chapter 2
Chapter 2: Methods in Psychology
Chapter Outline
Empiricism: How to Know Stuff
Methods of Observation: Discovering What People Do
Methods of Explanation: Discovering Why People Do What They Do
Thinking Critically About Evidence
The Ethics of Science: Doing What’s Right
Empiricism: How to Know Stuff
Scientific Method: A set of principles describing the appropriate relationship between ideas based on empirical evidence.
Empiricism: The belief that accurate knowledge can be acquired through observation; a critical element in the scientific method.
Dogmatism: The tendency to cling to one’s beliefs without considering alternative perspectives.
Historical Perspective:
Euclid and Ptolemy believed vision was due to rays emitted from the eyes.
Al-Haytham (965-1039) challenged this view.
The Scientific Method
Scientific Method: Procedure for utilizing empirical evidence to establish facts.
Theory: An explanation of a natural phenomenon that can never be conclusively proven correct.
Hypothesis: A falsifiable prediction that originates from a theory.
The Art of Looking
Empirical Method: A set of rules and techniques designed for observation.
Challenges in Study: People are difficult to study due to their complexity, variability, and reactivity.
Methods of Research:
Methods of Observation
Methods of Explanation
A World of Difference
Gender Differences in Science
Observation on gender representation in science:
Historical limitations on educational and employment opportunities for women.
Noted differences in interests and talents between genders.
Men often show more interest in scientific topics, while women exhibit more variability in quantitative and visuo-spatial abilities.
The World’s First Scientist
Established by William Whewell in 1834, the term "scientist" refers to Mary Somerville, recognized as the first scientist.
Methods of Observation: Discovering What People Do
Observation: Utilizing one's senses to acquire knowledge about events or objects.
Limitations of Everyday Observation:
Inconsistency
Incompleteness
Scientific Techniques to Overcome Limitations:
Measurement
Description
What to Measure and How to Measure It
Operational Definition: The description of a property in measurable terms.
Construct Validity: The extent to which the measured entity accurately reflects the property in question.
Key Features for Measurement Tools:
Reliability: Consistency of measurement results upon repeated use.
Power: The ability of a measure to detect specific conditions as defined in the operational definition.
Measurement
Measurement involves two steps:
Defining the property to be measured.
Designing the method of measurement.
Power and Reliability at the Olympics
Example with Usain Bolt and Yohan Blake:
Bolt’s record of 9.58 seconds vs. Blake’s 9.75 seconds illustrates measurement power.
Ice Dancing Judging: Reliable and consistent ratings by judges reflecting athleticism and artistry, exemplified by Tessa Virtue and Scott Moir.
Demand Characteristics: Doing What Is Expected (Part 1)
Demand Characteristics: Features of an observational setting prompting individuals to behave according to what they believe is expected.
Challenges in behavior measurement due to these characteristics.
Demand Characteristics: Doing What Is Expected (Part 2)
Strategies to Mitigate Demand Characteristics:
Conducting naturalistic observation without subjects’ knowledge.
Application of unobtrusive observation strategies to collect scientific data in natural contexts.
How Do Researchers Measure “Pace of Life”?
Conducting naturalistic observations, exemplified by measuring pedestrian walking speeds across cities.
Demand Characteristics: Doing What Is Expected (Part 3)
Techniques for Avoiding Demand Characteristics:
Ensuring participant privacy
Control over the experimental environment
Keeping participants unaware of the study’s hypotheses.
Observer Bias: Seeing What Is Expected
Observer Bias: The tendency for the expectations of observers to influence what they report observing.
Reasons for Observer Bias:
Expectations can shape perceptions and observations.
Avoidance Strategies:
Employing double-blind studies to mitigate observer biases.
Example of Observer Bias
Illustration using Robert Parker, a prominent wine critic, questioning if his ratings influence the taste perceptions of wines.
Description
Population: The entire group of interest, often hard to measure.
Sample: A subset drawn from the population for analysis.
Psychologists utilize two techniques for data analysis:
Graphic Representations
Descriptive Statistics
Graphic Representations: Picturing the Measurements
Graphic Representations: Used to describe data visually.
Two forms include:
Frequency Distribution: A graph showing the frequency of each measurement value.
Normal Distribution: A defined frequency distribution where most results cluster around the mean.
Frequency Distributions
Investigating reported levels of happiness through frequency distributions.
Descriptive Statistics: Summarizing the Data
Descriptive Statistics: Summary statements capturing essential information from frequency data sets.
Measures of Central Tendency: Include mode, mean, median, and skewed distribution reflections.
Calculating Descriptive Statistics
Formulas and processes to compute various descriptive statistics.
Differently Shaped Distributions
Exploration of how different distributions can illustrate varying data characteristics.
Variability: How Wide Is the Distribution?
Variability: Reflects the degree of variation in measurement; indicates how much individual measurements differ from one another.
Range: The difference between the largest and smallest measurements in a frequency distribution.
Standard Deviation: A statistic indicating the average deviation of measurements from the mean of the distribution.
Distributions Can Differ in Variability or Central Tendency
Evaluating how distributions differ based on central tendency vs. overall variability.
Methods of Explanation: Discovering Why People Do What They Do
Correlation: Describes the relationship between two variables where changes in one are linked with changes in the other.
