Please enable JavaScript.

Coggle requires JavaScript to display documents.

Educational Research Designs - Coggle Diagram

- - - - Independent Variable (IV)
      - Dependent Variable (DV)
      - Control Group
      - Experimental Group
      - Randomization
    - - Experimental design is a research method that involves manipulating one or more independent variables to observe their effect on a dependent variable. It is a systematic approach to investigate cause-and-effect relationships
    - - Teaching Experiments: Studying the impact of different teaching methods on student performance.
      - Engineering Tests: Examining the effect of various materials on product durability.
      - Clinical Trials: Testing the efficacy of a new drug compared to a placebo
    - - High internal validity: Because the researcher controls the conditions and randomly assigns participants to groups (experimental and control), it is easier to establish cause-and-effect relationships. You can be more confident that the treatment actually caused the change.
        
        Clear and objective results: It produces strong, measurable evidence that is often more convincing than other research methods.
        
        Replication: Other researchers can repeat the experiment to verify the findings, which strengthens scientific knowledge.
        
        Useful for testing theories: It helps test “what works” in education — for example, whether a new teaching method really improves student test scores.
        
        Reduces bias: Random assignment helps balance out individual differences between groups.
    - - Artificial setting: Many experiments are done in controlled lab-like conditions, which may not reflect real classroom situations. This lowers external validity (how well the results apply to the real world).
        
        Ethical issues: Sometimes it is unethical to manipulate variables — for example, you cannot deliberately give students a harmful teaching method or withhold helpful support just for research.
        
        Time-consuming and expensive: Good experimental studies require careful planning, materials, training, and often long periods of time.
        
        Hawthorne effect: Participants may behave differently because they know they are being observed or part of an experiment.
        
        Limited scope: It usually focuses on a small number of variables, while real teaching and learning involve many complex factors interacting at once.
        
        Sample size and generalizability: Many experiments use small or non-representative samples (e.g., only one school), making it hard to apply the results broadly.
  - - - Comparison Group: A group that does not experience the same intervention, but is similar to the treatment group in important ways.
      - Lack of Randomization: Unlike true experiments, quasi-experiments often don't employ random assignment.
      - Treatment Group: A group that experiences a change or intervention, which may be naturally occurring or externally imposed.
      - Limited Control: Researchers have limited control over variables, which can result in less internal validity compared to experimental design.
    - - Evaluating the impact of a new teaching method in a real classroom.
        Assessing the effects of a public health intervention in a community.
        Investigating the impact of policy changes on crime rates
    - - Benefits
        
        More practical in real educational settings: It is often impossible or unethical to randomly assign students to different classes or schools. Quasi-experiments allow researchers to study real-world situations.
        
        Higher ecological validity: Because the study happens in natural environments (actual classrooms), the results are usually more applicable to everyday teaching and learning.
        
        More ethical: Researchers don’t need to split students randomly or withhold potentially beneficial treatments from a control group.
        
        Feasible and cost-effective: Easier to conduct in schools compared to true experiments, especially with large groups.
        
        Still supports causal inference: With good design (such as pre-tests and post-tests, matching groups, or interrupted time series), researchers can make reasonably strong conclusions about cause and effect.
        
