Research Methods and Statistics

Basic Vocabulary

• Hypothesis: A tentative explanation that must be falsifiable, meaning it should be possible to support or reject it.
• Operational Definition: A clear, precise, and quantifiable definition of variables, allowing for replication and reliable data collection.
• Qualitative Data: Descriptive data (e.g., eye color).
• Quantitative Data: Numerical data, which is ideal and necessary for statistical analysis.
• Population: The total group that the research could apply to.
• Sample: The specific individuals selected for the study.

Research Designs

Correlation

• Definition: Identifies the relationship between two variables.
• Advantages: Useful when experiments are unethical.
• Disadvantages: Correlation does not imply causation.
  • Directionality Problem: Uncertainty about which variable influences the other (e.g., does depression cause low self-esteem or vice versa?).
  • Third Variable Problem: Another variable may be responsible for the observed relationship (e.g., ice cream sales correlating with murder rates).
• Types of Correlation:
  • Positive Correlation: Both variables increase or decrease together.
  • Negative Correlation: As one variable increases, the other decreases.
  • Correlation strength is indicated by the absolute value of the correlation coefficient. Values closer to 1 (or -1) indicate a stronger relationship.

Experimental Research

• Placebo Effect: Observed behavior changes due to a placebo, highlighting the treatment's effectiveness.
• Blinding Techniques:
  • Double-Blind: Neither participants nor experimenters know who is assigned to which condition.
  • Single-Blind: Participants are unaware of their group assignment.
• Confound: An error or flaw in the study, also known as a confounding variable.
• Inferential Statistics: Methods for analyzing data to infer conclusions about the population from the sample.
• Random Assignment: Participants are randomly assigned to control or experimental groups, enhancing representativeness and allowing for causal inferences.

Other Study Types

• Naturalistic Observation: Observing people in their natural environments without interference.
• Experiments: Purposefully manipulate variables to establish cause-and-effect relationships.
  • Independent Variable: The variable that is altered by the researcher.
  • Experimental Group: The group receiving the treatment related to the independent variable.
  • Control Group: Receives no treatment or a placebo; serves as a baseline.
  • Dependent Variable: Outcomes measured in response to manipulations of the independent variable.

Statistical Concepts

• Statistical Significance: Indicates results are unlikely due to chance if p < 0.05.
• Effect Size: Indicates the practical significance of data (larger effect sizes indicate more meaningful results).

Ethical Guidelines

• Confidentiality: Participants' identities must be protected.
• Informed Consent: Participants must agree to be part of the study informedly.
• Informed Assent: Minors and their parents must consent.
• Debriefing: Participants must be informed of the study's true purpose after completion, especially in cases of deception.
• No Harm: Studies must not inflict mental or physical harm.

Additional Vocabulary

• Surveys: Often turned into correlations but subject to bias due to self-reporting errors.
  • Advantages: High real-world validity.
  • Disadvantages: Cannot establish cause-and-effect relationships.
• Case Study: Intense study of one person, providing detailed information but lacking causal inferences.
• Meta-Analysis: Combines multiple studies to increase sample size and evaluate effect sizes.

Descriptive Statistics

Measures of Central Tendency

• Mean: The average, useful in normal distributions.
• Median: The middle value, best for skewed distributions.
• Mode: The most frequently occurring value.
  • Bimodal Distribution: Two modes exist, indicating mixed results.
• Skewness: Data distribution can be positively or negatively skewed, affecting mean, median, and mode.

Sampling Techniques

• Random Sampling: Everyone has an equal chance of being selected, improving generalizability.
• Representative Sampling: Sample adequately reflects the general population regarding key demographics (e.g., age, ethnicity).
• Convenience Sampling: Selecting participants based on availability, often leading to bias and reduced generalizability.
• Sampling Bias: When the sample does not reflect the population accurately, affecting study outcomes.

Cognitive and Participant Bias

• Experimenter Bias: Researcher’s expectations affecting results.
• Participant Bias: Participant expectations influencing their behavior.
• Cognitive Bias: Includes confirmation bias (favoring information that supports preexisting beliefs) and hindsight bias (post-event conviction that one knew the outcome).
• Hawthorne Effect: Modification of behavior by participants due to awareness of being observed.

Measures of Variation

• Range: Difference between the smallest and largest values in a dataset.
• Standard Deviation: Average distance of data points from the mean, indicating score spread (larger values suggest more variability).