Research Methodology in Psychology
The Research Process
Psychologists perform research to test hypotheses and find evidence to support ideas. The process generally follows these steps:
- Development of an aim and hypothesis.
- Selection of a research method and experimental design.
- Definition and control of variables.
- Consideration of ethics.
- Selection of participants.
- Data analysis and drawing conclusions.
- Evaluation of the research for future improvements.
Experimental Research Methods
Experimental methods are used to establish cause-and-effect by investigating how an independent variable () causes changes in a dependent variable (). Researchers use experimental groups (which receive the ) and control groups (which do not) for comparison. Extraneous variables must be controlled to ensure validity.
Laboratory Experiments
Studies take place in artificial, well-controlled environments.
- Strengths: High standardization, easy to replicate, and high control over situational variables.
- Weaknesses: Low ecological validity (realism of environment) and low mundane realism (realism of tasks). Participants may show demand characteristics.
Field Experiments
Studies occur in the participant's natural environment (e.g., a classroom or workplace).
- Strengths: High ecological validity and lower demand characteristics.
- Weaknesses: Less control over extraneous variables, difficult to replicate, and potential ethics issues regarding informed consent.
Quasi or Natural Experiments
These use naturally occurring s (such as age, gender, or health status) where subjects cannot be randomly assigned.
- Strengths: Allows for the study of variables that cannot be manipulated.
- Weaknesses: Cannot randomly assign subjects, leading to potential influence from participant variables.
Experimental Research Designs
- Independent Measures: Subjects are placed into different groups, with each group performing only one level of the .
- Repeated Measures: The same group of people participates in every level of the . This reduces participant variables and requires fewer subjects but may lead to order, practice, or fatigue effects. These effects can be reduced through counterbalancing.
- Matched Pairs: A variation of independent measures where participants are paired based on similar characteristics.
Non-Experimental Research Methods
Non-experimental methods can show relationships between variables but cannot establish cause-and-effect.
Self-Reports
- Questionnaires: Written answers. Questions can be open-ended (qualitative) or closed-ended (quantitative). Closed-ended tools include Likert scales (e.g., strongly disagree to strongly agree) and rating scales (e.g., to ).
- Interviews: Spoken answers. They can be structured (same questions/order), unstructured (thematic, based on previous answers), or semi-structured.
Observations
Researchers record behaviors, often using behavioral checklists.
- Overt/Covert: Participants do or do not know they are being observed.
- Participant/Non-participant: Researcher is or is not part of the group.
- Sampling: Time sampling (recording at intervals) or event sampling (tallying every instance of behavior).
Case Studies
Detailed, in-depth analysis of an individual or small group over time (longitudinal). They provide rich data but are difficult to generalize.
Correlations
Measure the relationship between two quantitative variables.
- Types: Positive (both values move together), negative (values move in opposite directions), or no correlation.
- Strength: Measured from to . A strength of indicates no trend.
Temporal Research Descriptions
- Snapshot Study: Data is collected at one point in time to compare groups.
- Longitudinal Study: Follows the same group of participants over an extended period at specific intervals. This is used in case studies and repeated measures experiments to track changes over time or following specific events.