Principles of Psych Final

4 main sources of information

intuition
personal experience
authority
empiricism

intuition

trying to support a hypothesis because it "just makes sense"

availability heuristic

basing our decision on the number of examples that come to mind

confirmation bias

the tendency to look only at information that agrees with what we already believe

bias blind spot

the belief that biases do not apply to us

personal experience

trying to support a hypothesis using someones personal experience

authority

trying to support a hypothesis by relying on information from trusted authority figures

empiricism

trying to support a hypothesis by objectively collecting and analyzing data

empiricism

process of using evidence from the senses or from instruments as the basis for conclusions

quantifiable

able to be expressed or measured as a quantity

absolute deprivation theory

feelings of deprivation are determined by the absolute or objective amount one has/does not have

relative deprivation theory

feelings of deprivation are determined by the absolute or objective amount one has/does not have relative to others

what makes a good theory?

1. good theories are falsifiable
2. good theories are supported by data
3. good theories are supported not proven
4. good theories have parsimony

occam's razor

"plurality must never be posited without necessity"

scientific publication

make findings public to scientific community

to be a variable

it must have at least two levels

conceptual definition

a researchers definition of the variable in question at a theoretical level

operational definition

specifications for turning a conceptual variable into a concrete measured (or manipulated) outcome

self-report

operationalizes a variable by recording people's answer to questions about themselves in a questionnaire or interview

reverse-coding

wording a self-report question in the opposite direction the researcher will later "reverse code" this item to ensure that items are averaged together in the right direction

physiological measures

operationalizes a variable by recording biological data

behavioral/observational measures

operationalizes a variable by recording observable behaviors or traces of behaviors

measured variables

research simply observes and records

manipulated variable

researchers controls the variable by assigning people to different levels

quantitative variables

coded with meaningful numbers

nominal scale

label variable without quantitative values instead variables are divided into categories

ordinal scale

there is an "orderedness" to the data HOWEVER the distance each unit is not necessarily the same

interval scale

there is "orderedness" to the data AND the distance between each unit is now the same/equal
-no true zero point

ratio scale

there is "orderedness" to the data, the distance between each unit is now the same/equal, AND zero actually means "none"

frequency claims

describe a particular rate or degree of a single variable
-usually supported with a survey/poll

association claim

argues that one level of a variable is likely to be associated with a particular level of another variable
-usually supported with a correlation study

causal claims

goes further than a mere association - says that variables are related *because* one variable is responsible for changing the other
-must be supported by an experiment

verbs that imply causation

-causes
-affects
-changes
-makes
-may lead to
-lowers
-increases

validity

appropriateness of a conclusion or decision; a claim that is reasonable, accurate and justifiable

external validity

how well do the study's claims generalize to populations/contexts/situations besides those in the original study

construct validity

how well a variable is measured in a study

statistical conclusion validity

extent to which a study's statistical conclusions are reasonable and accurate

random (chance) variation

two scores are different because of random factors

meaningful variation

a "real" difference or "real" effect i.e., an effect that will repeat even when the other random factors do not

p-value

a lower p-value means a lower probability that results are "just chance"

statistical significance

the point at which we determine a p-value is low enough to conclude there is a "real" effect
-traditional cut-off is p

effect size

estimates size of the difference between two group means

confidence interval

a range of scores designed to include the true population value a high proportion of the time

internal validity

the extent to a causal claim is valid/justifiable

evidence needed to support a causal claim

1. temporal precedence- i.e. causes precede (come before) effects
2. covariance of X&Y - do X&Y move together?
3. internal validity/ruling out alternative explanations- is a change in X the only explanation for a corresponding change in Y? Or could a change in some other variable be responsible for effecting our outcome?

making a causal claim

-manipulating variables is the only way to ensure that these 3 criteria for causation are met
-experiments can support a causal claim because they manipulate the independent variable
-correlation studies CANNOT support a casual claim, because in a correlation design both variables are measured

reliability

does your measure give consistent results?

validity

does your measure assess what it says it does?

