Research Design refers to the blueprint to be followed by the researcher. As it provides a framework for the entire research project, the design will need to answer some important questions on how the goals of the research project will be accomplished. Why is the study being made? What type of data is required? Will there be a control group? What techniques and statistics will be used to collect and analyze the data? How will neutrality, reliability and validity be proven? And so on. A good research design constitutes a solid foundation for the research work to follow.
Below are descriptions, links to library resources from Reference Services and web resource links on the topics of research methodology and methods. This guide is by no means an exhaustive source.
"Research methodology is a way to systematically solve the research problem...research methodology has many dimensions and research methods do constitute a part of the research methodology. The scope of research methodology is wider than that of research methods. Thus, when we talk of research methodology we not only talk of the research methods but also consider the logic behind the methods we use in the context of our research study and explain why we are using a particular method or technique and why we are not using others so that research results are capable of being evaluated either by the researcher himself or by others" (Research Methodology: Methods & Techniques, 2004).
Descriptive Research focuses on the characteristics of the population or phenomenon that is being studied. This methodology focuses on the "who," "when" or "where" in relation to the outcome ("what"). BUT not the "why" or "how."
"The descriptive design is used to describe a single variable or population completely and thoroughly. Descriptive studies usually involve cross-sectional field research using both qualitative and quantitative data collection techniques" (Advanced Design in Nursing Research).
Aggregate, case report and case series studies are other types of descriptive studies.
Analytical Research tests hypotheses and answers the "how" and "why."
Experimental studies and observational studies are two types of analytical research. Experimental studies are usually randomized, meaning the subjects are grouped by chance, i.e.; randomized control trials (RCT). Cohort, case-control and cross-sectional studies can be grouped as observational.
Basic (or Fundamental) Research is mainly concerned with generalizations and formulations of theories. It is directed towards finding new information and ways of thinking that has a broad base of applications; thus, growing the already existing body of scientific knowledge.
Before the fields of computer science and atomic physics were advanced, there first needed to be an understanding of pure mathematics and what protons, neutrons, and electrons were composed of.
Note: Also called Pure Research.
Applied Research is applied in order to discover a solution for immediate problems facing a society, an industry, a business organization, or any other situation/population that is particular in its make-up. It uses the data directly for real world application.
A lot of advances/developments in science have been marked by complex mutual interactions between applied and basic research. The general and accurate theories laid down by basic research provides the foundations for which applied findings can be of use immediately.
"Conceptual Research is that related to some abstract idea(s) or theory. It is generally used by philosophers and thinkers to develop new concepts or to reinterpret existing ones" (Research Methodology: Methods & Techniques, 2004).
While conceptual research does not involve conducting any practical experiments, the quality of empirical research is dependent on the quality of the concepts on which the research questions and hypotheses are based, and which are taken as reference points for interpreting empirical data.
Empirical Research is data-based research yielding results capable of being verified by observation or experiment. "Empirical research is appropriate when proof is sought that certain variables affect other variables in some way. Evidence gathered through experiments or empirical studies is today considered to be the most powerful support possible for a given hypothesis" (Research methodology: methods & techniques, 2004).
With experimental methods, researchers have more control over confounding variables and can establish cause and effect relationships. There are three primary types of experimental design: Pre-experimental research design, True experimental research design and Quasi-experimental design.
"In pre-experimental designs, either a single group of participants or multiple groups are observed after some intervention or treatment presumed to cause change. Although they do follow some basic steps used in experiments, pre-experimental designs either fail to include a pretest, a control or comparison group, or both; in addition, no randomization procedures are used to control for extraneous variables. Thus, they are considered “pre-,” indicating they are preparatory or prerequisite to true experimental designs. Pre-experimental designs represent the simplest form of research designs. Together with quasi-experimental designs and true experimental (also called randomized experimental) designs, they make the three basic categories of designs with an intervention. Each contains subdesigns with specific strengths and weaknesses" (Encyclopedia of research design).
Qualitative Research is primarily exploratory research into 'how' and 'why' things occur, interpreting events and describing actions. It gathers data about lived experiences, emotions or behaviors, and the meanings (reasons/opinions/motivations) individuals attach to them. Qualitative methods assists in enabling researchers to gain a better understanding of complex concepts, social interactions or cultural phenomena; and can help develop ideas for potential quantitative (numerical) research.
Qualitative Research involves induction, subjectivity and context-oriented processes.
