Studied by 0 people
generalizability
the extent to which the results of the study can be applied to a broader context beyond the sample and the settings used in the study itself.
provides numerical data that allows for statistical analysis to produce knowledge that can be applied as widely as possible
population validity
the extent to which finding can be generalized from the sample to the target population
conditions for high population validity
when a sample is representative of the target population AND an appropriate technique is used
probability sampling methods
preferred because it can reduce error and bias in sample’s selection
the method attempts to ensure every member of the target population has an equal chance of participating in the study
sample size
can influence the statistical representativeness
minimum size is 10% of the target population and should not exceed 1000
the more statistically representative the sample size the more accurately the results should reflect the potential target population
sample characteristics
sample should reflect all essential characteristics with of the target population
must identify the essential characteristics relevant to the target population BEFORE highlighting characteristics that may not have been represented well
essential sample characteristics
age, gender, ethnicity, culture, sexuality, education, language, religion
ecological validity
the extent to which findings can be generalized from the experiment to other settings and/or situations
both the task and the settings must be considered when judging for ecological validity
real-life environment
environment needs to be judged based on how realistic or artificial it is
realistic task
participants should be asked to do tasks or behaviors that they would normally do in real-life
replications
when studies are replicated and achieve the same or similar results as the original study, it gives greater validity to the findings.
quantitative research
relies on objective, numerical, and/or measurable data and is analyzed statistically to identify patters
qualitative research
relies on subjective experiences, personal accounts, and/or documents that illustrate in detail how people think and/or respond in society. this is thematically/contextually analyzed to complete and gain insights and contextual depth
experimental research key characteristics
is manipulated and DV is measured, confounding variables are controlled, or possibilities of causal inferences
lab or field
distinguishable by experimental design sub-types
correlational research key characteristics
variables are measured not manipulated
no causation can be inferred
refer to another way variables are measured
defined by process not by method of data gathering
qualitative research key characteristics
give defining characteristics to the larger method
unique characteristic distinguishes hit from other sub-types in the same category
true lab experiment
random allocation
characterized by its controlled environment
causal relationship can be established
field experiment
naturalistic environment
causal relationship MAY be established
not all confounding variables may be accounted for
quasi-experiment
IV is not manipulated but a naturally occurring characteristic of the participants
setting can be field or natural
no causal relationship can be established
natural experiment
IV is not manipulated but a naturally occurring aspect that impacted participants
extraneous/confounding variables are not always controlled
comparison of behavior before and after the event can occur
causal relationship can not be established
correlational study
no IV or DV but co-variables
measures the linear relationship between co-variables
correlation coefficients measure -1.0 to 1 based on strength of relationship
no causal relationship can be established
utilized inferential statistics and statistical significance
data is collected through surveys or questionnaires typically
survey
gathers data, calculated correlations, and makes statistical inferences on a large number of participants
fata gathering techniques are typically questionnaires or something similar
often gathers a random sample
case study characteristics
an in-depth investigation of a unique individual/group
longitudinal
often uses different methods
case study pros
useful for unique phenomena
can contradict established theories
case study cons
researcher and participant bias
no generalizability
ethically confounding - hard to maintain confidentiality
observation
researcher visually monitors participants and record their behavior
no manipulation - no IV or DV
limited by methods of data generation (selective attention and interpretation)
naturalistic observation
observation is done in real-life settings not arranged for the study’s purpose
naturalistic observation pros
used to analyze behavior that is unethical to artificially simulate
participant behavior is not impacted by research procedure
naturalistic observation cons
time consuming
laboratory observations
observation carried out in specifically designed environments and the participants know they are participating
laboratory observations pros
possible to create rare circumstances uncommon to be naturally observed
isolates behavior of interest efficiently
laboratory observations cons
artificial environment may influence participants behavior
overt observation
observation where participants are aware they are under observation
overt observation pros
follows ethical guidelines and obligations
overt observation cons
participant expectations due to their knowledge of observation may influence behavior
covert observations
observation where the researcher does not inform participants that they were being observed for their behavior
the researcher may or may not be actively present
covert observations pros
participants behave authentically/ have natural reactions
covert observations cons
raises ethical concerns as the researcher does not follow the ethical guidelines and obligations because participants are not aware and did not consent to this observation
participant observation
the researcher is an active member who participates in the observation group
participant observation pros
allows researcher to experience phenomena and gain personal insights
participant observation cons
risk that the researcher becomes too involved and loses objectivity
non-participant observation
researcher remains removed from the observed group
non-participant observation pros
more impartial
non-participant observation cons
some details can only be understood from the perspective of an active participant of the group
interview
a self-report technique that has the researcher ask questions of participants to give answers on their personal experiences
self-reporting
give insights of the perspective of the participant
interview data that comes in the form of of an audio/video recording is subsequently converted to an interview transcript which is analyzed to identify patterns and themes (inductive content analysis)
structured interview
include a fixed list of questions and topics in closed questions asked in a fixed order
highly controlled
bound by an interview schedule
useful when several interviewers are involved to ensure the process is standardized
structured interview pros
does not require much training
data is easier to analyze
useful when several interviewers are involved to ensure the process is standardized
structured interview cons
unique circumstances can not be accommodated in this method
more context behind rationale of participant can not be fully explored
semi-structured interview
uses an interview guide with a list of questions and topics that need to be covered
allows for flexibility where elaborate answers can be provided
questions can be open or closed
semi-structured interview pros
fits the natural flow of a conversation better
better suited for socially sensitive issues