Variable: Any measurable property that can manifest in multiple values.
Hypothetical Data Showing the Relationship Between Sleep and Money
Participant | Hours of Sleep | No. of Countries Named in One Minute |
|---|---|---|
A | 0 | 11 |
B | 0 | 17 |
… | … | … |
L | 9 | 43 |
Synchronized Patterns of Variation
Measurement and analysis of pairs of variables to identify consistent patterns.
Understanding correlations allows prediction and inference without direct measurement of each variable.
Measuring the Direction and Strength of a Correlation
Correlation direction classified as positive or negative:
Positive Correlation: Characterized by increased values in both variables.
Negative Correlation: Characterized by decreased values in one variable tied to increased values in another.
Correlation Coefficient (r): Ranges from -1.0 (perfect negative correlation) to +1.0 (perfect positive correlation); 0 indicates no correlation.
Graphing Correlations
Visual representations of varying correlation strengths across different graphs.
Positive Correlations of Different Strengths
Analyzing the presence of strong versus weak correlations illustrated through graphical representations of data.
Causation
Caution against assuming causal relationships based solely on correlation.
Natural Correlations: Observed correlations in real-world settings.
Third-variable problem (Z): A scenario where a third variable may be influencing both correlated variables, obscuring direct causation.
Causes of Correlation
Possible relationships:
X → Y: X causes Y.
Y → X: Y causes X.
Z → X & Y: A third variable causes both X and Y.
Experimentation: Establishing Causation
Experimentation: A systematic approach to ascertain causal relationships among variables.
Conducting experiments involves manipulating conditions and observing outcomes through key features.
Three Steps of Experimentation
Manipulate: Selecting an independent variable to alter conditions.
Independent Variable: the variable that is modified in an experiment.
Measure: Taking observations on the dependent variable.
Dependent Variable: the measured outcome variable in an experiment.
Compare: Assessing and contrasting results across conditions.
Manipulation: Making Different Conditions
The act of changing the values of variables actively in an experiment to assess effects.
Random Assignment
Random Assignment: A method using chance to allocate participants into experimental or control groups, negating bias.
Self-selection: A bias introduced when participants choose their group placements affecting study outcomes.
Statistical Testing: Making Sure Conditions Don’t Differ by Chance
Calculation of likelihood showing whether random assignment in experiments has succeeded.
Results typically accepted only if there is less than a 5% chance (p < .05) that findings could be due to random assignment error.
The Real World: The Surprisingly High Likelihood of Unlikely Coincidences
Example regarding public beliefs in extrasensory perception (ESP) among college graduates.
Quotation by mathematician John Paulos highlighting misinterpretations of coincidence probabilities.
Drawing Conclusions
Internal Validity: The extent to which an experiment can establish causal relations.
External Validity: The degree to which operationally defined variables are reflective of the broader population context.
The Generalizability Restriction: “In the People We Studied . . .”
Case Method: Research technique focusing on a single individual for comprehensive analysis.
Random Sampling: A method ensuring equal chances of inclusion for population members in research samples.
Leads to representative samples allowing for generalizations.
Nonrandom Sampling: Acceptable when specific conditions are met, such as when similarity of sample to population is not critical.
Is Nonrandom Sampling a Fatal Flaw in Psychological Science?
Nonrandom sampling is not inherently problematic due to:
Representations potentially being irrelevant.
It can serve as a good base for insights even with limited generalizability.
The Reliability Restriction: “It Is Likely That…”
Replication: Repeating an experiment using the original procedures but with a new participant pool.
Type I Error: Incorrect conclusion suggesting a causal relationship exists when it does not.
Type II Error: Conclusion that indicates a non-existent causal relationship when it actually exists.
Other Voices - Psychology Is Not in Crisis
Examines the discourse around psychology facing research crises and replicability challenges.
Thinking Critically About Evidence
Critical Thinking: Involves questioning the integrity and full context of evidence presented.
Historical Insight: Sir Francis Bacon developed the scientific method but noted its limitations in practical application.
We See What We Expect and Want to See
Discusses cognitive biases that hinder critical thinking, stemming from evolutionary perspectives on perception standards in modern societies.
Hours Spent Partying or Studying at Some Canadian Universities
Analysis of a bar graph demonstrating weekly hours spent in social activities by students across various universities, questioning implications on reputational assessments without contextual inquiries.
The Ethics of Science: Doing What’s Right
Tri-Council Policy Statement (TCPS): Outlines foundational ethical principles in research involving human participants.
Respect for personal autonomy and decision-making
Concern for participant welfare, ensuring benefits outweigh risks
Justice: equitable distribution of research benefits and burdens.
Respecting People
Essential ethical guidelines governing psychological research include:
Informed consent
Freedom from coercion
Protection from harm
Risk-benefit analysis
Deception and De-briefing protocols
Upkeeping confidentiality
Respecting Animals
Guidelines from the Canadian Council on Animal Care (CCAC) for ethical use of animal subjects, featuring principles of Replacement, Reduction, and Refinement.
Respecting Truth
Emphasizes the integrity of scientific discourse in psychology regarding result transparency, ethical credit allocation, and data dissemination.