Chapter 2 Psychology 1000
SCIENTIFIC PRINCIPLES IN PSYCHOLOGY:
-uses scientific method
-inquiry driven by curiosity, skepticism, and open-mindedness
-Darley and Latane: the presence of multiple bystanders causes diffusion of responsibility:
-each person feels less personal responsibility for intervening violent crime on Kitty
-Steps of Scientific Process: ****see Fig. 2.1 pg 39 as relates to Darley and Latane experiment****
1) Identify a Question of Interest:
-from media or peers or world, scientists find question that interests them
2) Gather Information and Form Hypothesis
-determine whether any pre-existing studies, theories and other info might help answer it
-then, form hypothesis: specific prediction about phenomenon often in the form: If...then
3) Test Hypothesis by Conducting Research
4) Analyze Data, Draw Tentative Conclusions, and Report Findings to Scientific Community
-publishing research is essential to scientific progress: allows scientists to learn about new ideas
and findings, evaluate, challenge or expand on it
5) Build a Body of Knowledge
-ask further questions, formulate & test new hypotheses
-enough evidence may lead to theory: set of broader formal statements explaining how/why
events are related: usually relating behaviour and its causes
Two Approaches to Understanding Behaviour
1) Hindsight (After-the-Fact Understanding)
-looking back on something and using common sense and folk knowledge to explain it
-problem: related past events can be explained in many different ways: how do we know which
is most accurate?
-ex. Darley and Latane diffusion of responsibility explanation was based on info of the
incident from the media
2) Understanding through Prediction, Control and Theory Building
-the more scientific approach
-theories are strongest test of scientific understanding: integrated network of predictions
-incorporates existing facts and observations within a single broad framework
-testable
-supported by research findings
-law of parsimony: if 2 theories can explain the same thing, simpler one is preferred
-never considered absolute truth
-prediction does not require understanding
Defining & Measuring Variables
-variable: any factor or characteristic that can vary ex. people’s sex, GPA
-operational definition: defines variable in terms of specific procedures used to produce/measure it
-changes abstract concepts to measurable and observable ones (ex. self esteem)
Measurement techniques: often combine many within single study
Self-Reports
-self-reports measures ask people to report on their own knowledge, beliefs, experience, behaviour
-interviews and questionnaires
-may be distorted by social desirability bias: tendency to respond in socially acceptable manner rather than
true feelings or behaviour
-Paulhus: minimize it by wording questions in a sneaky way & ensuring anonymity and confidentiality -Over-Claiming Questionnaire: measure degree of SDB: see if participants claim familiarity with
non-existent items
Reports by Others
-ex. questionnaires to the person’s spouse, teachers, employer about their behaviour
Measures of Overt Behaviour (directly observable behaviour)
-ex. reaction time: how rapidly they respond to stimulus
-often coding systems are developed to categorize behaviour ex. “assists child” “praises child” “criticizes child”
-must be used properly to get reliable measurements: consistent observations
-humans/animals may behave differently when they know they’re observed
-unobtrusive measures: record behaviour in way that keeps participants unaware that certain responses
are being measured
-can also use archival measures: records and documents that already exists (ex. school transcript)
Specialized Psychological Tests:
-personality tests: typically contains questions how a person typically feels/behaves
-performance tests:
ex. intelligence tests: assemble objects or solve arithmetic problems
ex. neuropsychological tests: help diagnose normal/abnormal brain functioning using mental and
physical tasks, ex. recalling list of words or manipulating objects
Physiological Measures:
-ex. heart rate, hormonal secretions, brain functioning
-problem: don’t always understand what they mean
METHODS OF RESEARCH:
Descriptive Research: Recording Events:
-descriptive research: seeks to identify how humans/animals behave, usually in natural setting, may reveal
cause-effect relations
Case Studies:
-case study: in-depth analysis of an individual, group, event
-discover principles of behaviour true for all general people/situations
-advantages:
-enables scientists to closely study rare phenomena
-challenge validity of a theory or belief
-source of new ideas/hypotheses for further investigation
-disadvantages:
-poor sources of cause-effect
-may not generalize to other people/situations
-generalization needs more case studies, more controlled methods, diff. cultural groups
-measurement bias/observer bias: observer not objective, worrisome in case studies
-failure to thrive example: lack of rapid infant weight gain neg. impacts later physical/mental growth