more effective in studying the unique experience of each participant
semi-structured interview cons
less comparability across interviews as flexibility of questions are added
interviewer must be trained and now how to establish rapport
time consuming to analyze data
unstructured interview
involves an informal discussion on a particular topic and the next question is determined by the participant’s prior response
unstructured interview pros
very effective for unique cases or cases where no theoretical expectations exist that would inform the wording of questions
unstructured interview cons
more time consuming and results are difficult to analyze and interpret
requires extensive training
focus groups
a special type of semi-structured interview that is simultaneously conducted with 6-10 participants
participants are encouraged to interact with each other
group dynamics can be analyzed
interviewer can act as a facilitator to keep interactions focused on research questions
focus groups pros
more natural behavior as participants interact with each other rather than just the interviewer
interactions can reveal more than a one-on-one session
easier to respond to sensitive questions in group settings (typically)
multiple perspectives allow for a more holistic understanding
focus groups cons
dominant responders can disrupt group dynamics as their assertiveness can impact and distort other participants behaviors and their responses
more difficult to preserve and maintain confidentiality
especially demanding in terms of sampling and creating interview transcripts
considerably more training for a interviewer
types of probabilistic sampling research
random sampling
types of non-probabilistic sampling research
purposive sampling
convenience/ opportunity sampling
snowball sampling
volunteer/self-selected sampling techniques
probabilistic sampling technique
samples are not selected based on specific selection criteria
most rigorous approach to sampling for statistical research
non-probabilistic sampling technique
samples are selected based on specific selection criteria
random sampling
probabilistic sampling
when every member of the target population has an equal chance of selection
reduce chance of sampling bias
convenience/ opportunity sampling
non-probabilistic sampling technique
gathers participants who happen to be available for the study for both researcher and participant
quick, easy, and inexpensive
volunteer/self-selected sampling
non-probabilistic sampling technique
participants approach researchers and volunteer to be a participant
typically a form of marketing calls for participants
participants may have more commitment to the study since they advocated to participate
snowball sampling
non-probabilistic sampling technique
a group of initial participants act as a “seed” to invite more participants
sample keeps growing until desired size has been reached
useful when studying “hidden populations” (people who do not want others to know about them/ people generally hard to find)
purposive sampling
non-probabilistic sampling technique
participants are chosen for salient characteristics that are relevant to the researchers
easy to select
flexible as it can be supplemented with more participants need be
sampling
method of selecting a group of participants to undergo research which will create data for researchers to collect and analyze
goal of quantitative sampling
to find a sample that is statistically representative - large sample size and a chance all sub-groups of the target population are represented
goal of qualitative research
to select participants who are particularly informative about the research topics under investigation BUT generalization of findings is less important
temporal validity
an external validity that refers to the generalizability of a study’s results across time
construct validity
the extent to which results of the study can be generalized from operationalizations to theoretical constructs
when is construct validity high?
this leap is justified if the operationalization provides sufficient coverage of the construct
transferability
is the extent to which we can transfer the findings from one study to another context
case-to-case generalization
which refers to applying the findings from the setting of the research status to other setttings
how to achieve case-to-case generalization in qualitative research
thick & rich descriptions to ensure the reader has sufficient information and details on the study’s context
new context is similar enough to findings described in the report
replication of findings
theoretical generalization
generalization is made from observations to the formations of a broader theory
how to achieve theoretical generalization in qualitative research
rigorous analysis and interpretation for research findings
data saturation
thick, rich descriptions
in-depth analysis
free of bias
data saturation
further data does not add anything new to the already formulated conclusions and interpretations
sample-to-population generalization
applying the results of the study from the participants who took part in the study to the wider population
how to achieve sample-to-population generalization in qualitative research
bias
refers to factors that may negatively affect/alter the results of the study, decreasing accuracy
methodological bias/ procedural bias
flaws in the design, procedures, or materials of the study that were not avoided
how to avoid methodological bias
methodological triangulation
controls
replication
methodological triangulation
using a second method to gather data to support or refute ideas in the original method
participant variability
groups are not entirely equivalent at the start of the experiment
differentiation skews the impact of/ strength of relationship
order effect/ testing effect
DV is tested twice and prior experience impacts the true results of the participants behavior because they have done it prior
instrumentation
instrument measuring the DV changes slightly between measurements, compromising standardization of the measurement process
history
outside events that happen to participant during the experiment
maturation
natural changes that participants go through in an experiment
experimental mortality
participants drop out of the experiment
only a problem if the rate of dropping out is not the same in every experimental condition
researcher bias
when thoughts, beliefs, r ideas of the researcher are negatively influencing this study’s results
ways to avoid researcher bias
researcher triangulation
controls
replication
researcher triangulation
using more than one researcher in the process, reduces the chances that one researcher’s beliefs or opinions can affect the study
participant bias
when an individual’s existing thoughts, beliefs, or ideas influence their thinking or behavor
sampling bias
sample is not representative of the target population
why is quantitative research always biased?
non-probabilistic sampling techniques will introduce bias as they automatically do not equally represent the full target population.
ways to avoid participant bias
controls
replication
making sure that questions are more open ended
non-judgemental interviewers
establish a good rapport
keep dominant participants in check
have an awareness of cultural differences
participants feel safe to answer truthfully
how to avoid sampling bias
ensure a large enough sample size OR add more participants
recommend to attempt random sampling
demand characteristics
participants understand the true aim of the experiment and then alter their behavior intentionally or unintentionally as a result
correlational research bias
credibility is the extent to which the results of the study reflect the reality that is being investigated
bias in correlational research can occur on two levels:
the level of variable measurement
level of interpretation of findings
Note 
Flashcard (60)