Lecture 6 Flashcards
Within a postpositivist worldview, quantitative strategies are
used to answer research questions and test hypotheses related to:
- determining associations
- comparing groups
- developing and testing measures
- theory verification
5 step process of quantitative design process:
- determining basic questions to be answered
- determining study participants
- selecting methods needed to answer questions
- selecting analysis tools
- understand and interpret results
When using hypotheses, it is important that they are developed with _____ in mind.
theory
The research question is key as it guides the ____ selected.
method
An advantage of quantitative research is one can use _____ groups (_____) to potentially make ______ to the _____
population
- smaller
- sample
- inferences
- larger
Population:
An entire group or aggregate of people or elements having one or more common characteristic
Sample frame:
The group of accessible people that can be connected with about the study
Sample:
A sub-group of the population that can be managed by the researcher but will represent the population
Sampling:
The process a researcher uses to obtain a sample
from the target population
2 types of sampling:
- probability
- nonprobability
In probability sampling, samples are selected using ____ _____ ensuring that…
- random processes
- every unit in the population has an equal probability of being selected
In probability sampling, the probability of selecting each participant or element is _____.
known
In probability sampling, estimating sampling error is _____.
possible
In non-probability sampling, how are samples selected?
not selected at random
In non-probability sampling, the probability of selecting each participant or element is ______.
unknown
In non-probability sampling, it is difficult to say if your sample is… and in turn difficult to…
- representative of population
- generalize findings
Non-probability sampling is ____ expensive and ____ complicated.
- less
- less
4 types of probability sampling:
- simple random sampling
- stratified random sampling
- systematic sampling
- cluster sampling
Simple random sampling:
- every individual has equal opportunity of being selected
- selection of one member does not affect the chances of another member being chosen
Stratified random sampling:
- dividing population elements into subgroups (STRATA) the randomly sample from each
- ensures representation from each strata
Systematic sampling:
- sampling units are selected in series according to some preset criteria or sequence
- selection of the 1st element is random, but after this selection is not independent (ex. select every 10th entry)
Cluster sampling:
- participants are randomly selected from a natural occurring group or unit in a population
- researcher specifies the cluster, which becomes the sampling unit
When to use simple random sampling:
anytime
When to use stratified random sampling:
when concerned about under representing subgroups
When to use systematic sampling:
when you want to sample every kth element in a ordered set
When to use cluster sampling:
when organizing geographically makes sense
Advantage of simple random sampling:
- simple to implement
- easy to expalin
Advantage of stratified random sampling:
allows oversample of minority groups to ensure subgroup analysis
Advantage of systematic sampling:
does not require that you count through all of the elements in the list to find the ones randomly selected
Advantage of cluster sampling:
is more efficient than other methods when sampling across geographically dispersed areas
Disadvantage of simple random sampling:
requires a sample list to select from
Disadvantage of stratified random sampling:
requires a sample list to select from
Disadvantage of systematic sampling:
if the order of the elements is nonrandom, there could be systematic bias
Disadvantage of cluster sampling:
is usually not used alone, combined with other methods
3 types of nonprobability sampling:
- purposive sampling
- convenience sampling
- snowball sampling
Purposive sampling:
- participants purposefully selected because they have specific characteristics the researcher is interested in
- not randomly selected = limited generalizability
- commonly used with very small sample sizes
- common in qualitative research
Convenience sampling:
selecting research participants on the basis of being available, accessible, and convenient to the researcher
Snowball sampling:
enrolled participants nominate or recruit potential participants who may meet the eligibility criteria
When to use purposive sampling:
when you want to examine specific characteristics or experiences
When to use convenience sampling:
anytime
When to use snowball sampling:
hard to reach populations
Advantage of purposive sampling:
easily understood, implement, and explain
Advantage of convenience sampling:
easy to do
Advantage of snowball sampling:
can be used with no sampling frame
Disadvantage of purposive sampling:
limited external validity, likely to be biased
Disadvantage of convenience sampling:
very weak external validity, likely to be biased
Disadvantage of snowball sampling:
low external validity
2 questions when selecting methods needed in quantitative design process:
- how many measurements are being used?
- what types of measures or observations are being used?
Selecting methods needed includes identifying:
- variables
- measures
- design
Variable:
an attribute or a characteristic that may vary over time or across cases
Types of variables:
- independent
- dependent
- mediator
- moderating
- control
- confounding
Dependent variable (DV):
- The variable that is being
affected - it is the outcome being assessed as a result of the
independent variable(s) and is the main focus of the study
Independent variable (IV):
the variable that is being manipulated (also called treatment variable)
Mediator variable:
A variable that is proposed to at least partially explain the relationship between an IV and the DV
Moderating variable:
A variable that affect the relationship between two other variables (predictor and outcome)
Control variable:
A variable that could influence the outcome or results of the study . . . not the main focus of the study
Confounding variable:
An unmeasured variable that is controlled for in the study. It could be the variable could not be measured
4 types of measures:
- observational
- self-report measures
- objective
- estimates
Observational measures:
recorded by individual observing an action
Self-report measures:
an individual reports their own behaviour
Objective measures:
taken by instruments or other calibrated devices
Estimate measures:
subject matter experts provide best guesses
Validity asks…
does the measure do what i is supposed to do?
Reliability asks…
does the measure lead to consistent results?
3 common types of validity as it relates to measurements:
- construct
- content
- criterion
Construct validity:
how one translates the idea or construct into something real or concrete
Content validity:
a check of the operationalization against the relevant content domain of the construct
Criterion validity:
the validation of a measure based on its relationship to another independent measure as predicted by your theory of how the measures should behave
Reliability:
the repeatability or consistency of a test (or tester) or instrument
Reliability is important because any change in scores should reflect a true indication of one’s _____ and not….
- ability
- change over a short period of time
- depend on who is administering the test
A valid measurement is _____, but having ______ measurements does not always mean they are valid.
- reliable
- reliable
4 general classes of reliability estimates:
- inter-rater or inter-observer reliability
- test-retest reliability
- parallel-forms reliability
- internal consistency reliability
Inter-rater or inter-observer reliability:
assess the degree to which different raters/observers give consistent estimates of the same phenomenon
Test-retest reliability:
assess the consistency of a measure from one time to another
Parallel-forms reliability:
assess the consistency of the results of 2 tests constructed in the same way from the same content domain
Internal consistency reliability:
consistency of results across items within a test
Shooting-target metaphor: reliable not valid
you are hitting the target consistently but you are missing the centre of the target (it is consistent but not right)
Shooting-target metaphor: valid not reliable
- hits are randomly spread across the target
- seldom hit bulls-eye but on average are getting the right answer for the group (target)
- group estimate is valid, but inconsistent
Shooting-target metaphor: neither reliable nor valid
- hits are spread across the top part of the target but are consistently missing the bulls eye
- it is consistent and is not right
Shooting-target metaphor: both reliable and valid
- consistently hitting the centre of the target
- both consistent and correct