        Useful for evaluating real programs: Great for studying the impact of new teaching methods, curriculum changes, or school policies that are already being implemented.
    - - Lower internal validity: Without random assignment, there may be pre-existing differences between groups (selection bias). It’s harder to prove that the treatment caused the results rather than other factors.
      - Confounding variables: Other uncontrolled factors (e.g., teacher quality, student motivation, parental support) can influence the outcomes, making it difficult to isolate the real cause.
      - Weaker causal claims: Conclusions are not as strong or convincing as those from true experimental designs.
      - Threats to validity: More vulnerable to history effects, maturation, testing effects, and regression to the mean.
      - Complex data analysis: Often requires more advanced statistical techniques (e.g., ANCOVA, propensity score matching) to control for differences between groups.
      - Generalizability issues: Results may still be limited to the specific groups or contexts studied.
  - - - Questionnaire
      - Sampling
      - Data Collection Methods
      - Types of Survey – Cross Sectional / Longitudinal
    - - Political Polls: Gathering data on public opinion for elections
      - Employee Engagement Surveys: Measuring job satisfaction within an organization.
      - Customer Satisfaction Surveys: Assessing how
        satisfied customers are with a product or service.
    - - Can reach a large number of people: Surveys allow researchers to collect data from hundreds or even thousands of respondents quickly and efficiently.
      - Cost-effective and time-saving: Especially online surveys are relatively cheap and easy to distribute compared to experiments or observations.
      - Good for understanding opinions and attitudes: They are excellent for exploring what people think, feel, or believe (e.g., student satisfaction with online learning).
      - High external validity: If the sample is representative, the results can be generalized to a larger population.
      - Anonymous responses: People are often more honest when answering sensitive questions anonymously.
      - Easy to analyze: Responses (especially closed-ended questions) can be easily quantified and analyzed using statistics.
      - Flexible: Can be conducted in many ways — paper, online (Google Forms, Qualtrics), phone, or in person.
    - - Cannot prove cause and effect: Surveys show relationships (correlations) but cannot determine causation because there is no manipulation of variables.
      - Self-report bias: People may not answer truthfully — they might give socially desirable answers or misremember information.
      - Low response rates: Many people ignore surveys, which can lead to biased results (only motivated people respond).
      - Question quality matters a lot: Poorly worded, leading, or ambiguous questions can produce inaccurate data.
      - Limited depth: Surveys usually give surface-level information. They don’t provide rich, detailed explanations like interviews do.
      - Sampling bias: If the sample is not representative (e.g., only responses from one school or only tech-savvy students), the results may not reflect reality.
      - Social desirability and acquiescence bias: Respondents may try to look good or simply agree with everything.
  - - - Data Collection at a Single Point
      - No Manipulation of Variables
      - Comparative Analysis
    - - Market Research: Studying consumer preferences and behaviors.
      - Social Demography: Examining population characteristics, such as age or income distribution.
      - Health Surveys: Assessing the prevalence of diseases or risk factors in a population.
    - - Fast and time-efficient: Data can be collected relatively quickly compared to longitudinal studies that follow people over months or years.
      - Cost-effective: Generally cheaper and requires fewer resources than experimental or long-term studies.
      - Large sample size: It is easier to survey or study a big number of participants, which improves generalizability.
      - Good for describing current situations: Excellent for understanding the prevalence of something (e.g., how common exam anxiety is among students) or exploring relationships between variables.
      - Useful for generating hypotheses: Helps identify possible patterns or associations that can be tested later with stronger designs.
      - Practical in education: Easy to conduct in schools — for example, giving questionnaires to all Grade 10 students in one day to study the relationship between screen time and academic performance.
    - - Cannot prove cause and effect: Since all data is collected at one time, we cannot know which variable came first. We can only see correlations, not causation.
      - No sense of change over time: It doesn’t show how things develop or whether one factor influences another later (this is a major weakness compared to longitudinal designs).
      - Cohort or age effects: Differences found might be due to the specific group being studied rather than the variables of interest.
      - Reverse causation problem: It’s often unclear which variable influences the other (e.g., Does poor sleep cause low grades, or do low grades cause poor sleep?).
      - Selection bias: The sample may not represent the whole population if certain types of people are more likely to respond.
      - Social desirability bias: People may give answers they think are expected rather than what is truly accurate.
