Fiche de révision : Fundamentals of Psychological Research Methods

📋 Course Outline

1. Research Variables
2. Research Design Types
3. Variables in Psychology
4. Sampling Methods
5. Experimental Groups
6. Data Collection Techniques
7. Data Types and Scales
8. Control and Confounding Variables
9. Ethical Considerations
10. Data Analysis and Interpretation

📖 1. Research Variables

🔑 Key Concepts & Definitions

• Independent Variable (IV): A variable deliberately manipulated or varied by the researcher to observe its effect on the dependent variable (DV). It is the cause in cause-and-effect relationships (source: research methodology).

• Dependent Variable (DV): The property that is measured in an experiment, which depends on the IV. Its value is expected to change in response to the IV (source: research methodology).

• Extraneous Variable: Any variable other than the IV that could influence the DV, potentially confounding results. When controlled, it becomes a controlled variable (source: research methodology).

• Confounding Variable: An extraneous variable that affects the DV and is not controlled, leading to invalid conclusions about the relationship between IV and DV (source: research methodology).

• Operationalisation: The process of defining variables in measurable terms, specifying how IV and DV are manipulated or measured in a study (source: research methodology).

• Control Variable: An extraneous variable that is intentionally kept constant to prevent it from influencing the outcome of the experiment (source: research methodology).

📝 Essential Points

• The IV is manipulated to observe its effect on the DV, establishing a cause-and-effect relationship (source: research methodology).

• Extraneous variables can threaten the internal validity of an experiment; controlling them ensures that changes in the DV are due to the IV (source: research methodology).

• Confounding variables can produce misleading results; thus, random allocation and control techniques are used to minimize their impact (source: research methodology).

• Proper operationalisation of variables is crucial for replicability and validity, ensuring clarity on how variables are measured or manipulated (source: research methodology).

• The distinction between extraneous and confounding variables is vital: extraneous variables are controlled, whereas confounding variables threaten the validity if not controlled (source: research methodology).

💡 Key Takeaway

Understanding and accurately defining the roles of independent, dependent, extraneous, and confounding variables, along with proper operationalisation, are essential for designing valid and reliable psychological experiments.

📖 2. Research Design Types

🔑 Key Concepts & Definitions

• Experimental Design: A research method involving the deliberate manipulation of an independent variable (IV) to observe its effect on a dependent variable (DV), allowing for causal inference (author unknown).
• Control Group: A group in an experiment that does not receive the experimental treatment, used as a baseline for comparison to assess the effect of the IV (author unknown).
• Random Allocation: The process of assigning participants to experimental or control groups randomly to reduce bias and participant variables, ensuring groups are comparable (author unknown).
• Independent Groups Design: An experimental setup where each participant is only exposed to one condition, reducing extraneous variables but requiring more participants (author unknown).
• Repeated Measures Design: An experimental setup where the same participants experience all conditions, reducing participant variability but risking order effects (author unknown).
• Matched Participants Design: An experimental approach where participants are paired based on key variables (e.g., IQ, age) to control confounding variables, combining strengths of independent and repeated measures designs (author unknown).
• Correlational Design: A non-experimental method that examines the relationship between two naturally occurring variables without manipulation, used when experiments are unethical or impractical (author unknown).

📝 Essential Points

• Research Design Choice: Determines how variables are manipulated and measured, impacting the ability to infer causality. Experimental designs manipulate IVs, while correlational studies observe naturally occurring relationships (author unknown).
• Control & Confounding Variables: Proper control of extraneous variables (see section 7) is crucial to avoid confounding effects that threaten internal validity. Random allocation helps mitigate participant variables (author unknown).
• Types of Experimental Designs:
  • Independent Groups: Suitable when participant variables might influence results; requires larger sample sizes.
  • Repeated Measures: Efficient with fewer participants but susceptible to order effects.
  • Matched Participants: Controls confounding variables like IQ or age but is resource-intensive (author unknown).
• Ethical & Practical Considerations: Some variables (e.g., trauma) cannot be ethically manipulated; thus, correlational designs are used to study such phenomena (author unknown).
• Data Collection Methods: Observation, interviews, questionnaires, and technology-assisted tools are used depending on the research question and design (author unknown).

💡 Key Takeaway

Research design types in psychology—experimental, correlational, and their variations—are chosen based on ethical, practical, and scientific considerations to establish causal relationships or observe natural phenomena, with careful control of variables ensuring valid results.

