Experimental Methods in Psychology Chapter 5

Research Methods - Content Booklet

• Experimental method covers types of experiments (laboratory, field, natural, quasi).
• Observational techniques include naturalistic, controlled, covert, overt, and participant observation.
• Self-report techniques involve questionnaires and interviews (structured and unstructured).
• Correlations analyze relationships between co-variables, contrasting them with experiments.
• Content analysis examines the content of communication.
• Case studies focus on detailed investigations of individuals or small groups.
• Scientific processes include aims, hypotheses (directional, non-directional), and sampling methods (random, systematic, stratified, opportunity, volunteer).
• Pilot studies, experimental designs (repeated measures, independent groups, matched pairs), and observational designs (event sampling, time sampling) are also part of the specification content.
• Variables (independent, dependent, extraneous, confounding) need to be appropriately operationalised/defined.
• Control techniques like random allocation and counterbalancing are essential to ensure validity.
• Demand characteristics and investigator effects are key issues.

Scientific Processes: Aims & Hypotheses

• Aims define the study's general purpose.
• Hypotheses are precise, testable predictions.
• Alternative and null hypotheses exist in all studies (directional and non-directional)

Data Handling and Analysis

• Quantitative data represents numerical data (counts, measures).
• Qualitative data involves descriptive data (words, observations).
• Descriptive statistics include measures of central tendency (mean, median, mode) and dispersion (range, standard deviation).
• Inferential statistics involve using statistical tests to determine the probability of findings being due to chance.

Experimental Methods Summary

• Laboratory Experiment: High control over variables, artificial environment, which can limit generalizability and increase demand characteristics impacting external validity. However, cause-and-effect relationships may be more easily established.
• Field Experiment: Manipulated IV in a natural setting, but less control over extraneous variables, making causality harder to pin down but improves ecological validity.
• Natural Experiment: Naturally occurring changes in the IV, less control but high ecological validity; however, cause-and-effect is difficult to determine.
• Quasi-Experiment: An existing difference between groups (e.g., gender, age) is the IV, high ecological validity but no manipulation, limiting causal interpretations.

Observational Techniques

• Naturalistic Observation: Observing behaviour in a natural setting without intervention, high ecological validity but lacks control, making it difficult to establish causality.
• Controlled Observation: Observing behaviour in a structured environment, more controlled than naturalistic observation; however, it may lack ecological validity.

Self-Report Techniques

• Questionnaires: A series of questions, can collect large amounts of data efficiently but responses may not be truthful or properly understood.
• Structured Interviews: Pre-determined questions, good for replication, but can be inflexible and lack depth.
• Unstructured Interviews: Unstructured questions and responses, richer data, but harder to analyze & might contain subjective biases.

Correlations

• Correlational analyses explore the relationship between two co-variables but cannot conclude cause-and-effect.
• Correlations can be positive, negative, or zero, and are quantified by correlation coefficients.

Features of Science

• Objectivity: Unbiased research, no researcher influence.
• Empirical method: Data-driven approach.
• Replicability: Consistent findings when repeated.
• Falsifiability: Ability for a theory to be proven wrong.

Reporting Psychological Investigations

• Structure of a scientific report (abstract, introduction, method, results, discussion, references).
• Data handling using descriptive & inferential statistics
• Graphs (bar charts, histograms, scattergrams) and tables for visualizing data.

Ethics in Psychological Research

• Informed consent; Deception; Protection from harm; Confidentiality, Anonymity; Debriefing & Right to withdraw are essential, and are outlined in BPS guidelines. Researchers are responsible for ethical oversight.

Improving Reliability & Validity

• Reliability: Consistency of measurement.
• Validity: Accuracy of the measure.
• Various techniques may improve both depending on the research method and purpose.

Statistical Tests

• Statistical tests are used. Understanding statistical analysis is essential to interpret results.
• Different tests are used depending on the type of data (nominal, ordinal, interval) and the research design.