ERMS Flashcards
Methods of Knowing
which ones are scientific and which are not?
Non-scientific
1. Intuition
2. Authority
3. rationalism
4. empiricism
Scientific
5. The scientific method
Intuition
limitations
Relying on guts, emotions, & instincts to guide us
- intuition can be wrong
Authority
Limitations
Accepting new ideas becuase some authority figure states they are true
- not always trustworthy
Rationalism
limitations
using logic and reasoing to acquire new knowledge
if premises are wrong/errored then the conclusion will not be valid
empricalism
limitations
acquiring knowledge through obzervation ad experience
can be deceiving, cant or dont want to experience evrything
The scientific Method
limitations
systemically collecting and evaluating evidence to test ideas and answer questions
Capital T truth
truth that is the same for everybody
Features of science (3)
- Systematic empiricism
- empirical questions
- public Knowledge
systematic empricism
carefully planning, making, recording, and analyzing multiple obzervations
Empirical questions
questions asked about the facts of the world which can be answered by systemic empiricism
- there is a “true” answer
- not about values, morals, “oughts”
Public Communication
Scientific findings must be openly and truthfully communicated, put in context of other research, and critiqued.
Psudo science has
- anecdotes
- evidence =proof
- confirmation bias
- hand-pick evidence
- no peer review
- overstates findings q
Basic Research
attempts to answers fundamental questions about human behvior
applied research
attempts to apply theory to solve practical problems and develop potential solutions
research questions
questions about the research
Peer Review
process of scientists reading, critiqueing other work prior to publication to make sure it fits the neccasary rigor of science/is done correctly.
allows science to self-correct & build off of eachother
hypothesis
specific falsifiable prediction of how you expect a research question should be answered
operationalization
turning something thats a construct and making it measurable and testable
Null hypothesis
H0: no relationship
why is converging diverse evidence so strong
replication and conceptual replication
what is research strategy
general approad and goals for a study
- experimental
- quasi
- non-exp
research design
design of particular study
- group vs indiv.
- same indiv. vs diff indiv.
- lab vs naturalistic
- # of variabbles and types
- measurements and protocols
relationships can be
- general
- linear
- curvilinear
- positive
- negative
experiment
answers cause-and-effect questions about the relationship between two variables
quasi-experiment
atttempt to produce a semi cause-and-effect explanation, using a variables that can’t be manipulated
non-experiment
demonstrates relationship between variables =- does not attempt to explain it
within-in subject variables
participants experience all conditions or gtouping for that variable
between subject variables
participants are assigned to different conditions
how do we use manipulation
research manipulates one variable by changing/assigning tis calue to create 2+ conditions
experiment features
- manipulation
- conditions
- measurment
- control
- comparison
conditions
experiment levels to independent variable
- if I am studying the effects of 2 different pain medications of headaches, I may give people who have headaches either Tylenol or Bayer (the treatment groups; thus in this example there are two levels of the IV; one level is Tylenol, the other is Bayer, but both are part of the experimental condition).
measurement
a second vairbale is measured, resulting in a set of scores for every condition
control
all other varibales are controlled to be sure that they do not influence the variables being examined q
comparison
the scores inone condition are compared with the scores in another condition
independent vairbale (IV)
the variable that is manipulated by the researcher
dependent variable
the variable that is obzerved by the researcher
extraneous variables
all other varibales in the study other than IV and DV. can be participant variables, or eenviroment variable
steps to establish causation
- temporal precedent: demonstrate that the “cause” happens before the “effect” occures. (iv must happen firt, change in DV must follow)
- control: Demonstrate that one specific variables is responisible for changes in the other
Types of manipulation
- experimental condition
- control condition
- no-treatment control condition
- placebo control condition
- placebo affect
- waitlist control condition
experimental condition
treatment condition whichc receives an experimental manipulation
.
.