Quantitative Research gathers numerical data which can be categorized, classified, measured, or ranked through statistical analysis. It assists with uncovering patterns or relationships, and for making generalizations within populations. Along with establishing causalities or associations between variables, this type of research is useful for finding how many, how much, how often or to what extent.
Quantitative Research involves deduction, objectivity and generalities. Most importantly quantitative research generates reproducible knowledge because the data is collected and analyzed in a statistically valid way, producing results that can be standardized and shared among researchers.
"Qualitative and Quantitative Research are not the only alternatives, however. Mixed Methods Research makes it possible to do things that would be more difficult or even impossible to accomplish by operating solely within either the inductive-subjective-contextual or the deductive-objective-general packages that characterize the two more traditional approaches. This flexibility, however, comes at the cost of greater uncertainty about the purposes and procedures associated with Mixed Methods Research. The lack of a set of established traditions for Mixed Methods Research can make it more difficult to convince others of either the value of your research or the appropriateness of your research procedures" (Research Design and Research Methods (Ch. 3), 2013).
A pilot study is a preliminary investigation or "trial run." Pilot studies permit researchers to test and possibly reshape hypotheses, test the proposed data collection and analysis procedures, and other processes that can enable researchers to identify and head problems off at the pass before undertaking the major study.
An Exploratory Design is the least controlled design. Exploratory Research studies (also termed formulative) are primarily done to yield beginning theories. The emphasis is on the discovery of ideas and insights in order to formulate a more precise investigation.
"Historical Research is nonexperimental and is based on available written data, artifacts, or oral histories, all of which are usually qualitative data. Historiography can be cross-sectional or longitudinal...A critical step in historiography is establishing whether the data are primary or secondary, hearsay or eyewitness accounts, and true or falsified documents and whether the data sources are biased or objective observers of the times" (Advanced Design in Nursing Research).
Field Research is qualitative in nature and aims to observe, interact and understand subjects while they are in a natural environment. Field research can typically be conducted with direct observation, participant observation, ethnography, qualitative interviews and case studies.
Simulation models may be designed to match physical reality, or express alternative sets of laws to that researchers can explain phenomena that may be difficult, dangerous, or impossible to explore in reality.
Researchers strive for honesty, objectivity, integrity, carefulness, and openness in their research. It is expected that researchers will honor intellectual property and confidentially, and also be aware of legalities and social responsibilities. Researchers need to demonstrate competence, respect for colleagues, and responsible mentoring and publication practices. And if applicable, researchers must seek to provide proper animal care and maximize human subject protections.
"The Case Study method takes a situation as given and tries to find out what it particularly means to the participants. Commonly, case studies are associated with qualitative research, but often they combine different research techniques. They can illuminate quantitative findings and can incorporate quantitative data. The method usually involves the examination of one or, possibly, two or three particular cases in-depth and holistically. A case study can take months or even years to complete, which allows mature consideration of the findings, correction of misunderstandings, filling of gaps in the data, investigation of new ideas arising from the data and a longitudinal view" (Basic research methods : an entry to social science research).
Researchers in education and a number of social science fields often conduct this type of research method and it is also a method employed when there are not the resources for a large study.
"Case–control Studies represent a sampling strategy in which the population under study is selected based on the presence (case) or absence (control) of an event of interest. The underlying purpose of these studies is to identify causal factors of the events of interest by comparing characteristics of both groups. Case–control studies start by identifying the study population, from which cases are identified and their exposure status is determined retrospectively. Then, a control group of study subjects is sampled from the entire source population that gives rise to the cases. Case–control studies are frequently one of the first approaches used in the etiological study of a disease or health condition. This is in part due to the possibility of incorporating in the analysis many exposition factors simultaneously and relatively quickly and inexpensively. Therefore, case–control studies represent a cost-effective way of identifying risk and protective factors and generating hypotheses for subsequent, methodologically stronger studies" (Case-control studies in traffic psychology (Ch. 3, publisher summary only)).
"A Clinical Study involves research using human volunteers (also called participants) that is intended to add to medical knowledge. There are two main types of clinical studies: clinical trials (also called interventional studies) and observational studies" (ClinicalTrials.gov).
Cohort Studies look at a well-defined group of subjects who share a common characteristic or experience.
In epidemiology, cohort studies investigate those who are at risk of developing a specific disease or health outcome.
