Research methods Flashcards
Cohort study
To investigate the causes of disease, establishing links between risk factors and health outcomes. Cohort studies are usually forward-looking - that is, they are “prospective” studies, or planned in advance and carried out over a future period of time.
ex: Study of high school students from Baltimore, Maryland, and that looks at the differences in initiation of tobacco use between a cohort of adolescents that started working for pay and a cohort of adolescents that did not work.
Cross sectional study
type of observational study that analyzes data collected from a population, or a representative subset, at a specific point in time—that is, cross-sectional data.
ex: Look at women of a specific age with breast cancer.
Case-control study
A study that compares patients who have a disease or outcome of interest (cases) with patients who do not have the disease or outcome (controls), and looks back retrospectively to compare how frequently the exposure to a risk factor is present in each group to determine the relationship between the risk factor and the disease.
ex: Study of two participants one with lung cancer and one with no long cancer who both lived in a household with parents that smoked as children.
What is selection bias?
Methods used to select participants is not representative of the whole population. no true randomization.
what are other types of bias?
- observers bias
- Demand characteristics
- information bias
- confounding variables
- placebo effect
- detection bias
- performance bias
- experimenter bias
what is self selection bias?
individuals select themselves into a group
what is pre- screening or advertising bias?
Pre-screening or advertising bias occurs when the screening or
advertising process itself results in an unrepresentative sample. For example, advertisements
asking for volunteers for the same study that are worded as “Volunteers needed for an obesity
study,” or “Volunteers needed for a weight-loss study” will likely elicit different volunteers.
what is specific real area bias?
Bias based on location like surveying teenagers that do drugs but only those in a high school and not taking into account those that are homeschooled or dropouts.
what is exclusion bias?
excluding subjects that have recently moved into/ out of study area.
What is healthy user bias?
The kind of subjects that voluntarily enroll in a clinical trial and actually follow the experimental regimen are not representative of the general population. They can be expected, on average, to be healthier as they are concerned for their health and are predisposed to follow medical advice,
What is overmatching bias?
Matching with exposure but not disease,
what is Bergson’s fallacy?
the selection of participants from hospitals, where the participants are likely to be
LESS healthy than the general population (the opposite effect of the healthy-user bias).
How is accuracy different from precision?
Accuracy is the degree to which a value represents the true or correct value. Precision is the measure of the degree to which repeatedly takes measures are the same.
What is the difference between reliable and valid?
results are consistent and repeatable while valid measures what it was meant to measure and meets scientific standard criteria.
What is r^2 and what does it measure?
Correlation coefficient –> linear regression analysis.
What is the value for the perfect r^2?
1
show good internal reliability
What does correlation coefficient explains?
The amount of variance in y accounted for by x.
What are the Hill’s criteria?
Criteria used to evaluate whether or not a causal relationship exist.
- temporality
- strength
- consistency
- specificity
- plausibility
- dose-response relationship
- testable by experiment
- coherence
- analogy
What is single blind vs. double blind?
Aspects of the study is hidden from person doing the assessment .
Double blind - both participant and person assessment don’t know what’s going on.
Population ?
NYC population
sample?
random sample of NYC population that represents population.
statistic?
fact or piece of data from study of a large quantity of numerical data.
parameter?
conditions experiment is performed
mean
average
median
middle number/ midpoint
mode
numbers that appear the most
range
range of data points
standard deviation
quantity calculated to indicate extent of deviation of group as a whole. atop
when are results significant?
p
Do we reject or concur with the null hypothesis if p
we reject null
what is a type I error?
claimed difference between groups when there is none
what is type II error?
did not claim a difference between groups when one DID exist
what’ standard error of the mean (SEM)?
quantification of how precisely the mean represents the true mean of the population.
Does SEM increase of decrease as sample size increases?
decreases
Where is the error margin?
the thin extending bar above the graph or bar graph –> the bigger it is = the bigger the error.
