Section Overview
Understanding various research study designs and methods is crucial for postgraduate students aiming to produce high-quality, impactful research. This resource delves into the fundamentals of research study design, from formulating your research questions and hypotheses to selecting appropriate methodologies and data collection techniques.
Research Explained
Research is a systematic process of inquiry to attempt to answer a research question, or to produce new knowledge or information. In short, you find answers to questions!
| Goal | Task |
|---|---|
| Exploration | Do certain study habits correlate with academic performance in Higher Education? |
| Description | If a particular study habit is associated with improved academic achievement, you want to further investigate this. Does it depend on other factors? |
| Explanation | The relationship between caring for an individual with heart failure and caregiver burden. |
| Prediction | A-level results are used to predict performance at university. |
| Action | Ban the sale of confectionary at tills in supermarkets as part of the plan to reduce the obesity epidemic. |
Researchers can take an inductive or deductive approach when formulating theories or generating a hypothesis.
Using an inductive approach, researchers start with a specific observation or empirical evidence and then move towards developing a broader understanding or development of theories. This involves making observations, identifying patterns, and then formulating hypotheses or theories based on the patterns identified.
A deductive approach is when researchers start with a general theory or hypothesis and they test the hypothesis.
Your research aim and objectives dictate the type of research study design you need to employ, not the other way about.
There are strengths and limitations to each research study design and to determine the correct study design to answer your research question you need to consider the resources needed for each type.
When designing research studies, there is a certain level of trade-off between rigor and the practical considerations.
Here are some things for you to consider whilst designing your research project:
- What type of data to collect (e.g. qualitative or quantitative data).
- Who to collect it from (i.e. the sampling strategy).
- When to collect it.
- Where to collect it.
- How to collect it (i.e. the data collection method).
- How to analyse it (i.e. the data analysis methods).
A good methodology explains not only what methodological choices were made but also why they were made, i.e. why the choices that were made are the best fit for your research.
Research Paradigms
This is a framework into which the theories and practices of your discipline integrate to create the research plan.
It encompasses a set of beliefs, assumptions, values and methods that shape how researchers conceptualise and study phenomena. Research paradigms influence everything from the formulation of research questions to the interpretation of findings.
Types of research paradigms:
- Positivism (quantitative)
- Interpretative or Constructivism (qualitative)
- Pragmatism (philosophical partner of mixed-methods)
- Critical Theory
Researchers may choose a paradigm based on its strengths, or the nature of their research questions and objectives.
To critically reflect on how they shape their research practices and interpretations, researchers must be aware of the paradigm's assumptions and implications.
For further reading on these paradigms
Hypothesis Development
A hypothesis is a specific statement of prediction to describe a relationship between variables that you will prove or disprove through the research process.
The goal of a hypothesis is to help explain the focus of the research clearly and concisely.
A hypothesis should state:
- the purpose of the research
- identify what variables are to be used
- the key concepts in the research and
- identify its relationship to both the problem statement and the literature review
A hypothesis can be simple or complex, directional or non-directional, associative or casual.
| Hypothesis | Explanation | Example |
|---|---|---|
Simple | Predicts a relationship between 1 independent variable and 1 dependent variable | Patients with better inhaler technique have better-controlled asthma |
Complex | Predicts relationships between 2 or more related independent variables and/or 2 or more related dependent variables | Patients with better inhaler technique who attend the asthma clinic regularly have better-controlled asthma and less reliance on salbutamol inhalers than those with poorer technique who do not attend the asthma clinic regularly |
Directional | You are stating the direction of change, for example when X increases Y increases | Taking antidepressant medication decreases depressive symptoms |
Non-directional | You are simply stating there is a relationship between X and Y | There is a relationship between antidepressant use and depressive symptoms |
NULL and ALTERNATIVE Hypotheses
In research, we present the NULL and ALTERNATIVE hypotheses.
Null hypothesis – H0 – the hypothesis to be tested
- There is no statistically significant difference/association/effect between the variables.
- There is no relationship between students' attendance in class and academic achievement.
The null hypothesis is presumed to be true until statistical evidence nullifies it.
Alternative hypothesis – H1 – the research hypothesis
- There is a statistically significant difference/association/effect between the variables.
- There is a positive relationship between students’ class attendance and academic attainment.
Aims and Objectives
The aim of a research study is the overarching goal or purpose. It outlines what the researcher intends to achieve with the research study. It is more general compared to the hypothesis.
Objectives are specific and measurable steps that the researcher will take to achieve the aim of the research study. When writing your objectives consider the following points:
- Objectives help define and limit the scope of the research.
- They create and maintain the overall structure.
- You should specify the variables and the methods of quantifying the variables.
- Use SMART targets (Specific, Measurable, Achievable, Realistic & Time-bound).
- Use action words such as:
- measure, compare, assess, evaluate, identify, examine, investigate etc.