Example: The Air Force Health Study (AFHS) is a longitudinal, prospective epidemiologic study of more than 2,700 men followed for approximately 20 years. The cohort consisted of Air Force personnel who had participated in Operation Ranch Hand.
Cross-sectional Studies make observations of either an entire population or a representative subset of the population at a specific point in time. Like a photographic "snap-shot."
It is an observational type of study, and trends can be monitored over time however, when serial cross-sectional studies are conducted. But unlike longitudinal studies (performed over time), where the variables of the research can change during a study, a cross-sectional study observes a single instance with all variables remaining the same throughout the study.
In education a cross-sectional study is particularly helpful in understanding how students who scored within a particular grade range in the same preliminary courses perform with a new curriculum. This type of study can also help market researchers identify target demographics.
Experimental Studies are ones where researchers introduce an intervention and study the effects. Experimental studies are usually randomized, meaning the subjects are grouped by chance.
Randomized controlled (intervention) trial (RCT): Eligible subjects are randomly assigned to one of two or more groups. One group receives the intervention (such as a new drug) while the control group receives nothing or an inactive placebo. The researchers then study what happens to subjects in each group. Any difference in outcomes can then be linked to the intervention.
Natural/Quasi Experiments: Natural experiments are often used when controlled experimentation is not possible; such as when subjects can not be practically or ethically assigned treatments.
Experimental Settings: Experiments can be conducted in laboratory (artificial) and field (real-life) settings. Field research can help resolve gaps in data and collect supporting/ancillary material. "However, because field studies, by their very nature, do not control extraneous variables, it is exceedingly difficult to ascertain which factor or factors are more influential in any particular context. Bias can also be an issue if the researcher is testing a hypothesis" (Encyclopedia of Research Design).
Focus Groups will consist of several or more participants discussing a particular topic or a set of questions in a controlled environment. Focus groups provide insights into how people think and provide a deeper understanding of the phenomena being studied. Market researchers find focus groups useful in testing new products and concepts. Researchers can be facilitators or observers.
Interviews are one-on-one conversations where the interviewer is asking questions to gather specific information from the participants. Interviews are good complements or enhancements to data derived from surveys and questionnaires. It can be a good technique to employ when probing for information on sensitive topics. Field researchers might employ informal, conversational interviews to both help develop relationship with participants and incorporate the data into their field notes.
Observational Studies are types of study in which subjects are observed or certain outcomes are measured, without any attempt to affect the outcome (e.g., no treatment is given). Ecologists conduct observational studies in order to measure basic attributes of organisms, populations and communities in their natural setting without disturbance. Such data collected can help to understand behavioral or environmental changes.
Some observational research methods include case reports, chart reviews, and longitudinal studies of large cohorts followed over time.
Extremely important to recognize, however, is that observational studies cannot prove cause-and-effect. Researchers should be aware of, and address, the limitations of observational studies in their research.
Ever wanted to observe an elusive and crazed creature but were too scared to venture out? Read on! An Analysis of Consumer Behavior on Black Friday
Why use sampling? Sampling efficiently gets information on a large population when you can otherwise not get information on every subject in the population. Therefore, a representative sample of a population will suffice.
There are two main branches of sampling methods, probability sampling and non-probability sampling.
When choosing a sampling method, consider characteristics of the population to be studied, the level of precision required and the resources available.
Probability Sampling: every unit of the population has a known chance of participating in the study.
Non-probability Sampling: probability of selection is unknown or zero, so not every unit of the population has a chance to participate in the study.
Survey Research: This is an organized method of obtaining data from research participants via written, electronic, or oral questioning, which can be analyzed to reach conclusions about the question being investigated within a sample of the population. Surveys are frequently used by multiple disciplines to look for trends, behavior and the "bigger picture;" however, unless an experimental design is used, survey research does not allow for the drawing of causal inferences.
A questionnaire is a set of questions that can be qualitative, quantitative, or mixed, in nature. Questions can be open-ended (which requires the participant to formulate answers in their own words) or closed-ended (which requires the participant to choose a response from a list of options).
Delve into the world of research with the Researcher's Bootcamp. In these sessions you will learn how to create a search string, use reliable research resources, find law and legislation, and use reference managers to keep your research organized. The session order is as follows:
1. Conducting Internet Research Part 1
2. Conducting Internet Research Part 2
3. Finding Laws and Legislation
4. Reference Managers