  - - - Repeated Data Collection
      - Temporal Sequence
      - Tracking Individuals or Groups
    - - Educational Research: Evaluating the long-term impact of educational interventions.
      - Health Studies: Tracking the health trajectories of specific populations.
      - Developmental Psychology: Studying cognitive,
        emotional, and social development in individuals from childhood to adulthood.
    - - Can show change over time: It allows researchers to observe how variables develop or change (e.g., how students’ motivation changes from Grade 7 to Grade 12).
      - Stronger evidence for cause and effect: Because data is collected in sequence, researchers can see which variable comes first, making causal relationships clearer than in cross-sectional studies.
      - Tracks individual development: It shows individual patterns and trajectories, not just average group results. This helps understand differences between people.
      - Good for studying long-term effects: Excellent for examining the long-term impact of teaching methods, early interventions, parenting styles, or educational policies.
      - Reduces cohort effects: Since it follows the same people, it avoids some problems that appear when comparing different age groups at one point in time.
      - Rich and detailed data: Provides deep insights into how and why changes happen.
    - - Very time-consuming: Studies can take many years to complete, which delays results.
      - Expensive: Requires significant funding, resources, and long-term commitment from researchers.
      - High dropout rate (attrition): Many participants may leave the study over time (move schools, lose interest, etc.), which can bias the results.
      - Practice or testing effects: Repeated testing may influence participants’ behavior or performance (they get better at tests just because they take them often).
      - External influences: Long-term studies are easily affected by outside events (e.g., pandemics, economic changes, new education policies) that can confuse the findings.
      - Difficult to manage: Keeping the same sample for many years is challenging, especially with students who graduate or change schools.
  - - - Measurement of Variables
      - Statistical Analysis: the strength and direction of relationships
      - Correlation Coefficient
    - - Social Sciences: Investigating the link between income and life satisfaction.
      - Epidemiology: Assessing the correlation between smoking and lung cancer.
      - Psychology: Studying the relationship between stress levels and academic performance.
    - - Easy and practical to conduct: Can be done using surveys, existing school records, or questionnaires without interfering with normal classroom activities.
      - Good for real-world settings: Works well in schools where it is difficult or unethical to do experiments.
      - Can study many variables at once: Researchers can explore relationships between several factors simultaneously.
      - Useful for prediction: Helps identify variables that can predict outcomes (e.g., attendance can predict final grades).
      - Time and cost efficient: Generally faster and cheaper than experimental or longitudinal studies.
      - Generates hypotheses: Strong correlations can suggest possible cause-and-effect relationships that can be tested later with experimental designs.
      - High external validity: Data is often collected from natural settings, so results can apply well to real life.
    - - Cannot prove cause and effect: This is the biggest weakness. Correlation does not mean causation. Just because two things are related doesn’t mean one causes the other.
      - Directionality problem: Unclear which variable influences the other (e.g., Does low self-esteem cause poor grades, or do poor grades cause low self-esteem?).
      - Third-variable problem: There may be other hidden factors affecting both variables (e.g., family income affecting both study time and academic performance).
      - Can be misleading: Strong correlations sometimes lead people to assume wrong conclusions.
      - No control over variables: Researchers cannot control outside influences, so results may be less accurate.
      - Risk of spurious correlations: Sometimes two variables appear related purely by chance.
- - - - Intensive Exploration: Researchers collect a vast amount of data and information about the case.
      - Holistic Approach: The entire context and multiple facets of the case are examined.
      - Qualitative Data: Typically, case studies involve the collection of qualitative data, such as interviews, observations, and documents.
    - - Business: Analyzing the success or failure of a particular company's strategies
      - Education: Investigating a unique teaching method's effectiveness in a specific classroom.
      - Clinical Psychology: In-depth analysis of a single patient's psychological condition and treatment.
      - Case studies are empirical investigation of phenomena within their environmental context, where the relationship between the phenomena and the environment is not clear.Case study methods involve systematically gathering enough information about a particular person, social setting event, or a group to permit the researcher to effectively understand how it operates or functions" (Berg, 2001, p. 225).In a case study the researcher collects detailed informationthrough a variety of data collection procedures over a sustained period of time.
        
