Research Methods and Stats Final

intuition

knowledge based on "hunches" and what feels right

limitations of intuition

cognitive and motivational biases affect perception
illusory correlation

illusory correlation

tendency to perceive relationships where there is none

authority

acceptance of knowledge because it is acquired from highly respected sources

limitations of authority

information may be inaccurate
information may be biased

logic and reasoning

knowledge based on logical reasoning (or inferring conclusions that are implied by the premises and the relations between premise)

limitations of logic and reasoning

the conclusion of a logical argument is limited by the premise making up the argument

scientific approach =

empirical

empirical

to make observations

scientific method

generate a problem
make an observation
ask a question
search prior literature
form a hypothesis
conduct an experiment
collect data
draw conclusions

ideas have to be ______ and _______

testable and faslibiable

peer review

expert scientists carefully evaluate research

serotonin hypothesis

no evidence that serotonin inbalance is the causal explanation for depression (lacks validity)

characteristics of pseduoscience

anecdotal evidence
appeals to authorities
claims are vague
uses scientific sounding terminology

pseudosciene

tries to intentionally pass off something as scientific when it is not

population

entire group that a researcher wishes to study

sample

set of individuals selected from a population
intended to be selective of the population

parameter

characteristics that describes a population

statsistics

characteristics that describe a sample

variable

a factor or element one expects to change, vary, or have several different vaules

well define variables =

height, weight, eye color

psychological constructs =

aggression, depression, intelligence, memory, political attitudes

conceptual definition

defining variables in theoretical terms

operational definition

how the researcher uses the variables in the study

descriptive research (describing variables)

examines individual variables and no relationship between variables

correlational research (relationships between variables)

examines relationships between variables by measuring two or more variables for each participant

experimental and non-experimental research (comparing groups)

examines relationship between 2 or more scores

direct replication

copy the original study

systematic replication

very similar to the original study but with some version

conceptual replication

based on the idea of the original study but very different from it, usually with significant changes to measures or manipulations

replication crisis

50% of replication of famous psych studies do not match

theory

systematic body of ideas about a particular phenomenon

research question

identifies and describes the broad topic of investigation

hypothesis

testable prediction about the relationship between two or more variables

null hypothesis

no relationship between variables

good hypothesis

logical, specific, falsifiable, supportable

Barnum effect

observations that individuals will give high accuracy ratings to descriptions of their personality that supposedly are tailored specifically for them, yet are very vague and general enough to apply to a larger pouplation

falsifiability (refutable)

when you can imagine an observation that would cause you to reject the hypothesis

supportable

predict differences rather than no differences

research ethics

refers to the behavior of the researchers

researchers are responsible for

protecting participants, research assistants, and readers of the reports

scientific integrity

commitment to intellectual honesty and adherence to ethical principles in scientific research

benefience

maximize benefits and minimize harm
need to do a risk-benefit analysis and justify any remaining risk

physical harm

medical procedures, sleep deprivation, studies alcohol/caffeine

psychological harm

asking about traumatic events, anxiety, giving unfavorable feedback

breach of privacy/confidentially

bathroom/internt privacy

anonymity is _______ to protect privacy

best

informed consent

treat individuals as autonomous

threats to autonomy

coercion, deception (counteract with debriefing)

justice

participants were treated impartially without bias on gender, wealth, sexuality etc.

instituional review board

responsible for the review of research conducted within the institution (they decide if the benefits outweigh the risks)

mediators

variable that proposes the mechanism that comes between a cause and its effect
(possible explanation for relationship between X and Y)

mediators do what?

explain the relationship between 2 variables (causal)

moderators

variables used to determine when ----> in what situation and for whom the cause has effected
(affects the magnitude of the effect of X ad Y)

moderators do what?