-mentally challenged mothers went through training by Surrey Place Centre Toronto for feeding and
nutrition: it worked
-but, some other change could have been responsible for this
Naturalistic Observation
-naturalistic observation: researcher observes behaviour as occurs in natural setting and avoids influencing it
-Jane Goodall: chimpanzees display behaviours, ex. using tools, formerly believed to only be human capability
-cannot make clear cause-effect conclusions, too many variables involved in natural setting
-observer bias possible
-presence of observer may change participant behaviour -try to disguise
-wait for habituation: delay data collection until participants get used it/ignore their presence
Survey Research:
-info obtained by questionnaires/interviews, strongest advantage: can closely portray whole population
-Terracciano: while cultural stereotype of national character may define national identity, does not reflect actual
assessed personality traits of members of the culture
-2 key concepts:
-population: all individuals whom we are interested in drawing conclusion
-sample: subset of individuals drawn from larger population of interest, must be representative
-representative sample: proportionally reflects important characteristic of population
-random sampling: every member of population has equal probability of being chosen for study
-stratified random sampling: first divide population into subgroups based on
characteristics (gender, culture, age)
-internet questionnaires: no method for random sampling, no control of data quality/honesty, but still
comparatively reliable
-disadvantages:
-no cause-effect conclusions can be made
-rely on self-report: can be distorted by SDB, interviewer bias, inaccurate perception of questions and
own behaviour
-sometimes by chance can be non-representative sample (<5%) causes faulty beliefs of generalization
Correlational Research: Measuring Associations between Events
-correlational research: studies association between naturally occurring events/variables
-measure, NOT manipulate
1) researcher measure variable X, ex. birth order
2) measures a second variable Y, ex. personality trait
3) statistically determines whether X and Y are related
-bidirectionality problem (two-way causality): does X influence Y, or Y influence X, or both
-spurious problem: non-genuine relationship between variables
-third variable problem: Z is responsible for what looks like relation between X and Y
-Therefore, CANNOT draw cause-effect relations (confounding of variables) major disadvantage
-Advantages: (mostly helps extend other research or pave a path for new research, not so great on own)
-helps test generalization of cause-effect relations found in laboratory
-find relations for further lab investigation
-practically and ethically, some questions cannot be research in experiments, only correlational
-allows to make predictions
Correlation Coefficient: (r)
-correlational coefficient: statistic indicates direction and strength of relation between variables
-(+r) positive correlation: higher scores on X associated with higher scores on Y
-(-r) negative correlation: higher scores on X associated with lower scores on Y
-ranges from: +1.00 to -1.00
-the closer to -1 or +1 the stronger the relation, 0 means no relation
-scatterplots: graphs that show correlation between 2 variables
Experiments: Examining Cause-Effect
-can establish cause-effect relations MANIPULATE, not just measure like other methods
1) researcher manipulates/controls one or more variables
2) researcher measures whether this manipulation influences other variables
3) researcher attempts to control extraneous factors that might influence outcome
-Approach:
-start out with equivalent groups of participants
-treat them equally in all respects except for variable of interest -isolate variable and manipulate it
-measure how groups respond
-independent variable: factor that experimenter manipulates/controls
-must have 2 levels: usually an experimental level and control level, or 2 experimental levels
-dependent variable: factor measured by experimenter & may be influenced by IV (can be many DVs per IV)
Experimental and Control Groups:
-experimental group: receives treatment or an active level of IV, often multiple
-control group: not exposed to treatment or receives 0-level of IV
-provides standard of behaviour to compare to
Two Ways to Design Experiment:
1) between groups/between subjects design: each group composed of a different set of participants
-to avoid un-equal groups: random assignment: each person has equal likelihood of being
assigned to either group: used to balance any differences
2) repeated measures/within subjects design: each participant exposed to all conditions of IV
-must make sure they don’t get fatigued, more confident, habituated, etc. for the second test
-counterbalancing: order of conditions varied so there’s no advantage of one condition
over the other (ex. half group drives w/o cellpho
More
Less