W1 ways of knowing Flashcards
(23 cards)
3 ways of knowing
- intuition (common sense)
- authority/experts/institutions
- science
1. way of knownig
Intuition
Intuitive decision making
choices based on instinct, gut feeling and past experiences
- fast
Wilson 1993
Intuitive vs deliberative
intuitive - fast, but faulty decision making
deliberative - thoughtful analytic
participants chose 1 art poster out of 5
measured - 3 weeks later found intuitive gut choice more satisfaction
Availability Heuristic
mental shortcuts that enable efficient intuitive decisions
- vivid, imaginable events appear more common eg shark attack
- negativity bias - recall of negative easier
media influence eg 9/11 more ppl travel by car
Representativeness heuristic
ppl estimate probability of an event by how similar to a known situation/stereotype
eg. Linda studies social issues, is she bank teller or bank + feminist?
Confirmation bias
people tend to search out/recall info that confirms existing beliefs
eg. Watsons 2-4-6 task
seek confirming evidence of increase by 2, rather than falsifying it (any increases 10-20-30)
2 way of knowing
Authority/expert/institution
expert opinion changes - eg Fauci says masks ineffective for Covid, then later recommends them
AI - people have different views
trust in US gov has decreased overtime
3 ways of knowing
The scientific method
Observation - notice and define problem/question
Hypothesis - formulate testable prediction
Experiment - design one to test
Analysis - analyse data, statistics objective
C - does data support hypothesis - refine, replicate
Reproducibility project
Replication crisis
attempted to replicate 100 experimental+correlational studies
97% of original studies said significant results p<.05
BUT only 36% relplications had statistical significance
mean effect size in replications was half of original effect
CRISIS = original study significant, but replication didnt find any
Critique of replication ‘crisis’
p-value = probability of an observed result assuming null hypothesis is true
p=<.05
higher = more frequent the data??
if 100 ppl throw basketball again, wont get 10 in a row again
Out of 100 candidates, 5 get 10 Bball throws in a row
will they get 10 in a row again?
NO
due to regression to the mean
- extreme values tend to move closer to the mean (average) when measured again
eg. really tall mother have a child that is extra taller? = unlikely
Replication science is to
determine real effect size - correctional
Replication crisis impact
pre-registration - share research plan
open data/materials
improved practices - larger sample, statistical power
publication of null results
replication science = large scale collaboration
MORE TESTS! MORE COLLABORATE! MORE REPLICATE!
goals of psychological science
- describe behaviour
- predict future behaviour
- determine if causal
- mechanisms of causality
- Describe behaviour
overtime at population level
eg. 1848 Phineas suffered brain injury to skull
- following surgery, his personality changed –> frontal lobe for personality and decisions
2 - predict future behaviours
by looking at co-occurring relationships
eg. socio-economic status + life expectancy
- gap between riches 1% of popn and poorest in 14.6 years m, 10.1 f
3 - determine if relationships are causal
and example
Correlation is NOT causation
eg. self esteem + academic achievement
- one may predict the other
- other variables too – intelligence, social status
Requirements for causality
- theory is a significant correlation
- A) temporality (time) where a precedes b
- B) theoretically justified
if only B theory - it is only correlation, cannot say a influences B
4 - mechanisms underlying causal relationships
causal explanations 💪🏼 by 👀 causal mechanisms
correlation - temporality - mechanism
eg. hypothesise that exersise causes thirst, feel thirsty causes person to drink water
c = exercise and drinking
t = exercise before drinking
m = cause (exercising) linked to effect (water) via thirst
Theories
what is + steps
= a proposed explanation whose status is conjectural and subject experimentation
= an idea
steps
- identify question of interest
- hypothesis
- test hypothesis
- analyse data, findings, collect and develop more
identify question of interest
- testable predictions and hypothesis
- subjective observations
eg. phrenology brain sizes
form hypothesis
eg. diffusion of responsibility (bystander effect)
- all watching, no one intervenes
- awareness and consequence
test hypothesis
independent variable - predictor
dependent variable - outcome
eg. smoke in room
