Research Methods Flashcards
cross-sectional study
- snapshot of a population at one point in time
- type of observational study
- sample recruited based on exposure and outcome
longitudinal study
- long period of time
- uses specific cohort of people
case control study
- individuals with a condition are matched to similar indiiduals without condition to identify factors that may have led to condition
- recruit samples based on outcome, then determine what exposure was
confounding variable
variable (not independent variable) that influences the dependent variable
related to research design
usually try to CONTROL for these variables
mediating variable
prove a MECHANISM to explain the relationship between the 2 variables
can help produce CAUSAL relationships
moderating variable
NOT a mechanism, but affects the STRENGTH of the relationship
affects how much the independent variable affects them
affects the already-established relation between IV and DV
placebo effect
effect produced by a placebo treatment that cannot be attributed to the properties of the placebo itself, so its due to person’s BELIEF about treatment working
FINER method
if a study is Feasible, Interesting, Novel, Ethical, Relevant
Hill’s criteria
help determine strength of a causal relationship
only temporality is necessary for causality!!!
SD distribution
- 68% within 1 SD
- 95 % within 2 SD
- 99.7% within 3 SD
selection bias
sample differs from the population
eg. not randomly selected
detection bias
educated professionals using their knowledge in an inconsistent way by searching for an outcome disproportionately in a certain population
hawthorne effect
behaviour of subjects is altered due to the fact they know they are being monitored
social desireability bias
type of response bias, tendency to answer in a way that would be favoured by others
positive control
- ensure a change in dependent variable when a change is expected
- A positive control is a control group that is not exposed to the experimental treatment but that is exposed to some other treatment that is known to produce the expected effect.
negative control
- ensure NO change in dependent variable when NO change is expected
- A negative control group is a control group that is not exposed to the experimental treatment or to any other treatment that is expected to have an effect.
validity
ACCURACY
ability to measure true value
reliability
PRECISION
read consistently / narrow range
4 ethics in medicine
- beneficence
- nonmaleficence
- justice
- autonomy
3 ethics in research
- respect for persons
- beneficence
- justice
internal validity
support for causality
when the dependent variable changed as a result of the independent variable
external validity
support for generalizability, higher with large samples or more representative samples of larger population
statistical significance
not the result of random chance
p values
clinical significance
is there a worthwhile change in health status as a result
type I error
reject null when the null is true
alpha
type II error
fail to reject null when the alternate is true
beta
within subject design
Controls for individual variations in a measurement by comparing the scores of a subject in one condition to the scores of the same subject in other conditions
the subject serves as its own control
repeated measures design
retrospective VS prospective study
- retrospective = information collected about individual’s past
- prospective = followed over time and data gathered
ethnographic study
study social interactions in real settings
observing people in their everyday environment to understand their experiences, perspectives, everyday activities
gain insight into cultures/groups
sampling bias
when some members of a population are morel likely to be sampled than others
limits generalizability of findings
between subjects design
subjects randomly assigned to an experimental group
only participate in ONE group
mixed methods research design
use both qualitative and quantitative methods in the study
attruition bias
occurs when participants dro out of a long term longitudinal study
attruition bias
occurs when participants dro out of a long term longitudinal study
reconstructive bias
Reconstructive bias is a type of bias related to memory. Most research on memories suggests that our memories of the past are not as accurate as we think, especially when we are remembering times of high stress.
construct validity
extent to which your test/measure accurately measures what it is supposed to
eg. The study aims to examine how social class influences perception of race. However, if the categories of “Black” and “White” are viewed by participants as being categories of social class rather than race, then they are not reflecting what the researchers intend to measure (participants’ views of race). This is a problem with construct validity, or the manner in which the terms of the study are defined
hindsight bias
Hindsight bias refers to the tendency to believe that past events were highly predictable, even though those events were not predictable before they occured.
This example describes current researchers who believed that a past event (the effectiveness of an intervention) was highly predictable, when it was not easy to predict at the time it occured.