affects the magnitude of effect of x and y

categorical variables

takes a finite number of categories or distinct group (discrete)
(typically qualitative)

examples of categorical variables

race, political party

continuous variables

take on an infinite number of variables between any 2 values
(typically numerical)

examples of continuous variables

age, income

nominal scale

levels represent different categories or groups

examples of nominal scale

pet type, color, fruit

ordinal scale

adding order to a nominal scale
categories order in meaningful scale

examples of ordinal scale

the hotness of peppers, class grade, letter grade in glass

interval scale

difference between the numbers on the scale is meaningful and equal in size
no absolute zero (no absence of measure)

examples of interval scale

temperature (C or F)
time of day

ratio scale

equal and meaningful differences between numbers on the scale
absolute zero (represents absence of measure)

examples of ratio scale

weight, reaction time, length

Likert scale controversy =

ordinal scale

measurement reliability

degree to which a measurement is consistnet

less measurement error =

more reliable the measure

systematic error (bias)

varies systematically or consistently pushes scores higher or lower in a way that leads to inaccurate or misleading conclusions

random error

variation from one's true score due to unsystematic or chance factors that occur in a haphazard, disorganized, or arbitrary way

examples of systematic error (bias)

observer bias, bias in administering the measure, participant bias (social desirability and interpretation of the study = demand characteristics)

examples of random error

forgetting to start the timer on one trial, entering a measuring incorrectly, fire alar going off during an experimental session, participant misreading a question and answering incorrectly

test-resets

same test is taken twice
assesses overall effects of random error

alternative form

two different form of same test created and used
assesses overall effects of random error

test-retest coeffcient

the percentage of variation is score that is NOT due to random error

split-half reliability

correlation between total score one half of the test with total score on another half of the test
assesses amount of random error due to the participant

cronbach's alpha

average of all possible spilt-half reliability coefficients
assesses amount of random error due to the participant

internal consistency reliability assesses

amount of random error due to the participant

interrater reliability

extent to which rater agree in their observations
assess amount of random error due to the observer/scoer

interrater realibilty assesses

amount of random error due to the observer/score

standardization

treat each participant in the dame way

reliability is a _________ for validity

prerequiste

construct validity

degree to which a measurement device accurately measures the theoretical construct it is designed to measure

face validity

the extent to which the measure appears to measure what its supposed to
"Does the scale appear to be measuring the variable?"

content validity

extent to which a measure represents all facets of a given construct
"Do the items on the scales represent the various aspects of the variables being measured?"

convergent validity

the extent to which scores on the measure in question are related to scores on other measures of the same construct or similar constructs
"Does the scale relate to other measures of similar constructs?"

discriminant validity

the extent to which measurements that are not supposed to be related are, in fact, unrelated
"Does the scale relate to measures of unrelated variables?"

predictive validity

extent to which a score on a test predicts scores in some criterion measure
"Does the scale relate to relevant outcome or behavior that occurs in the future, after the scale is completed?"

concurrent validity

people in groups known to differ on a construct should differently on the measure
"Does the scale differentiate in a manner that helps to predict a relevant outcome or behavior currently?"

sensitivty

ability to detect differences among participants on a given variable

sensitivity measures should be

valid, reliable, provide a variety of scores

correlational hypothesis

there is some relationship between x and y
there is a positive relationship between x and y. As x is increasing, so is y.
There is a negative relationship between x and y. As x increases, y decreases.
Null: There is no relationship between x and y.

experimental hypothesis

x and y will differ on _________
x will be more than y on _________
x will be less than y on _______
Null: x and y will not differ on _______

pearson =

r value
numerical index of the strength of the relationship between two variables

-1 to 1
closer to 1/-1 = stronger relationship
closer to 0 = weaker relationship

sig (2 tail) =

p value

risk-benefit analysis

some greater good must come out of psychologists using human subjects (good for subjects and mankind)

Three Types of Moderators

1. Weakens/Eliminates X-Y relationship
2. Strengthen X-Y relationship
3. Reverse X-Y relationship

scale of measurement

a classification system used to measure a variable