Types of Research Study Designs
Research designs can be categorised as descriptive or analytical and examples of each are included here.
-
Case Report
The experience with a patient’s medical history, symptoms, signs, diagnosis, and treatment are discussed in detail.
They often focus on complex situations or rare cases, providing valuable insights.
-
Case Series
Several patients with similar experiences are grouped and discussed.
The aim is to describe the characteristics and outcomes of the patients rather than to perform statistical analysis or hypothesis testing.
-
Survey
This is a method used to gather information from a sample of participants.
Surveys can range from simple questionnaires to more complex questionnaires.
The questions can include a combination of closed (quantitative data) and open-ended (qualitative data) questions.
-
Randomised Controlled Trial (RCT)
This is a type of experimental study to evaluate the effectiveness of an ‘intervention’.
Participants are randomly allocated to either an intervention arm or control arm and in some cases, there can be more than two arms.
Having a control arm adds rigor to an RCT and the control arm is sometimes referred to as normal care or placebo, depending on the research area.
-
Observational Study
In this type of study, researchers observe and collect data on individuals or groups without intervening or manipulating any variables.
-
Cross-Sectional Study
This is a type of observational study that gathers data at a single point in time.
The aim is to gather information on a cohort's characteristics, behaviours or attitudes at that time.
-
Longitudinal Study
This is a type of observational study, also referred to as a cohort study.
It involves collecting data from participants on more than one occasion, over some time.
Data collection begins before the outcome of interest occurs and can continue after the outcome occurs.
You would use this method when studying disease prognosis or the consequences (positive or negative) of a particular exposure (e.g. Mediterranean diet or asbestos).
-
Prospective Study
This is a type of observational study where participants are recruited and followed over time to observe the occurrence of outcomes or events of interest.
Prospective studies focus on observing future events based on the participant's exposure or risk factors identified at the start of the study.
They focus on identifying predictors of future outcomes and assessing cause-and-effect relationships over time.
They are commonly used to study disease incidence, risk factors and preventive strategies.
Prospective studies typically analyse the association between exposures or risk factors measured at baseline and subsequent outcomes using methods such as logistical analysis.
-
Retrospective Study
This is a type of observational study where researchers analyse data from the past to investigate relationships between variables or to examine outcomes.
-
Case Control Study
This is a type of observational study that begins with the identification of participants with a particular outcome (cases) and participants without the outcome (controls).
The aim is to determine what makes one group of individuals different from the control or comparison group.
The study compares variables or characteristics between the cases and controls to identify factors from the latter that might have caused the outcome.
Often case-control studies recruit matched participants for the control groups, for example participants could be matched by diagnosis, age, gender and socioeconomic status.
Research Design Approach
Research designs can be described as quantitative, qualitative, or a mixture of both. It is important to remember some research questions require both approaches.
The section below summarises each of the approaches.
-
Quantitative
Quantitative research design is the process of collecting data that’s numerical and can be analysed using statistical software packages.
Results are presented in the form of numbers, statistics, graphs, and charts.
-
Qualitative
There are often instances when collecting numerical data and applying statistical analysis will not provide the answer to your research question.
Qualitative research design involves collecting and analysing non-numerical data such as text or audio. It can include open-ended questions used in surveys that collect numerical data, or data collected through conducting structured or semi-structured interviews.
This method is employed when you want to understand perceptions, beliefs or people's experience of something (intangible topics). This method allows the researcher to explore people's values and feelings to better understand complex processes.
-
Mixed-Methods
Mixed methods is an approach where researchers collect and analyse quantitative and qualitative data in the same research study.
When interpreting the results, researchers purposefully mix the quantitative and qualitative data, and an interpretation is made with both in mind. This is a common approach and helps you “triangulate” the data by supporting quantitative findings with qualitative findings and vice versa.
For example, you could ask people who care for a loved one with heart failure to complete questionnaires to assess stress and burden and then follow this up with interviews or focus groups to gain a more in-depth understanding of the impact of their caring role on their quality of life.
Qualitative Research Methodologies
These include many different approaches, some of which have been described here.
Grounded Theory
A systematic, yet flexible procedure, grounded theory follows an inductive approach, where theories about social phenomena are developed from qualitative data during the research process.
The iterative nature of the process involves constant comparison and refinement of categories and concepts as new data are collected and analysed.
Ethnography
This is employed to understand cultural phenomena within communities or specific social groups.
The researcher becomes embedded or immersed in a particular culture or community and closely observes, records and engages in the community under investigation, and then reports the outcomes.
It enables researchers to uncover the social and cultural processes shaping individuals\' beliefs, and behaviors to construct meaning within specific cultural contexts.
Action Research
This focuses on addressing a practical problem or real-world issue through a collaborative and iterative process of inquiry followed by action and reflection.
It aims to generate actionable knowledge with the support of key partners through all phases of the research process, to promote positive change within the context of the area of study.