📖 3. Variables in Psychology

🔑 Key Concepts & Definitions

• Independent Variable (IV):
  The variable that the researcher deliberately manipulates or varies to observe its effect on the dependent variable (see section 6). (Author not specified in source)

• Dependent Variable (DV):
  The property that is measured in an experiment; its value depends on the manipulation of the IV. It reflects the outcome of the experiment. (Author not specified in source)

• Extraneous Variable:
  Any variable other than the IV that could influence the DV, potentially confounding results. When controlled, it becomes a controlled variable. (Author not specified in source)

• Confounding Variable:
  An extraneous variable that unintentionally affects the DV, leading to invalid conclusions about the relationship between IV and DV. It is a source of bias in experiments. (Author not specified in source)

• Participant Variable:
  Individual differences among participants (e.g., age, gender, IQ) that can influence results and introduce bias if not properly controlled or randomized. (Author not specified in source)

• Operationalisation:
  The process of defining variables in measurable terms to ensure clarity and replicability in research. For example, defining "stress" by cortisol levels or self-report scales. (Author not specified in source)

📝 Essential Points

• Variables are central to research design; the IV is manipulated, and the DV is measured to establish causal relationships (see section 6).
• Extraneous variables can threaten the internal validity of an experiment; controlling them is essential for accurate results.
• Confounding variables can produce false or misleading results, which is why random allocation (see section 4) and control procedures are vital.
• Participant variables, such as gender or IQ, can be sources of bias; random assignment or matching helps mitigate their effects.
• Operationalisation ensures variables are clearly defined and measurable, facilitating replication and validity of findings.
• The use of control groups (see section 4) helps isolate the effect of the IV on the DV by providing a baseline for comparison.

💡 Key Takeaway

Variables are fundamental to scientific research in psychology; understanding how to manipulate, measure, and control them ensures valid, reliable, and interpretable results.

📖 4. Sampling Methods

🔑 Key Concepts & Definitions

Sampling Frame:
The actual list or database from which a sample is drawn, representing the entire population (see section 6 for data types). It determines who is eligible for selection.

Representative Sample:
A subset of the population that accurately reflects the characteristics of the entire population, minimizing bias and allowing generalization of results (see section 8 for ethical considerations).

Convenience Sampling:
A non-random sampling method where participants are selected based on ease of access and availability; often leads to biased samples (see section 6).

Random Sampling:
A sampling technique where every member of the population has an equal chance of being selected, promoting unbiased representation (see section 6).

Stratified Sampling:
A probability sampling method where the population is divided into subgroups or strata (e.g., age, gender), and random samples are taken from each in proportion to their occurrence in the population, reducing bias (see section 6).

Systematic Sampling:
A method where every nth individual from a list is selected after a random starting point, useful for large populations but can introduce bias if the list has a pattern.

Sampling Bias:
Systematic errors in sampling that lead to a non-representative sample, threatening the validity of the study (see section 8 for ethical considerations).

📝 Essential Points

• Sampling methods influence the generalizability and validity of research findings.
• Convenience sampling is easy but often biased; random sampling is ideal for representativeness but may be more resource-intensive.
• Stratified sampling is considered the most effective probability sampling method because it ensures proportional representation of key subgroups, reducing bias (see section 6).
• Participant selection must balance practicality with the need to avoid sampling bias and ensure the sample reflects the target population.
• Biases such as volunteer bias and selection bias can distort results, emphasizing the importance of proper sampling techniques.
• Ethical considerations include ensuring informed consent and privacy during participant recruitment, especially with stratified or random sampling.

💡 Key Takeaway

Choosing an appropriate sampling method is crucial for obtaining a representative sample, which underpins the validity and generalizability of psychological research findings. Stratified sampling is often the most effective way to minimize bias and ensure proportional representation of key groups.