test-retest reliability

study participants get pretty much the same score each time they are measured

internal reliability

participants to give consistent responses, no matter how the question is phrased

Reliability

how consistent is your measure? does it give the consistent scores across time/people?

inter-rater reliability

are observers ratings consistent? applicable when we have multiple observers rating/coding participant responses and/or behavior

face validity

your measure appears "on its face" to measure what it says it does. in other words, it just "looks like" what you want to measure

content validity

measuring all parts of a defined construct

criterion validity

when a scale correlates with a concrete outcome or behavior

known-groups paradigm

researchers see whether scores on the measure can discriminate among two or more groups whose behavior has already been confirmed.

convergent validity

your self-report measure is more strongly associated with self-report measures of similar constructs

discriminant validity

when a scale doesn't correlate very highly with other self-report measures of a different construct

population

the entire set of people in which you are interested

census

getting responses from every single person in the population of intrest

sample

a smaller set of individuals taken from the population

bias

consistent leaning one direction vs. the other

biased sample

some members of the population of interest have a much higher probability of being selected into the sample

convience sampling

researcher selects only those who are easy to contact

self-selection

sample contains only those who initially volunteer to participate

non-response bias

concern if the people who chose to respond are fundamentally different than those who do not chose to respond

representative (unbiased) sample

every person in the population of interest has an equal chance of being selected

probability sampling

every member of the population of interest has an equal and known chance of being selected for the sample regardless of whether they are convenient or motivated to volunteer

simple random sample

simply a completely random set of individuals from a population

cluster sampling

-instead of randomly selecting individuals the researchers randomly selects clusters of people
-list of clusters are obtained and researcher randomly selects a sample of these clusters
-an option when people are already divided into arbitrary groups

stratified random sampling

researcher purposefully selects meaningful categories and randomly selects individuals within each categories proportionate to their assumed membership in the population

oversampling

-variation of stratified random sampling
-used when we are interested in groups that have a very low base rate in the population
-researcher purposefully oversamples these groups and then uses weights to adjust to actual percentage in population

quota sample

-similar to stratified random sampling in that the researcher selects demographic categories
-participants are not randomly selected

snowball sampling

after being recruited a participant recommends other potential participants

larger sample sizes

-yields a smaller confidence interval/margin of error
-are better but are not automatically trustworthy

frequency claim

describe a particular rate or degree of a single variable

survey/poll

a method of posing questions to people, online, in personal interviews, or in written questionnaires

observational study

when a researcher watches people or animals and systematically records how they behave or what they are doing

types of questions

-likert-type scale
-semantic differential
-forced choice
-open ended

likert-type scale

people rate their agreement with a statement on a scale with anchor points from "strongly disagree" to "strongly agree"

semantic differential

rate a target statement using a numeric scale that is anchored with adjectives

forced-choice

people give their opinion by picking the best of two or more options

open-ended

allows respondents to reply any way they want

framing effect

when our decisions are influenced by the way information is presented

loaded-questions

when a question makes an assumption that isn't warrented

double-barreled questions

asking two questions in one

leading questions

when a questions wording leads people to a particular response

push poll

a marketing technique that is designed to manipulate/change peoples opinions

self-report

operationalizes a variable by recording peoples answer to questions about themselves in a questionnaire or interview

socially desirable responding

if participants are embarrassed, shy, or worried about giving an unpopular opinion, they will not tell the truth

response set

when respondents adopt a consistent way of answering all the questions

acquiescence response set

answering "yes" or "strongly agree" to all questions

fence sitting response set

playing it safe by answering in the middle of the scale

naturalistic observation

naturally observing behavior in the field (vs. laboratory)

field study

a study conducted "in the field" vs is the lab

structured observation

researchers observe specific behaviors in a more structured setting

participant observation

researchers become active participants in the group or situation they are studying

ecological validity

extent to which a study finding generalizes to the "real world"

observer bias

observers expectations influence their interpretation of the participants behavior or the outcome of the study

observer effects

when observer expectations change participants behavior/responses