research methods Flashcards
types of data
qualitative: experiences, detailed
quantitative: comparable, generalizable, simple data
independent variable
what we manipulate/change, different conditions
dependent variable
what we are measuring
true experiments (lab)
one IV and one DV, goal is to establish a cause-and-effect relationship between two variables, experiments are performed under highly controlled conditions
a research hypothesis
a clear, precise, testable statement that is written at the beginning of an investigation. it states the relationship between the variables being investigated
types of hypotheses
- null
- experimental/alternative: direction and non-directional
null hypothesis
IV will have no effect on the DV, any change will be due to chance
researcher’s goal is to reject the null hypothesis
directional hypothesis
expected direction that the results will go in can be predicted
non-directional hypothesis
cannot make a clear prediction, or if evidence is mixed. difference will be found but we dont know what it is
extraneous variables
variables that might interfere with the relationship between the IV and DV
determining cause-and-effect is threatened by EV’s
controls
something you do to minimize the possibility that an extraneous variable can affect your results
how to write a conclusion
- identify the relationship between the variables in the study
- identify the results
- apply this to the aim of the study
sampling methods
- self selected
- opportunity
- random
- stratified
secondary sampling techniques:
- purposive sampling: looking for people with a very specific set of traits
- snowballing sampling: type of purposive sampling used when looking for participants from a specific group which would not respond to an ad in a paper
external validity
when a study can be generalized to the target population
self-selected sampling
participants select themselves by volunteering
+quick and easy way to recruit individuals and have wide coverage
- not representative, share similar characteristics
random sampling
every member of the target population has an equal chance of being selected. takes into account all possible essential characteristics of the target population
+ easily generalizable, random sample of sufficient size is representative
- difficult to achieve with a larger population
opportunity (convenience) sampling
when you recruit participants that are more easily available
+quick, easy and cheap
- lack of representativeness, researcher bias
stratified sampling
sample that reflects the sub-groups within a target population
+ very representative
- time consuming
experimental designs
- repeated measures
- independent measures
- matched pairs
repeated measures
when the same group of participants takes part in both conditions of the IV
+ no participant variables
+ less participants needed
- order effect
independent measures
when you divide your participants into different groups that take part in different conditions of the IV
+ no order effect
- participant variable
- more people needed
matched pairs
when you split pairs of participants with similar characteristics into two different groups through matching tests
+ participant variables reduced
+ no order effects
- time consuming
- if one participant drops out you lose 2 data sets
- impossible to match people exactly
controlling for researcher and participant bias
- single blind
- double blind
single blind
when only the participant or researcher doesn’t know if they are in the ‘treatment’ group or ‘control’ group (placebo)
- treatment group experiences the factor that the researchers hypothesize will have an effect
- the control does not receive this treatment, but something in its place
double blind
when neither the participant nor the person gathering the dependent variable data knows which group the participant is in
- reduces researcher bias
interference effect
when one condition of the IV interferes with the other, distraction tasks are used to reduce this
ethical consideration
Consent (informed consent)
Anonymity and confidentiality
Right to withdraw
Deception: commissioned and ommissioned
Undue stress or harm
Debrief
confounding factors
(extraneous variables) influence the cause-and-effect relationship between the IV and DV, decreasing internal validity
validity
the extent to which the study accurately reflects or assesses the specific concept or variable it is intended to measure
construct validity
characterizes the quality of operationalisations. refers to how well a test or tool measures the construct that it was designed to measure
internal validity
characterizes the methodological quality of the experiment. how certain we are that it was the change in IV that affected the DV
when confounding variables are not controlled this decreases internal validity.
external validity
characterizes the generalizability of the findings in the experiment
- population validity
- ecological validity
ecological validity
the extent to which findings from an artificial environment in an experiment can be generalized to other environments
population validity
the extent to which findings can be generalized to the target population
researcher bias
when the beliefs or opinions of the researcher influence the outcomes or conclusions of the research
- confirmation bias
- questionable practice of p-hacking
- funding bias that leads to publication bias
controlling researcher biases
- decide on hypothesis before carrying out the research and do not go back and adjust it in response to results, instead run another experiment for new hypothesis
- to control for confirmation bias use researcher triangulation to improve inter-rate reliability
- double blind control is the standard control for researcher bias
demand characteristics
any aspect of the study that may lead participants to figure out how their supposed to act
- participant variability, (selection bias)
- expectancy effect
- social desirability effect
- reactivity effect
- screw-you-effect
- evaluation apprehension effect
participant variability
(selection bias), when participants are selected in a way that does not make them representative of the target population
expectancy effect
when participants change their behaviour as a result of behaving in the way that they are expected to
social desirability effect
participants change their behaviour because they have a desire to be liked by other people
reactivity effect
when participants don’t behave as they normally would because they are being watched
screw-you-effect
when participants act in a way that might sabotage the researcher’s aim
evaluation apprehension effect
(spotlight effect) when participants are anxious about being evaluated on a particular skill, which might change their behaviour
controlling for demand characteristics
- deception- single blind procedure
- field experiment (natural environment)
- independent measures design (can’t figure out the aim)
order effects
- practice effects
- fatigue effects
- interference effects
practice effect
type of order effect, participants may become better at a certain task as they become more familiar with the testing environment
fatigue effect
type of order effect, participants may perform worse near the end of an experiment because they’ve become fatigued from performing the same task multiple times
interference effect
type of order effect, when previous responses disrupt performance on a subsequent task. they are more likely when the second task quickly follows the first and the responses required conflict with each other.
controlling for order effects
- distraction task
- counterbalancing
- long wait between tasks
types of experimental methods in psychology
- true experiment
- field experiment
- quasi experiment
- natural experiment
field experiment
conducted in a real life setting
researcher manipulates the IV
lack of control over EV’s due to natural environment
+ high ecological validity
- low internal validity due to high confounding variables
quasi experiment
designed similar to true experiments (controlled environment) however no random allocation because participants are grouped based on their characteristics
when researcher is interested in IV’s that cannot be assigned or manipulated
variable is an innate characteristic of the participant
natural experiment
- tests effect of IV on DV
- IV= naturally occuring environmental factors e.g introduction of TV
- type of quasi experiment
correlational studies
different from experiments, no variable is manipulated so no causation can be inferred
two or more variables are measured and the relationship between them is mathematically quantified
illustrated graphically through a scatter plot
bidirectional ambiguity
seen in correlational research.
because there is no IV, can’t know if x causes y or y causes x, if even interact to cause behaviour, or whether it is coincidental or a third variable causing the pattern
case study
in-depth investigation of a unique individual or group over a long period of time
- longitudinal
- research method triangulation: uses multiple research methods
- unique individual or group
- less focus on generalizability and sampling: you are interested in this participant not the population
observation methods
naturalistic observation:
- dont manipulate any variables
- subject’s behaviour observed in a naturalistic environment
- field notes/data gathering techniques used
- may be followed by interviews
types of naturalistic observation:
- covert: subjects are not aware that they are being observed
- overt: subjects are aware of observation
- participant: researcher interacts and becomes part of the group they are observing
- non-participant: researches stays removed from the observation
