Psych 303 Exam 1
informed consent
part of autonomy learn about the project, know the risks and benefits, and decide if they want to participate
snowball sampling
participants recommend a few acquaintances for the study
zero association
two variables are not related
deception
researchers giving false information to participants
split half reliability
test if one half of the test correlates with the other half
content validity
the measure spans the universe of all content of a construct
translational research
the use of lessons from basic research to develop and test applications to healthcare, psychotherapy, and other forms of treatment and intervention bridge from basic research to applied research
availability bias
things that come to mind easily guide our thinking
negatively worded questions
unnecessarily complicated can cause confusion
empiricism
use evidence from the sense or from instruments that assist the senses as the basis for conclusions
known groups/concurrent validity
use the measure to differentiate between groups and behaviors
quantitative variable
variable that is coded with meaningful numbers
correlational design
variables are measured, not manipulated efficient way to get lots of data higher external validity observe relationships and make predictions
motivational bias
want to think what we want examples: present/present bias, confirmatory hypothesis testing
hypothesis
way of stating the specific outcome the researcher expects to observe if the theory is accurate
relationship between reliability and validity
we can have reliability without validity, but we cannot have validity without reliability
present/present bias
we fail to look for absences and only notice what is present
double-barreled question
asks two questions in one
cluster sampling
clusters of participants within a population are randomly selected an all individuals in the cluster are used
theory data cycle
collect data to test, change, or update theories
institutional review board
committee responsible for interpreting ethical principles and ensure research with human participants is conducted ethically
comparison group
compare what would happen both with and without the thing we are interested in used in systematic data collection
things to watch when writing questions
complexity double-barrleed loaded negative wording response sets
operationalization
concrete definitions of variables that can be manipulated/measured
internal consistent reliability
consistent pattern of answers no matter how question is phrased measured with cronbach's alpha and split-half reliability
journal
contains research papers that have been peer reviewed different from journalism
cronbach's alpha
correlational-based statistic shows if scales have internal reliability closer the number is to 1, the better the internal reliability
reliability
how consistent the results of a measure are 3 types: test-retest, inter-rater, internal consistent
positive association
if one variable increases, so does the other
negative association
if one variable increases, the other decreases
convergent validity
if the measure correlates strongly with other measures of the same construct
discriminant validity
if the measure correlates with measures of different constructs
order of questions
important questions first simple questions first, sensitive questions later with multiple topics, use logical order background/demographic questions last
case study
in depth study of one case
cognitive bias
influenced by a story that makes sense or what comes to mind easily example: availability bias
causal claim
one variable is responsible for change the other must have correlation, temporal precedence, and high internal validity
response set
nondifferentiation type of shortcut respondents can take consistent way of answering all questions
naturalistic observation
observe behaviors in real world settings
longitudinal research
observe same subjects over a long period of time
purposive sampling
only want to study certain kinds of people so recruit in a non-random way
categorical variable
operational variable that is a category
two types of biases
1. cognitive biases 2. motivational biases
3 conditions for causal relationships
1. correlation/covariation 2. temporal precedence 3. elimination of confounds
4 ethical classifications
1. exempt (no risk) 2. minimal risk (chance of harm) 3. greater than minimal risk 4. not reserach
features of good scientific theories
1. supported by data 2. falsifiable 3. parsimonious
theory data cycle steps
1. theory 2. research questions 3. research design 4. hypothesis 5. data 6. data either supports theory, or leads to revisions in theory and/or research design
manipulated variable
a researcher controls its levels, usually by assigning participants to different levels
commission
actively lying
construct validity
adequacy of operational definitions does it measure what we want it to measure
parsimony
all other things equal, the simplest solution is the best
confirmatory hypothesis testing
also called positive test bias select questions that lead to a particular, expected answer not scientific
confound
alternate explanation occurs when you think one thing caused an outcome, but other things changed as well so you don't really know what caused the outcome hard to isolate in personal expieriences, can control variables in a research setting
dependent variable
always measured predicted/affected by the independent variable
experimental design
an independent variable is manipulated only way to establish causality other confounding variables are controlled with experimental control and random assignment
convenience sample
biased using sample or people who are readily available to participate
random assignment
can eliminate alternative explanations increases internal validity
conceptual definition
definition of a construct at theoretical level
research design
depends on the nature and the number of variables involved types: experimental, correlational
frequency claim
describes rate or degree of a single variable
survey/poll
descriptive observe and describe attitudes, opinions and behaviors
problems with interviews
discomfort if sensitive topics interviewer bias
Belmont report
done because of tuskegee and milgram study laid out basics for doing research 1. beneficience 2. justice 3. autonomy
applied research
done with a practical problem in mind want to directly apply findings to the solution of that problem in a particular real-world context
enticing people to do surveys
easy to do not too long small rewards anonymity/confidentiality
interval scale
equal distances between levels no true 0 (cannot get it)
ratio scale
equal distances between levels true 0 (0 is meaningful)
historical/archival
examine existing data to test hypotheses
mediating variable
explains association between two variables needs to be shown statistically
internal validity
extend to which our finds provide evidence of causality 3 conditions
external validity
extend to which research claims generalize to other samples, situations, and settings
face validity
extent to which it appears to experts that an operational definition is a possible measure of the variable
scientific fraud
fabricating data
omission
failing to mention certain details
principle of justice
fair balance between types of people who participate and the kinds of people who benefit from it
inter-rater reliability
get the same data regardless of who measures
test-retest reliability
get the same result every time the measure is used
basic research
goal is to enhance the general body of knowledge may be applied to real-world problems later on
measured variable
levels are observed and recorded
problems with correlational studies
little control need larger samples because harder to find effects cannot establish causality with certainty (but can rule out causality if no correlation is found)
panel study
longitudinal research design same people participate for several years
problems with experiments
low external validity some variables cannot be manipulated
independent variable
manipulated in experiments, measured in non-experimental designs predicts/affects dependent variable
criteron-related validity
measures validity of construct three types: known groups/concurrent, convergent, discriminant
autonomy
principle of respect for persons individuals are free to make up their mind if they want to participate some people have less autonomy so are entitled to special protection can withdraw at anytime without penalty don't have to answer all questions if they don't want to
principle of beneficience
protect participants from harm and ensure their well-being
difference between journalism and journal
public reads journalism scientists read journals journalists may publish research, beneficial if the story is important and accurate
simple random sampling
raffle anyone could get selected
multistage sampling
random clusters are selected then random samples are taken from those clusters
probability sampling
random sampling everyone has an equal chance of being selected
ordinal scale
ranked order
plagiarism
representing the ideas or words of others as ones own
socially desirable responding
respondents give answers that make them look better
problems with questionnaires
response rates and missing data misrepresentation of participants misunderstanding
alternatives to deception
role playing simulation studies honest experiments
association claim
says that one level of a variable is likely to be associated with a particular level of another variable correlates/covaries/relates two or more variables, measured not manipulated
stratified random sampling
selects particular demographic categories on purpose then randomly selects individuals within each category
theory
set of statements that describes general principles about how variables relate to one another
falsifiable
some hypotheses that when tested may fail to support the theory
study validity
success with which a study has demonstrated a relationship two types: internal and external
debrief
telling participants about the study's hypotheses after the study deception revealed communicate purpose and importance return participants to original state of mind