Research methods- final exam

All of the chapters

Uses the scientific method to find new knowledge

Researcher producer

Reads about knowledge found by research producers, often with the goal of applying it to make decisions or improve life

Research consumer

Using evidence from the senses (or instruments that assist the senses) as the basis for conclusions

Empricism

Scientists question everything, including their own theories and results

Skepticism

Scientists tend to avoid any biases that could influence the design, interpretation, or reporting of their research; not get too attached to it

Disinterestedness

The theory-data cycle

Theory, research question, research design, hypothesis, data, support or revision of research design or theory

change absolutely nothing about the study design, besides who the participants are

Exact/direct replication

Change other things about the study (stimuli or environment)

Conceptual replication

Good theories are ________; they can fail when tested

Falsifiable

Scientists never _____ theories wrong; they only find support for or against theories

Prove

Scientists often work within ____ and different teams communicate with each other

teams

conducted to increase knowledge about some topic

basic research

uses lessons from basic research to test applications of those lessons in the real world

translational research

conducted to solve a practical problem

applied research

scientists publish their research in _______-________ journals for other scientists to read

peer-reviewed

Journalists write about scientific resarch for the public, sometimes ________

incorrectly

Weaknesses of personal experience as a source of information

No comparison group and confounds

occur when two things are changing at the same time

confounds

Weaknesses of intuition as a source of information

Availability heuristic, confirmation bias, and bias blind spot

things that come to mind easily tend to guide our thinking

availability heuristic

looking only for evidence that supports our beliefs

confirmation bias

belief that we are immune to bias, even though we recognize other people are biased

bias blind spot

Weaknesses of trusting authorities as a source of information

not all authorities are equally trustworthy on a subject, experts make mistakes, some people claim to be experts when they aren’t

Findings do not explain all cases all of the time, they can explain some of the cases

probabilistic

new data, research questions being tested

empirical articles

can be written by anyone, reviewing past information, not new information

review articles

a lot of information in a quick way

chapters in edited books

written at the level that the general public could read, have references, not peer reviewed

trade books

something that varies, changes

variables

observed and recorded variable

measured variable

controlled by the researcher, has at least two levels/conditions

manipulated variable

only have one level that is the same for every person in the study

constant

construct, theory level, used by researchers when discussing theories, used by journalists in news reports

conceputal variable

data level, variable that is concrete, specific, used by researchers when discussing particular studies

operational variable

3 types of claims

frequency, association, causal

describe the rate, degree, or frequency of a single variable

frequency claim

what to look for to identify a frequency claim

focus on one variable, variable is measured, not manipulated, usually has statisics but not always

argue that two variables are associated with/related to one another

association claims

What to look for to identify an association claim

two variables, variables are measured

as one variable increases, the other increases

positive correlation

as one variable increases, the other decreases

negative correlation

levels of one variable are associated with all levels of the other variable

zero correlation

argue that the change in one variable is responsible for (causes) the change in another variable; the strongest type of claim

causal claim

What to look for to identify a causal claim

at least one manipulated variable and at least one measured variable, can wrtie using stronger language

3 requirements for a causal claim

covariance, temporal precedence, no other explanations

two variables must be related/correlated; the relationship cannot be zero

covariance

the causal (manipulated) variable must come first in time, the effect (measured) variable must come later in time

temporal precedence

accomplished through experimental control and randomly assigning participants to condition

no other explanations for the relationship

can’t tell which variable causes which; rules out temporal precedence

directionality problem

there could be a 3rd un measured variable that causes both measured variables

3rd variable problem

4 big validities

construct, statistical, internal, external

how well the variables in your study are measured or manipulated; important for frequency, association, and causal claims

construct validity

how well the statistical results support the claim, how precise, reasonable, and replicable a study’s statistical conclusions are; important for association, causal claims

statistical validity

how well a study was designed so that alternative explanations for the findings can be ruled out; important for causal claims

internal validity

how well the results of a study represent people and contexts besides the people and contexts within the study; important for frequency claims

external validity

best for finding psych-specific articles across a variety of journals, does not include every psych-related journal

PsychINFO

best for finding psych-specific articles in journals published by the APA

PsychARTICLES

actual validated instrument without the rest of the junk, can be helpful to find psychological measures, doesn’t find everything that you need

PsychTESTS

good for broadening search to other, related fields, might have free pdfs

Google Scholar

good to find work by a particular author, will have free pdfs

Author’s personal website

looking at reference section of your article to find related research published prior to your article

backward citation tracing

looking for articles published after your article that cited your article

forward citation tracing

Structure of an empirical article

Title, abstract, introduction, method, results, discussion, references

Core ethical principles from the Belmont Report

principle of respect for persons, principle of beneficence, principle of justice

core ethical principles from the APA

fidelity and responsibility, integrity

people should be free to decide whether to participate in research, vultnerable populations should receive special protection

principle of respect for persons

insinuation that people will be punished for not participating

coercion

offering an incentive too large to turn down

undue influence

legal caregiver can give consent because they are adults, for the welfare of the children

voluntary informed consent

change our language for their age of development to explain what’s going to happen

assent

prisoners, children people with intellectual or developmental disabilities

vulnerable populations

researchers must do their best to protect participants from harm; minimize these risks and provide benefits that will be equivalent; includes risk of personal information being revealed

principle of beneficence

no identifying information in data

anonymous studies

identifying info is in dataset but is protected

confidential studies

must be fair in terms of who can participate in the research and who can benefit from it; participants bear the burden of research risk, so their group should be able to benefit from the research

principle of justice

establish relationships of trust, set appropriate boundaries

fidelity and responsibility

strive to be accurate and honest as a researcher, teacher, or clinician

integrity

APA’s five general principles

IRB review, avoid/minimizing deception, debriefing, avoidance of research misconduct, avoidance of plagarism

researcher’s definition of the variable at the theoretical level

conceptual definition

researcher’s decision for how to measure or manipulate the variable

operational definition

people provide information on some aspect of themselves through a questionaire or interview

self-report measures

recordings by researchers of observable actions of physical traces of behavior

observational/behavioral measures

recordings of biological data

physiological measures

variables in which the levels are categories, names; often coded with numbers but are meaningless

categorical/nominal

variables in which the levels are meaningful numbers

continous/quantitative

variables are rank-ordered on some dimension, but the interval between rankings is not consistent

ordinal

numbers in which the intervals between numbers are equal, but there is no “true zero”

interval

numbers in which the intervals between numbers are equal and the value of ratio means “none”

ratio

reliability=

consistency

validity=

accuracy

each participant should get approximately the same score each time they are measured using the same instrument

test-retest reliability

multiple researchers obtain the same score when measuring the same participant

interrater reliability

a participant gives a consistent set of answers across a variety of items designed to measure the same construct; measurement or scale

internal reliability

subjective assessments of validity

face, content

empirical (objective) assessments of validity

criterion, convergent, discriminant

does the measure look, on it’s face, like a measure of the conceptual variable?

face validity

does the measure capture all parts of the conceptual variable?

content validity

is the measure associated with the behavior it should be associated with? is that association higher than the association between a different measure and the behavior?

criterion validity

does the measure correlate with other measures of the same or similar conceptual variables? (should)

convergent validity

does the measure correlate with the measures of different conceptual variables? (shouldn’t)

discriminant validity