The process of scientific research in psychology 3

The Process of Scientific Research in Psychology

Basics of Research Methods

• Research Methods Course: PSYB19-104 by Zsolt Horváth, PhD

• Email: horvath.zsolt@ppk.elte.hu

Planning Research

• Key Considerations:

  • WHAT to measure (variables involved)

  • Design plan detailing HOW to answer the research question

  • SAMPLE determination: whom to study and how to find participants

  • Analysis method: type of evidence and its format

Variables in Research

• Definition: Phenomena/concepts that can change and be measured

• Types of Variables:

  • Quantitative Variables: Represented mathematically/statistically (height, attitudes, etc.)

  • Categorical Variables: Non-numerical, discrete categories (marital status)

  • Measured Variables: Numerical values indicating degree along a scale (extroversion, anxiety)

• Variability:

  • Within individuals over time

  • Between different individuals

Examples of Variables

• Height Distribution: Example showing variation in a population's heights

• Relationship Satisfaction Example: Comparison of participants' satisfaction at different time points (e.g., October 2024 vs. December 2024)

Measuring Variables

Challenges in Measurement

• Easy-to-Measure Variables: E.g., weight, height, age

• Difficult-to-Measure Variables: E.g., attitudes, feelings, anxiety, extroversion

• Crude Measurements: Depending on subjective ranking which lacks qualitative insights

Operational Definitions

• Importance: Distinction between constructs and their measurements

• Example of Aggression:

  • Measurement using behaviors (e.g., recorded arguments) and questionnaires

Naturalistic Observations

• Context: Identifying bullying episodes through observed behaviors in children's interactions

• Reliability Measures: Ensuring consistency in identifying bullying incidents (interrater reliability measures)

Constructs and Measurements

Hypothetical Constructs

• Examples: Anxiety, self-esteem, intelligence believed to explain observed behaviors or phenomena

• Measurement Issues: Constructs often rely on observable behaviors to infer underlying concepts which are not directly measurable.

Reliability and Validity in Research

• Reliability: Consistency of measurements across time and observers

• Validity: Degree to which a measure accurately reflects the concept it intends to measure

  • Examples: Psychological scales and the potential pitfalls in wrongly attributing constructs (e.g., measuring aggression instead of assertiveness)

Samples and Sampling Methods

Population and Sample Definitions

• Population: All potential members of a group

• Sample: Subset of the population for investigation

• Participant Definition: Individuals participating in a psychological investigation

• Sampling Frame: The specific portion of the population accessible for study

Issues in Sampling

• Sampling Bias: Systematic representation issues can lead to inaccuracies in generalizing findings

  • Example: Convenience sampling often leads to over- or under-representation of specific subgroups.

• Types of Samples:

  • Random Sample: Each member has an equal chance of selection

  • Stratified Sampling: Ensures representation of known subcategories in the population

Causality in Research Designs

Understanding Causation

• Causal Relationships: Some psychological events may influence others but require strong evidence to establish causality.

• Common Issues:

  • Correlation does not imply causation, presence of confounding variables, and limitations of data measurements

Research Design Types

• Cross-Sectional Studies: Measure variables at a single time point with no causal inference possible

• Longitudinal Studies: Measure changes over time, providing more robust evidence for causality

• Randomized Experiments: Participants randomly assigned to conditions to control for confounding variables

Developing Hypotheses

Importance of Hypotheses

• Definition: Statements predicting the relationship between variables based on existing research

• Types:

  • Causal vs. non-causal

  • Directional vs. non-directional

  • Examples of directional hypotheses indicating expected outcomes

Statistical Hypothesis Testing

• Null Hypothesis (H0): No relationship exists between the variables

• Alternative Hypothesis (HA): A relationship is presumed to exist

• Testing Methods: One-tailed vs. two-tailed testing approaches and their applications

Concluding the Research

• Final Steps: Formulating research aims, defining hypotheses, and deciding on research design and required statistical analysis.