Psychological Research Methods and Statistics

Research Methods and Sampling

• Sample Selection:

  • Sample: A representative subgroup chosen from the total population under study to ensure manageable and accurate data collection.
  • Random Sample: Selection method where every individual in the population has an equal likelihood of being chosen.
  • Stratified Sample: Deliberate selection where population subgroups are proportionally represented within the sample.

• Non-Experimental Research Methods:

  • Naturalistic Observation: Observing subjects unobtrusively in everyday environments without experimental intervention to record natural behavior.
  • Case Study: Intensive, qualitative investigation of an individual or small group; generates detailed hypotheses but findings cannot be generalized to broader populations.
  • Surveys: Data gathering via interviews or questionnaires; interviews offer conversational flexibility while questionnaires maximize standardization and eliminate experimenter cues.
  • Longitudinal Studies: Assessing the same cohort repeatedly over extended periods to track developmental stability and change.
  • Cross-Sectional Studies: Comparing distinct age groups simultaneously to identify age-related differences efficiently.

• Correlational Research:

  • Correlation: A statistical measure describing the relationship between two variables; does not identify causation or rule out third-factor influences.
  • Directionality: Positive correlations indicate variables change in the same direction; negative correlations indicate variables change in opposite directions.

• Experimental Framework:
  • Hypothesis: An educated, testable prediction defining the expected relationship between variables.
  • Independent Variable (IV): The factor deliberately altered or manipulated by the experimenter.
  • Dependent Variable (DV): The variable measured to assess the effect of the independent variable.
  • Experimental Group: Participants exposed to the independent variable manipulation.
  • Control Group: Participants treated identically to the experimental group but unexposed to the independent variable; serves as the baseline for comparison.
  • Replication: Duplication of an experiment with different participants to verify results and eliminate hidden flaws.

• Research Ethics:
  • Ethical guidelines established by the American Psychological Association in $1992$ require minimizing potential harm, safeguarding dignity and confidentiality, securing signed informed consent, and strictly limiting deceptive practices.

Research Hazards and Experimental Controls

• Biases and Methodological Hazards:

  • Self-Fulfilling Prophecy: An experimenter unknowingly communicates expectations through subtle cues (such as nodding or eyebrow raises), leading participants to behave in alignment with those expectations.
  • Hawthorne Effect: Alterations in participant behavior resulting strictly from the awareness of receiving special experimental attention rather than the experimental treatments tested.
  • Placebo Effect: Measurable changes in a participant's physical condition or behavior resulting entirely from their belief in the efficacy of a treatment rather than an active therapeutic agent.

• Experimental Control Procedures:

  • Single-Blind Experiment: Research design where participants remain unaware of whether they are assigned to the active treatment group or the control (placebo) group.
  • Double-Blind Experiment: Research design where neither the participants nor the experimenters directly interacting with them know who received the active treatment or the placebo, preventing researcher bias.

• Demonstrations of Bias and Influence:

  • The Milgram Experiment ($1960\text{s}$): Tested obedience to perceived authority; despite displaying extreme distress, $65\%$ of original male participants ($26$ of $40$) administered maximum shocks up to $450\,\text{V}$. A replication at Swarthmore College resulted in an $88\%$ full obedience rate.
  • Clever Hans Case ($1911$): Oskar Pfungst determined that the horse Hans was not calculating mathematical solutions but responding to unconscious, involuntary visual cues (such as leaning forward or nostril dilation) given by the questioner.

Statistical Evaluation and Analysis

• Descriptive Statistics:

  • Frequency Distributions: Organized presentations of raw data showing how often each score or value occurs, frequently illustrated through histograms or frequency polygons.
  • Normal Curve: Symmetrical, bell-shaped distribution where values peak at the center and decrease evenly toward both tails.

• Measures of Central Tendency:

  • Mode: The score that occurs most frequently within a data set.
  • Median: The middle value that divides a rank-ordered distribution exactly into halves ($50\%$ above and $50\%$ below).
  • Mean: The arithmetic average calculated by summing all individual values and dividing by the total number of observations ($N$).

• Measures of Variance:

  • Range: A crude estimate of spread computed by subtracting the lowest score from the highest score and adding $1$.
  • Standard Deviation: A comprehensive measure of dispersion reflecting the average distance of scores from the distribution mean; larger values indicate widely dispersed data, while smaller values indicate tightly clustered data.
  • Standard Score: Transformed score denoting relative distance from the mean in standard deviation units to allow comparison across distinct distributions.

• Correlation Coefficients:

  • Pearson Correlation Coefficient ($r$): Quantifies the strength and direction of association between two variables, ranging from $-1.0$ to $+1.0$.
  • Magnitude: An $r$ value between $0$ and $\pm 0.30$ denotes a weak relationship; $\pm 0.30$ to $\pm 0.60$ denotes a moderate relationship; $\pm 0.60$ to $\pm 1.0$ denotes a strong relationship; $\pm 1.0$ indicates a perfect correlation.

• Inferential Statistics and Significance:

  • Purpose: Mathematical procedures used to determine whether observed findings reflect genuine population patterns or mere chance occurrences.
  • Statistical Significance: A conclusion that results are unlikely due to random variation, conventionally accepted when the probability ($p$) that findings are due to chance is less than $0.05$ ($5\%$) or $0.01$ ($1\%$).