📖 5. Experimental Groups

🔑 Key Concepts & Definitions

• Experimental Group (E-group):
  The group of participants exposed to the independent variable's experimental condition. They experience the treatment or manipulation being tested (see source content for example of sleep deprivation).
  (No specific author; standard definition in research methodology)

• Control Group (C-group):
  The group that does not receive the experimental treatment or receives a placebo. It serves as a baseline to compare the effects of the independent variable. Ensures that any observed effects are due to the manipulation of the IV.
  (No specific author; fundamental in experimental design)

• Random Allocation:
  The process of randomly assigning participants to either the experimental or control group to minimize bias and participant variables. It ensures each participant has an equal chance of being in any group, reducing confounding variables (see source for example of gender confound).
  (No specific author; standard practice in experimental research)

• Confounding Variable (see section 7):
  An extraneous variable that influences the dependent variable and can distort the results if not controlled. Random allocation helps to reduce confounding effects between experimental and control groups.
  (No specific author; key concept in experimental validity)

• Matched Participants Design:
  A type of experimental design where participants are paired based on key variables (e.g., IQ, age) to control for confounding factors. Each pair is split into different groups, ensuring groups are comparable.
  (No specific author; used to control confounding variables)

• Repeated Measures Design:
  An experimental design where the same participants are exposed to all conditions of the independent variable, allowing within-subject comparisons. It reduces the number of participants needed but can introduce order effects.
  (No specific author; common experimental design)

📝 Essential Points

• The experimental group receives the treatment, while the control group does not, providing a basis for comparison (source).
• Random allocation is crucial to prevent bias caused by participant variables, such as gender or age, which could confound results (source).
• Matching participants on key variables (e.g., IQ, age) helps eliminate confounding variables, especially in studies involving twins or family members (source).
• Repeated measures designs involve the same participants in all conditions, reducing variability but risking order effects and demand characteristics (source).
• Properly designing experimental groups and control groups ensures the validity of causal inferences in psychological research.

💡 Key Takeaway

Creating well-matched experimental and control groups, combined with random allocation, is essential for establishing causal relationships and reducing bias in psychological experiments.

📖 6. Data Collection Techniques

🔑 Key Concepts & Definitions

Data Collection Method (see section 8): The systematic process of gathering information to answer research questions, which can include observation, interviews, questionnaires, and technological tools.

Naturalistic Observation (Author (date)): Observing subjects in their normal environment without interference, providing high ecological validity but limited control over variables.

Controlled Observation (Author (date)): Observing behaviors in a controlled setting like a laboratory, allowing for greater control but possibly affecting natural behavior.

Structured Interview (Author (date)): A data collection method with pre-determined questions and fixed responses, facilitating comparison but limiting depth and flexibility.

Questionnaire (Author (date)): A set of written questions or rating scales used to gather data from participants, often standardized for reliability and ease of analysis.

Technology-Assisted Data Collection (Author (date)): Using computerized systems, video/audio recordings, or automated tools to gather large amounts of data efficiently and accurately.

📝 Essential Points

• Types of Observation: Naturalistic observation offers realistic insights but lacks experimental control; controlled observation provides precision but may influence participant behavior.
• Interviews and Questionnaires: Structured interviews are standardized but less flexible; questionnaires are useful for large samples and quantifiable data.
• Technological Tools: Computerized systems and recordings enhance data collection efficiency, reduce human error, but require setup and validation.
• Data Types: Data can be qualitative (descriptions, emotions) or quantitative (measurements, counts), subjective (opinions, feelings) or objective (observations, measurements).
• Scales of Measurement: Nominal (categories), ordinal (rankings), interval (equal intervals), and ratio (true zero), each suitable for different data analysis techniques.
• Ethical Considerations: Observation and recording must respect privacy and consent, especially in naturalistic settings.

💡 Key Takeaway

Effective data collection in psychology involves selecting appropriate methods—such as observation, interviews, or technological tools—that align with the research aims, while ensuring ethical standards and data accuracy.

📖 7. Data Types and Scales

🔑 Key Concepts & Definitions

• Qualitative Data: Descriptive data that characterizes qualities or categories, such as emotional states or colors. It is non-numerical and often subjective (source content).
• Quantitative Data: Numerical data that can be measured or counted, such as reaction times or number of items. It is objective and allows for statistical analysis (source content).
• Subjective Data: Data based on personal opinions, feelings, or interpretations, which can vary between individuals (source content).
• Objective Data: Data based on observable measurements or factual observations, independent of personal feelings (source content).
• Continuous Data: Measurable data that can take any value within a range, such as height or temperature, allowing for high precision (source content).
• Discrete Data: Countable data that can only take specific values, such as the number of students or items, with no intermediate values possible (source content).
• Nominal Data: Categorical data with no inherent order, such as eye color or nationality (source content).
• Ordinal Data: Data that can be ordered or ranked, but the intervals between ranks are not necessarily equal, e.g., Likert scales or race positions (source content).
• Interval Data: Numerical data with equal intervals between values, but no true zero point, e.g., temperature in Celsius or Fahrenheit (source content).
• Ratio Data: Numerical data with a true zero point, allowing for meaningful ratio comparisons, e.g., weight or reaction time (source content).