        Case study research can be based on either a single case or multiple cases (Berg, 2001; Yin, 1993). When two or more settings are studied, they are considered to be multi-case studies (Bogdan & Biklen, 1998) or collective case studies (Stake, 1994).
        Single case design is ideal for studying extreme cases, to confirm or challenge a theory or for cases where a researcher did not have access previously. Multiple cases are attempted when the purpose is to strengthen the understanding of a broader context within which a particular phenomenon is occurring (Stake, 1994) Multiple case design serves to confirm evidence which enhance the reliability and validity of a research work.
  - - - Educational Research: Studying classroom dynamics and learning processes in a specific school.
      - Healthcare: Exploring the cultural factors influencing healthcare-seeking behavior in a particular community.
      - Cultural Anthropology: Investigating the customs, traditions, and rituals of an indigenous tribe.
    - - In ethnographic research the researcher studies an intact cultural group in a natural setting over a considerable period of time.
      - A cultural group can be any group of individuals who share a common social experience, location, or other social characteristic of interest.
        Ethnography involves exploring the nature of phenomena and working with unstructured data, analysing data through interpretation of the meanings attributed by research respondents.
      - The ethnographic method needs considerable time and fieldwork commitment by the researcher. It can be extremely time consuming as it involves the researcher spending a long time in the observation period and jotting down field notes.
    - - Jot down notes immediately and as soon possible during observationas
      - Keep count of the number of phrases used by subjects
      - Never neglect anything as insignificant
      - Record sequence of events chronologically and period of events
      - Avoid evaluative judgments or summarising retrieved facts from respondents.
  - - - Exploring Subjective Experience: Researchers focus on how individuals interpret and make sense of their experiences.
      - Bracketing: Researchers temporarily set aside their own preconceptions and biases to fully immerse themselves in the participants' worldviews.
      - In-Depth Interviews: Data is typically collected through in-depth, open-ended interviews that allow participants to describe their experiences in their own words.
    - - Psychology: Investigating the lived experiences of individuals with a particular mental health condition.
      - Nursing: Exploring the experiences of caregivers in the context of long-term patient care.
      - Education: Studying the experiences of students with disabilities in inclusive classrooms.
    - - Phenomenological studies are a qualitative research method that aims to explore and understand the lived experiences, perspectives, and meanings of individuals regarding a particular phenomenonResearchers seek to uncover the essence and structure of these experiences.
  - - - Data-Driven Approach: The theory is grounded in the data, emerging through a systematic process.
      - Constant Comparative Analysis: Data is continually compared and analyzed to generate categories and concepts.
      - Theoretical Sampling: Data collection is guided by emerging theories, with researchers selecting new participants or data sources strategically
    - - Sociology: Creating theories that explain the process of identity formation in marginalized communities.
      - Nursing: Developing theories on patient experiences and coping mechanisms during illness.
      - Business: Exploring the emergence of new management practices in the modern workplace.
  - - - Cyclical Process: Action research typically involves iterative cycles of planning, acting, observing, and reflecting to make improvements.
      - Collaboration: Practitioners and researchers work together to identify and address problems.
      - Practical Orientation: The goal is to create practical solutions and effect positive change in specific contexts.
    - - Education: Teachers conducting action research to improve teaching methods and student outcomes.
      - Healthcare: Healthcare professionals using action research to enhance patient care procedures.
      - Community Development: Community members collaborating with researchers to identify and solve local issues.
    - - Action planning: Determine the actions to be undertaken to address the research
      - Action taking: Conduct and monitor the planned actions
      - Diagnosing: Identifying the research question
      - Evaluation: Determine if the actions have addressed the research question
  - - - Media Studies: Analyzing news articles to identify trends in reporting on a specific topic.
      - Marketing: Studying advertisements to understand how products are marketed to different demographics.
      - Social Media: Analyzing social media posts to track public sentiment on a particular issue.
    - - It is a detailed systematic examination of the contents of a particular material for identifying patterns or themes. Content analysis involves the concurrent acts of identifying substantive points in the raw data, coding them and constructing categories for sorting the coded pieces of information (Gillham, 2000; Merriam, 1998;Patton, 1987).
    - - Systematic Coding: Researchers develop a coding scheme to categorize and analyze content based on predefined criteria.
      - Quantitative and Qualitative Approaches: Content analysis can be both quantitative (counting and categorizing) and qualitative (interpreting meanings and themes).
      - Objective and Replicable: The method is designed to be objective and replicable, with clear procedures for data analysis.
    - - Define the characteristics or qualities to be examined in precise terms
      - Break into small and manageable segments
      - Identify the specific body of material which needs to be explored
      - Scrutinise and sort the materials