Research Methods in Psychology

Research Methods

• Case studies: Detailed investigations of individuals, groups, or events, often yielding a large amount of information.
• Content analysis: A method used to identify the presence of certain words, themes, or concepts in qualitative data.
• Controlled observation: Observing participants in a lab setting, providing high control and reliability but potentially reducing ecological validity.
• Correlation: Measures the relationship between two variables.
• Covert observation: Observational method where participants are unaware of being observed, potentially reducing demand characteristics but raising ethical concerns about consent..
• Experiments: Investigate hypotheses by manipulating an independent variable to observe its effect on a dependent variable.
• Field experiments: Conducted in real-life settings, improving ecological validity but decreasing control over extraneous variables.
• Interviews: Self-report method where participants answer questions posed by an interviewer. Allows for flexibility in gathered information.
• Laboratory experiments: Conducted in highly controlled environments, allowing precise control over extraneous variables, potentially compromising ecological validity.
• Natural experiments: Observe pre-existing independent variables in natural settings, offering high ecological validity but reducing control over extraneous variables.
• Naturalistic observation: Participants are observed in their natural environment, maintaining high ecological validity but with less control.

Observation

• Non-participant observation: The observer does not participate in the activity being observed, potentially increasing objectivity.
• Participant observation: The observer participates in the activity being observed, possibly gaining deeper insights but potentially compromising objectivity.
• Overt observation: Participants know they are being observed.
• Covert observation: Participants do not know they are being observed.
• Observation: A general approach to data collection where behavior is observed.

Quasi-experiments

• Quasi experiments are similar to true experiments but do not involve the manipulation of variables by researchers.
• They investigate relationships between variables rather than cause-and-effect.
• They are useful when manipulation is unethical or impractical, providing useful information.
• See "Natural experiment".

Questionnaires

• Instruments where set questions are given to participants, often paper or electronic.
• Allow for anonymity, a key feature for some studies.

Interviews

• Structured interviews: Pre-set questions, allowing for replication and comparability; these lack flexibility.
• Unstructured interviews: Natural conversations, providing rich, detailed data, with less control over content and no repeatability.

Scientific Processes

• Abstract: Summary of the complete report in research reports.
• Aims: The purpose or objective of an experiment.
• Bias: A predisposition or inclination that could influence results or judgments.
• Behavioural Categories: Categorizing behaviors into specific components; observable and measurable units.
• Closed Questions: Questions that require a limited range of responses (e.g., yes/no).
• Concurrent Validity: The extent to which a test measures what it aims to measure in comparison to an existing well-established test.
• Confounding Variables: Variables that are correlated with the independent variable and also effect the dependent variable. These distort relationships.
• Control Variables: Variables that remain unchanged during an experiment, isolating the influence of the independent variable.
• Counterbalancing: An experimental design technique where participants experience conditions in different orders to control for order effects.
• Demand Characteristics: Participants modify their behavior in response to cues or expectations about the experiment.
• Dependent Variable: The variable that is measured in an experiment to determine the effect of the independent variable.
• Directional Hypothesis: Hypothesis specifying the expected direction of a relationship between variables.
• Ecological Validity: The extent to which results relate to the natural environment.
• Event Sampling: Recording specific behavior events as they occur.
• Extraneous Variables: Variables other than the independent variable that may impact the results.
• Face Validity: The extent to which a test seems valid at a superficial level.
• Falsifiability: The capacity for a hypothesis to be proven wrong.
• Generalisation: Applying results from a sample to the broader population.
• Hypotheses: A prediction of the outcome of an experiment.

Experimental Designs

• Independent Groups: Different participants are in each condition of the experiment.
• Matched Pairs: Participants in different conditions are matched based on relevant characteristics to limit participant variables.
• Repeated Measures: Same participants take part in all conditions of the experiment.

Data Handling and Analysis

• Bar charts: Graphical depiction of categorical data using numerical values.

• Coding: Simplifying large amounts of text using keywords and themes.

• Correlation: The relationship between two variables.

• Correlation coefficient: A numerical value (-1 to +1) describing the strength and direction of a correlation.

• Interval: A level of numeric measurement with equal distances between values.

• Mean: The average of a group of data points.

• Median: The middle number of a sorted set of data.

• Mode: The most frequently occurring score in a data set.

• Nominal: Categorical level of measurement; values are not ordered numerically.

• Normal Distribution: A bell-shaped curve, common in many natural phenomena.

• Ordinal: A level of measurement where data can be ranked but distances between values aren't equal.

• Operationalisation: Defining abstract concepts in concrete, observable terms.

• Opportunity Sampling: Gathering participants who are available at the time.

• Paradigm: Set of core concepts that guide work in a specific field.

• Paradigm Shift: Major shifts in a field's foundational concepts.

• Peer Review: Evaluation of research by expert colleagues before publication.

• Pilot Studies: Initial trial runs to evaluate and refine research methods.

• Qualitative Data: Non-numerical data, e.g. descriptions of behavior or experiences.

• Quantitative Data: Numerical data, e.g. counts, measurements.

• Random Allocation: Assigning participants randomly to different conditions within an experiment.

• Random Sampling: Selecting participants randomly from a population to ensure representation.

• Range: The difference between the highest and lowest values in a data set.

• Reliability: Consistent results when a test or study is repeated.

• Repeated measures: Same participants complete all conditions.

• Sampling Techniques: Methods of selecting participants for a study (such as Random Sampling, Opportunity Sampling, Volunteer Sampling).

• Scattergrams: A graph showing the relationship between two variables; used for correlations.

• Secondary Data: Data collected by others, pre-existing.

• Skewed Distribution: A non-symmetrical distribution of data.

• Standardisation: Ensuring that procedures and actions in an experiment are identical each time.

• Standard Deviation: A measure of the spread of data around the mean.

• Stratified Sampling: Division of the population into strata followed by selection of participants from each stratum.

• Systematic Sampling: Selecting participants at fixed intervals from a population list.

• Temporal Validity: The extent to which results can be applied across time periods (different time periods).

• Test-retest Reliability: Consistency of a test over time.

• Thematic Analysis: Identifying recurring patterns and themes in qualitative data.

• Time Sampling: Observing and recording behavior at specified intervals

• Validity: Accuracy of measuring what a test intends, whether it's face or concurrent.

• Volunteer Sampling: Participants actively choose to participate.

Data representation

• Bar charts: Display categorical data with bars of varying heights, often using numerical values.
• Scattergrams: Show the relationship between two variables by plotting data points on a graph; useful for correlations.