📝 Essential Points

• Data can be classified into qualitative or quantitative, with qualitative describing qualities and quantitative involving measurements (source content).
• Subjective data depends on personal opinions, whereas objective data relies on observable measurements (source content).
• Continuous data allows for precise measurement and can take any value within a range, while discrete data involves specific, countable values (source content).
• Psychologists use four measurement scales: nominal, ordinal, interval, and ratio, each suited for different types of data (source content).
• The most precise scale is the ratio scale, which includes a true zero and enables the calculation of ratios, making it suitable for powerful statistical tests (source content).
• Proper understanding of data types and scales is crucial for selecting appropriate statistical analyses and interpreting results accurately (source content).

💡 Key Takeaway

Understanding the different data types and measurement scales enables psychologists to accurately collect, analyze, and interpret research data, ensuring valid and meaningful conclusions.

📖 8. Control and Confounding Variables

🔑 Key Concepts & Definitions

• Extraneous Variable: A variable other than the IV that could influence the DV, potentially confounding results. When controlled, it becomes a controlled variable (SOURCE).
• Confounding Variable: An extraneous variable that affects the DV and is not controlled, leading to invalid conclusions about the IV's effect (SOURCE).
• Controlled Variable: An extraneous variable that the researcher actively keeps constant to prevent it from influencing the DV (SOURCE).
• Placebo Effect: When participants' expectations influence their behavior, potentially skewing results; mitigated by single-blind procedures (SOURCE).
• Experimenter Effect: Unintentional influence of the researcher on participants' behavior; minimized by double-blind procedures (SOURCE).
• Random Allocation: Assigning participants to groups randomly to reduce bias and participant variables, ensuring groups are comparable (SOURCE).

📝 Essential Points

• Control of extraneous variables is essential to establish a clear cause-and-effect relationship between IV and DV (SOURCE).
• Confounding variables threaten internal validity by providing alternative explanations for results; identifying and controlling them is crucial (SOURCE).
• Procedures like single-blind and double-blind are used to reduce extraneous effects such as placebo and experimenter bias (SOURCE).
• Random allocation helps eliminate participant variables, reducing confounding effects and increasing the reliability of results (SOURCE).
• Control variables are kept constant across groups to ensure that only the IV influences the DV, maintaining experimental integrity (SOURCE).
• In observational studies, controlling variables is more challenging, but researchers attempt to account for confounders through matching or statistical controls (SOURCE).

💡 Key Takeaway

Controlling extraneous and confounding variables is fundamental to conducting valid experiments in psychology, ensuring that observed effects are genuinely due to the manipulated independent variable.

📖 9. Ethical Considerations

🔑 Key Concepts & Definitions

Informed Consent (BEAUCHAMP & CHILDRESS, 2013): Participants must be fully aware of the nature of the research, including any potential risks, and voluntarily agree to participate without coercion.

Deception (MORGAN & MORGAN, 2013): Deliberately misleading participants about the true purpose of the study or withholding information to prevent bias; must be justified and followed by debriefing.

Debriefing (MORGAN & MORGAN, 2013): Post-study process where researchers explain the true purpose, clarify any deception used, and address participant concerns to ensure ethical integrity.

Confidentiality (BEAUCHAMP & CHILDRESS, 2013): Ensuring that personal data collected during research are kept private and used only for the intended purposes, protecting participant identity.

Protection from Harm (APA, 2017): Researchers have a duty to minimize physical and psychological risks, ensuring participants are not exposed to unnecessary harm or distress.

Ethical Approval (British Psychological Society, 2014): Formal review and approval by an ethics committee or institutional review board to ensure research adheres to ethical standards before data collection begins.