It contributes to positive social change by bridging the gap between theory and practice and encouraging collaborative problem-solving processes.
Sampling Methods
This is the technique used to select participants to conduct research. Each sampling method has its strengths and limitations, and when deciding on a sampling method you need to consider the research objectives, the characteristics of your population, as well as time and resource constraints.
Here are some common sampling methods:
-
Simple Random Sampling
In this method a random sample is selected from a population so that each individual or object has an equal chance of being chosen. The researcher will require a list of all the individuals in the target population from which to choose a random sample. Free random sampling software is available.
-
Stratified Sampling
In this method the population is divided into distinct subgroups or 'strata' based on certain characteristics such as gender, race or age. Samples are then selected from each of the subgroups randomly, in proportion to the size of the stratum of the population. This ensures each subgroup is represented in the sample and allows a more targeted analysis of specific groups within a population.
-
Cluster Sampling
In this method, the population is divided into clusters based on a naturally occurring, well-defined divisions such as schools and households. A subset of clusters is then randomly selected and all individuals within the selected cluster are included in the sample.
-
Purposive Sampling
This can also be referred to as selective sampling, where the researcher intentionally selects participants that meet specific characteristics relevant to the research question. This method is often used in exploratory research to gain insights into a relatively unknown topic.
For purposive sampling, the process of selecting the sample population is not systematic. This means that each participant does not have an equal opportunity to be selected.
-
Convenience Sampling
In this method, the researcher recruits participants according to their availability and accessibility, such as colleagues, acquaintances, friends, and family.
For convenience sampling, the process of selecting the sample population is not systematic. This means that each participant does not have an equal opportunity to be selected.
-
Snowball Sampling
This is a nonprobability sampling technique where research participants are used to recruit additional participants who also meet the inclusion criteria, thus your sample "snowballs".
This is sometimes used when recruiting participant from minority groups or support groups.
Quantitative Data Collection Techniques
This involves collecting data via a questionnaire, from a sample of sufficient size and representation to enable generalisations to be made against a wider population.
They can be conducted in person, on the telephone, through the mail or online.
The researcher can facilitate the completion of the questionnaire or the participant can self-administer.
Survey data can also be collected via interviews and by observation of activities in practice.
JISC is an online survey tool that can be used to develop surveys. Students who require access should place a request with their research supervisor.
Microsoft 365 Forms is another online tool for creating a survey and is free to use by all students at Ulster University.
This test is administered and scored in a consistent or standard method.
This makes it possible to compare abilities or performance, to estimate a student's position within a cohort of similar students.
These can include a specific instrument such as a blood pressure monitor, Tanita scales, a stadiometer or a tool such as a standardised questionnaire.
The use of sensors to detect and measure information based on the physical environment around them.
Biomarkers are indicators of biological processes or conditions and include things like hormone levels, enzyme activity, urinary analysis or genetic markers.
Other types of biomarkers include standardised tests such as blood pressure readings.
This refers to data that was previously collected by someone else for a different purpose.
This data can be reused to answer other research questions.
Researchers at Ulster University have collated a disease sample biobank that includes data of genomic, proteomic and metabolomic determinants.
Further information is available on the biobank developed by the Personalised Medicine Centre.
Qualitative Data Collection Techniques
Qualitative Data Collection Techniques
This enables the researcher to collect rich, detailed data.
They can be conducted in person, online or via telephone.
Researchers use a semi-structured interview guide or schedule to help inform the discussion.
To allow for data analysis interviews are often audio recorded and transcribed verbatim.
Further reading:
- SAGE publications - Doing Interviews
- SAGE publications - A Practical Introduction to In-Depth Interviewing
These enable researchers to gain insights into specific topics.
Researchers often use focus groups to access 'vulnerable' participants or to discuss 'sensitive' topics.
Further reading:
- SAGE publications - Doing Focus Groups
These types of questions encourage respondents to provide detail using their own words in their response.
These can be used in a survey in combination with closed questions.
Open-ended questions capture in-depth respondent perspectives and experiences.
This is a type of interview where the researcher asks the participant to share their story as they see fit and without a pre determined interview guide.
Life story interviews seek to explore how changes across the globe or within societies affect peoples lives.
Further reading:
Observational research involves systematically observing and documenting behaviors or phenomena as they naturally occur without any manipulation by the researcher.
This approach is useful in the following contexts:
- Naturalistic observations
- Behavioral science
- Market research
Further reading:
Data Analysis
Data analysis is the process of systematically applying techniques to the data in order to answer your research question.
The data analysis method will vary depending on the type of data you collect (quantitative or qualitative).
Quantitative Analysis
Quantitative data can be analysed using statistical software packages such as IBM SPSS or R.
Qualitative Analysis
Qualitative data analysis includes thematic analysis, content analysis, narrative analysis and discourse analysis.
Quote
“Research is something that everyone can do, and everyone ought to do. It is simply collecting information and thinking systematically about it.”