RESEARCH METHODS

VARIABLES AND CONTROLS

Independent Variable (IV)

• Definition: A variable that is manipulated to test its effect on the dependent variable.

• Conditions: At least two conditions; typically includes an experimental condition and a control condition.

  • Experimental condition: Group exposed to the IV (e.g., doodling group in Andrade).

  • Control condition: Group not exposed to the IV for comparison (e.g., control group in Andrade did not doodle).

• Allocation: Participants can be assigned to groups by researcher choice or random allocation.

  • Advantages: Reduces bias; increases validity since participant types vary.

  • Weaknesses: Random allocation may result in individual differences between conditions affecting validity (e.g., intelligence variability).

Dependent Variable (DV)

• Definition: The variable being measured to determine the impact of the IV.

• Example: In Andrade, the DV was the participants' scores on monitoring and recall tasks.

Confounding Variables

• Definition: Variables other than the IV that may affect the DV, introducing confusion and lowering validity.

• Types:

  • Participant variables: Characteristics like personality, age, gender, intelligence, and memory.

  • Situational variables: Conditions inherent to the study environment such as lighting, noise, etc.

  • Uncontrollable variables: Confounding variables that cannot be eliminated from the study, negatively impacting validity.

Controls

• Definition: Measures taken to minimize or eliminate confounding variables in a study.

• Example: Splitting participants into AC condition and control condition to compare concentration effects.

• Importance: Controlling variables enhances the validity and reliability of the results. High levels of control standardize procedures enhancing replicability.

VALIDITY AND ITS TYPES

Validity

• Definition: Refers to how accurately a study measures what it is intended to measure.

• Experiment validity hinges on ensuring that only the IV affects the DV, without confounding variables interfering.

• Demand Characteristics: Participants knowing the study's true aim might change behavior, reducing validity.

• Socially Desirable Responses: Participants may answer in a way they think is socially acceptable rather than truthfully.

Enhancing Validity

• Double-blind Technique: Neither participants nor observers know which condition participants are in to avoid demand characteristics and researcher bias.

Ecological Validity

• Definition: The degree to which a study's findings can be generalized to real-life settings.

• Higher ecological validity arises from natural settings (field experiments), while artificial lab settings generally reduce it.

• Mundane Realism: The similarity of tasks in a study to everyday tasks (e.g., helping a person vs. giving an electric shock).

Temporal Validity

• Definition: Relates to whether a measure reliably reflects traits over time.

Criterion Validity

• Definition: Measures how well one variable predicts another.

• Types:

  • Predictive Validity: Measure's ability to predict future outcomes (e.g., personality tests predicting job performance).

  • Concurrent Validity: Measure correlating well with criteria assessed simultaneously (e.g., depression scale correlating with a clinical diagnosis).

RELIABILITY AND ITS TYPES

Reliability

• Definition: Study's consistency realized through high controls leading to replicable and repeatable results.

• Standardization is critical, ensuring uniform procedures across all participants.

Types of Reliability

1. Inter-Rater Reliability: Consistency between two observers rating the same behavior.

2. Inter-Observer Reliability: Observers report the same consistent behaviors rather than rate them.

3. Test-Retest Reliability: Checking for consistency of a questionnaire or task over separate occasions.

4. Split-Half Method: Assessing questionnaire consistency by splitting it into halves and comparing results.

GENERALISABILITY

• Definition: The extent to which study findings can be applied to a broader population.

• Representative Sample: Larger, diverse samples yield higher generalizability compared to small or homogeneous samples.

• Example: A study involving 10 women vs. 5000 multigendered individuals spanning various ages across regions.

ORDER EFFECTS

• Definition: Changes in participant behavior due to task order.

• Types:

  1. Practice Effects: Improvement on repeated tasks due to memorization.

  2. Fatigue Effects: Decreased performance from tiredness or boredom.

• Solutions:

  • Randomization: Random distribution of task sequences.

  • Counterbalancing: Balancing order (AB/BA) to mitigate order effects.

  • Independent Measures Design: Prevents exposure as different participants engage in each condition.

DATA TYPES

Quantitative Data

• Objective, numerical data suitable for comparisons but lacks insights on 'why' outcomes occur.

Qualitative Data

• Detailed, subjective, behavioral insights explaining participant actions.

ETHICS

Ethical Guidelines for Humans

1. Deception: Ethical concerns regarding misleading participants.

2. Informed Consent: Required permission from participants, with clarity on study purpose.

3. Right to Withdraw: Participants may exit at any time.

4. Protection from Harm: Ensuring physical and psychological safety.

5. Confidentiality: Safeguarding personal data.

6. Debriefing: Explaining the actual purpose post-study to alleviate any distress caused by deception.

Ethical Guidelines for Animals

1. Numbers: Using the fewest animals necessary for valid results.

2. Replacement: Using alternatives to animal testing where feasible.

3. Pain and Distress: Minimizing animal suffering during research.

4. Reward and Housing Considerations: Ensuring enrichment and appropriate social conditions.

RESEARCH METHOD TECHNIQUES

Experiments

• Types: Lab, Field, Natural.

Observations

• Types: Overt/Covert, Participant/Non-Participant, Structured/Unstructured, Naturalistic/Controlled.

Self-Report Methods

• Types: Interviews (Structured, Unstructured, Semi-Structured), Questionnaires.

Case Studies

Correlations

• Positive and Negative correlations detailed in findings context.

Longitudinal and Cross-Sectional Studies

• Differences in participant observation across time frames.

Experimental Design

1. Independent Measures/Groups Design

2. Repeated Measures/Groups Design

3. Matched Pairs Design

SAMPLING METHODS

Types

1. Opportunity Sample (convenient, but may lead to bias).

2. Volunteer Sample (self-selecting, ethical, but time-consuming).

3. Random Sample (ensures equal chance, but may limit generalizability).