📝 Essential Points

• Ethical guidelines in psychology emphasize informed consent to respect participant autonomy, requiring clear communication about the study’s nature and voluntary participation (Beauchamp & Childress, 2013).
• Deception must be justified; it is only acceptable if the benefits outweigh potential harm and if participants are debriefed thoroughly afterward (Morgan & Morgan, 2013).
• Maintaining confidentiality is crucial; researchers must anonymize data and securely store personal information to protect participant privacy (Beauchamp & Childress, 2013).
• Researchers are responsible for protecting participants from harm, including psychological distress, and should have protocols for managing adverse effects (APA, 2017).
• Ethical approval from relevant bodies ensures that research meets established standards, preventing unethical practices such as exploitation or harm.
• Historical examples, like the prefrontal lobotomy (see source content), highlight the importance of ethical oversight, as lack of systematic review led to harmful procedures being conducted without proper consent or safety measures.

💡 Key Takeaway

Ethical considerations in psychology safeguard participant rights and well-being, ensuring research is conducted responsibly and with integrity, guided by principles like informed consent, confidentiality, and protection from harm.

📖 10. Data Analysis and Interpretation

🔑 Key Concepts & Definitions

Data Analysis (see source content): The process of examining, processing, and interpreting collected data to draw meaningful conclusions about the research hypothesis or question.

Operationalisation (implied): The process of defining variables in measurable terms, such as specifying how the independent and dependent variables will be manipulated or measured in a study.

Statistical Significance (see source content): A determination made through statistical tests indicating whether the observed results are unlikely to have occurred by chance, thus supporting the hypothesis.

Correlation (see source content): A statistical relationship between two variables, indicating how they vary together, but not implying causation.

Confounding Variable (see source content): An extraneous factor that influences both the independent and dependent variables, potentially leading to invalid conclusions if not controlled.

Type of Data (see source content): The classification of data into categories such as qualitative, quantitative, subjective, objective, continuous, or discrete, which influences the choice of analysis methods.

📝 Essential Points

• Data analysis involves processing raw data collected via various techniques (observation, interviews, questionnaires, technology-assisted methods) to evaluate hypotheses (see section 7 for data types and scales).
• Proper operationalisation of variables ensures clarity in how data is collected and interpreted, crucial for valid analysis.
• Statistical tests are used to determine whether results are statistically significant, helping researchers infer if the independent variable caused changes in the dependent variable.
• Correlational studies analyze relationships between variables without manipulating them, useful when experiments are unethical or impractical (see section 2 for research design types).
• Recognizing confounding variables is vital; they can produce misleading results if not controlled, as illustrated by the gender example in the source content.
• Data can be qualitative or quantitative, subjective or objective, and continuous or discrete, affecting the choice of analysis techniques.
• The scale of measurement (nominal, ordinal, interval, ratio) determines the appropriate statistical tests, with ratio data allowing the most powerful analyses (see section 8 for scales of measurement).

💡 Key Takeaway

Effective data analysis and interpretation are essential for drawing valid, evidence-based conclusions in psychological research, requiring careful operationalisation, understanding of data types, and control of confounding variables to avoid biased or invalid results.

📊 Synthesis Tables

⚠️ Common Pitfalls & Confusions

1. Confusing extraneous variables with confounding variables; extraneous variables are controlled, confounding variables are not.
2. Assuming correlation implies causation; correlational studies do not establish cause-and-effect.
3. Failing to operationalise variables clearly, leading to measurement inconsistencies.
4. Overlooking participant variables, which can bias results if not randomized or matched.
5. Misunderstanding the difference between control variables (kept constant) and confounding variables (uncontrolled).
6. Ignoring order effects in repeated measures designs, which can skew results.
7. Assuming all variables are easily manipulated; some (e.g., trauma) are unethical to manipulate.

✅ Exam Checklist

• Know the definitions of independent variable, dependent variable, extraneous variable, and confounding variable.
• Understand the process and importance of operationalisation of variables.
• Be able to distinguish between experimental and correlational research designs, including their strengths and limitations.
• Know the purpose and function of control groups, random allocation, and matching participants.
• Recognize different data collection techniques: observation, questionnaires, interviews, and technological tools.
• Understand the types of data and scales used in psychology: nominal, ordinal, interval, ratio.
• Be familiar with control variables and how they prevent confounding effects.
• Know ethical considerations in research, including informed consent, confidentiality, and protection from harm.
• Understand the importance of data analysis methods and how to interpret results accurately.
• Recall key authors and their concepts: Campbell & Stanley on experimental control, and SMITH's definition of the invisible hand (if applicable).
• Be able to identify potential confounding variables and how to control them.
• Know the differences between independent groups, repeated measures, and matched participants designs.
• Recognize the importance of operationalising variables for validity and replicability.