no-treatment control condition
condition which does not receive any manipulation; answers the measurement (DV) but nothign else
placebo control condition
condition which receives an ineffective treagtment that is equivalent to the experiment condition in all ways except expected effectivness
placebo effect
effect occurs simply due to thinking they’re receiving treatment
wait-list control condition
control condition which recieves nothinhg for the duration of the experiment but receives treatment after
Controlling nature
an experiment must control all aspects of the world to establihs a cause and effect relationship
confounding variable
a specific type of extraneous variable that changes systematically along with the other 2 variables being studied. 3rd variable that influences both the IV and DV
ex: CV: hot temp -> IV: Ice cream -> DV: Sunburn
enviroments as confounds
location, tempature, time, procedure, order of tasks, experimenter
individuals as confounds
race, age, income, personality, sexual orientation, intelliegence
how are extranious variables controlled
randomization: use a random prcoess to help avoid systemmatic variation with the IV
Hold the variable constant” keep the extranious variable in the same in each condition
Procedure manipulation
participants are asked to do something behavioral; conditions are based on doing different things
stimuli manipulation
participants are exposed to something; conditions are based on serving different things
nominal ( or named )
numbers represent named differences, not value differences
- number have no qualitative value or intrinsic order
vignette (aka scenerio)
a short description of an event, person, etc. that you want a participant to evaluate and respond to
instruction manipulation
particpants an experiement might experience different instructions based on conditions
manipulation check
making sure manipulation happened
Scales of measurement
se4t of categories used for classification for varibales
categorial variables
conditions differ by type
nominal ( or named )
numbers represent named differences, not value differences
- number have no qualitative value or intrinsic order
ordinal (ranked)
number represent differences in rank or order
continious variables
a variable that can be an infinite or uncountable set of variables (height, weight, tempature)
- Interval and ratio variables
interval or integer
number represent measurement along a scale when each unit is the same
ratio
quantitative data, measure varibales on continious scale, ratio has “true 0”
example: money
likert-type scale
scale from 1-5
how do different scales compare
1.nominal: reveals whether difference exists
2. ordinal: indicates the direction of the diffference (which is more and which is less)
3. interval: determines rhe direction and the magnitiude of a difference
4. ratio: determines rhe direction, magnitude, and ratio of a difference
ways to measure variables
- self-report
- bevbahiors
self-report
- what are the porblems witht his
- ask individuals and they answer about themselves
1. social desirbality
2. baises
3. demand characterisitics
4. retrepoective bias `
behvaiors
how do we measure them?
actual actions
1. behavioral obzervation: what the researcher sees happen
2. behavioral trace: evidence of behavior left behind
3. behavioral choices: behavior indicates a perference
different types of Measurment error
random error: variation form true score due to chance
bias: systematic variation from true score in a consistent way
reliabilty
stability or ocnsistency of the measurements produced by a specific measure
ex, if u score 95 and then score 96
validity
accuracy of measurement produced by a specific measurement
what are parts of Surveys
- carefully worded instructions
openened, closed ended, text options, numberical options
population
set of people your interested in
sample
subset of people youre interesred in
representstive sample
matches people, charactorisits, places, etc your interestred in
rnadom sample
everyone in pop nhas the same chacnce of being in your study
convienience sample
sample selected conviently
Sampling issues
WEIRD: western, education, industrialized, rich, democratic
sampling size: larger = more representative
smale -> less
non response bias
people who dont participate in the study may be differnet in a meaningful way than those who do
selection bias
people who volunteer may be differnt than people who dont
attribution bias
people who dont fisn might be differnt than rthose who do in a meaningful way
score
measurement for each individual (referred to as x ((sometimes y ))
N =
number of poeple in population
n =
number of peopole in sample
frequency
number of individuals in each score
distrubution
the way scores are distributed acroess the levels of a variable
normal distrubution
symmetrical distrubution with greatest frewuency in middle
temporal precedent
demonstrate that the “cause” happens before the “effect” occures. (iv must happen firt, change in DV must follow)
chicken or eggs example
Folk psychology
peoples intutition about human behavior
critical thinking attitudes
skepticism and tolerance for uncertainty
psychology research cyclical model
research question about research literature leads to empirical study, the results are published and become research literature.
developing research question
- is it interesting?
- does a fill a gap?
- does it have important practical application?
what method do psychologists use
hypothetico-deductive method
Inferential statistics
allows researchers to decide if their findings are statistically significant
type 1 and type two errors in knowing if our conclusions are correct
type 1: concluding an effect is real when it is not
type 2: concluding there is no effect
when there actually is a real effect in the population
four moral principles
- weighing risks against benefits
- acting responsibly and with integrity
- seeking justice
- respecting people’s rights and dignity
experiments and tudies are high in______
internal validitiy
what are the two tyes of non-exp research
Correlational research:
statistical relationships between variables that are measured but not manipulated
observational research: participants are observed and their behavior is recorded without the researcher
interfering or manipulating any variables.
what is the internal validity of research method types?
experimental: high internal validity
Quai-exp: in between
correlational: Low internal validity
what does correlation research do if not establish causation?
establishing
reliability and validity, providing converging evidence, describing relationships, and making predictions
what is the purpose of comples correlational research
explore possible causal relationships among variables
using techniques such as partial correlation and multiple regression
aproaches to obzervational research
participant observation, structured observation, case studies, and archival research.
