in both.
Conducting historical research involves steps such as formulating a research idea, planning how to access sources, gathering data, analyzing data, and assessing data veracity through source analysis . Source analysis is crucial as it evaluates the validity and reliability of data sources, distinguishing between primary and secondary accounts. This step ensures that researchers accurately interpret historical events and ground their findings on credible evidence, minimizing bias and distortion in understanding past occurrences, which is essential for deriving valid conclusions and predictions .
A researcher might select an experimental research design to establish cause-and-effect relationships between variables in a controlled environment, enabling manipulation of the independent variable to observe its impact on the dependent variable . This design is ideal when seeking empirical evidence supporting hypotheses. The three common types of experimental designs include pre-experimental, true experimental, and quasi-experimental designs. Pre-experimental lacks random assignment and control but tests initial approaches; true experimental, with random assignment, provides strong causal inferences; and quasi-experimental lacks randomization yet attempts to approximate experimental conditions for practical constraints .
A well-constructed research design should possess neutrality, reliability, validity, and generalization. Neutrality ensures unbiased results free from researcher influence. Reliability guarantees consistent results across different iterations of the study. Validity confirms that the research measures what it intends to measure, ensuring accurate findings. Generalization allows for extrapolating results to broader contexts beyond the study sample. These characteristics are important as they uphold the scientific integrity and applicability of research findings .
Quantitative research design involves the use of mathematical and statistical analysis to examine variables, often using numbers, graphs, and tables as expressions. This approach typically requires the participation of a large number of people and often employs fixed design data collection methods such as multiple-choice questions . On the other hand, qualitative research design is more flexible, emphasizing open-ended inquiries and focusing on understanding context and complexity through subjective analysis. It generally involves a smaller participant group and employs techniques such as interviews or case studies to gather non-numerical data .
A researcher might choose a descriptive research design when the goal is to observe and describe phenomena as they naturally occur without manipulating variables, such as examining how the housing market in a city has changed over time . It is suitable for understanding trends and patterns without establishing cause-effect relationships. However, its limitations include the inability to infer causal links due to the lack of experimental control and potential observer biases impacting data collection and interpretation .
When choosing between a fixed and a flexible research design, researchers should consider the research question's nature and the type of data sought. Fixed designs are suitable for quantitative studies where objective, replicable measurements are needed, often using pre-determined methods like multiple-choice questions, and are favorable when testing specific hypotheses with controlled variables . Flexible designs are better for qualitative studies, allowing for data collection methods such as open-ended inquiries to evolve in response to gathered insights and requiring fewer participants. They accommodate exploratory phases in research where understanding context and subjectivity is crucial .
Diagnostic research design identifies and examines the underlying causes of specific issues or phenomena, assisting in clarifying the nature and origin of those problems. By progressing through stages like the inception of the issue, diagnosis, and solution proposal, this design helps in systematically investigating factors contributing to clients' challenges . It facilitates informed decision-making by providing insights into problem mechanisms and suggesting actionable strategies to address them, thereby directly supporting practical solutions and organizational strategies .
Cohort studies involve a form of longitudinal research where data is collected from a group of individuals sharing a common characteristic over set time intervals, allowing researchers to observe changes and developments within the cohort over time . Cross-sectional studies, conversely, collect data from a population or a representative sample at a single point in time, offering a snapshot of prevalence or relationships among variables at that instance. These studies are prevalent in social science and medical research . The longitudinal nature of cohort studies provides insights into temporal dynamics, whereas the cross-sectional approach provides immediate, comprehensive data without tracking changes over time.
Explanatory research design uses methods such as literature research, in-depth interviews, focus groups, and case studies. Literature research helps identify existing knowledge gaps and informs new inquiries. In-depth interviews provide detailed insights and deep understanding of individual perspectives. Focus groups gather diverse views and facilitate interaction between participants, enriching data. Case studies offer comprehensive explorations of specific instances within their real-life contexts. These methods collectively allow researchers to uncover causal relationships and delve into the underlying reasons behind phenomena .
Correlational research design effectively investigates relationships between variables by measuring the degree and direction of associations without researcher intervention or manipulation . It is advantageous for exploring real-world natural settings, offering insights into how variables may relate in actual contexts. However, while it reveals correlations, it cannot establish causation due to the potential presence of confounding variables. Therefore, results from correlational studies should not be interpreted as indicating direct causal relationships .






