Research Design Flashcards
What the difference between Research and Evidence?
-Research is results from individual studies, whereas evidence is an
accumulation of results across studies
-Evidence is the synthesis of all valid research studies that answer a specific
question
-Evidence-based practice involves tracking down available evidence, assessing
its validity, & using the best evidence to inform treatment decisions
What is Validity?
closeness to the truth; degree to which design & method provide accurate
investigation of event in question
What is Internal Validity?
(selection, maturation, instrumentation)
Are the results valid for the patients in the study?
Was the study performed correctly?
Was any difference
between groups seen?
Threats (yikes):
Selection bias: sample is collected where some members of the intended population are less likely to be included
What is Maturation?
Wrong age group or it ages out
What is Instrumentation?
How you measure, are you doing it appropriately?
Random assignment to groups addresses many threats to internal validity but….
not all!
What is External Validity?
(publication, financial)
Ability to generalize findings: beyond subjects in the study, beyond environmental
constraints of current study and to other temporal periods
Are the results valid for patients not involved in the study?
Does the study population represent the larger group?
As controls increase (increasing internal validity)….
the generalizability of findings may suffer (decreasing external validity)
Threats: Publications & Financial
Quantitative data (numbers) vs. Qualitative data (interpret)
Inquiry rooted in empiricismbased in hermeneutics
Only measurable phenomena is real Interpretation/contextual
Measures often numeric scales meaning
Subjective measures dependent on perceptual biases
What are tests for Data Examples?
Chi-square for nominal/categorical data
Mann/Whitney for ordinal data (severity of perio-disease)
T-tests or ANOVVA for continuous/interval or ratio data (pocket depth)
Systemic Reviews vs. Meta Analysis:
-Systematic review is formal approach to review all relevant literature on a topic..
whereas meta-analysis..
is a statistical method used to combine the numerical results
from such studies, if possible to do so. Systematic reviews often include a meta-
analysis
What are Randomized Control Trials?
Strongest evidence for demonstrating causes and effect
Reduces effect of bias due to intervening variables
Assumes that confounding conditions will be equally distributed among groups
What do Randomized Control Trials include?
● At least one “varied condition” (treatment vs no treatment)
● Concurrent enrollment
● Random assignment to groups
● Follow up
What does Blinding do?
Blinding attempts to reduce bias due to expectations or preconceptions of
patients or investigators (double blind)
Describe Experimental Research:
Quantitative
Investigates cause
Controls or manipulates variables under investigation
Describe Observational Research (Cohort & case control studies):
Quantitative or qualitative
Without experimental controls (may include comparison of natural group)
“Quasi experimental”
Investigation of relationships, but not cause
i.e. researcher investigates relationship between smoking and lung cancer
(considers exposure and outcomes instead of testing an intervention like a drug)
What are examples of Variables?
Any factor relevant to particular study; may be known or unknown; examples → age,
SES, ethnicity, disease history
What are Independent Variables (changes) vs. Dependent Variable?
Independent =Variable:
Factor changes naturally, Observed variable in a study.
=Intentionally Manipulated, Changes determined by level
> > ???
Grouping variable of independent variable(s)
Known & controlled by experimenter Factor directly affected by
“Causative factor”another factor
“Response” or “outcome
What are Confounding Variables/
Error- not measured by researcher and has impact
Extraneous variable correlates w/ both D & I variables
Factor not considered by researcher that has significant impact on the dependent
variable or outcome of interest
“Confounding influences” or “error”
What is the Difference between Prospective vs. Retrospective Research?
● Prospective: looks at events that haven’t happened yet or constructs that have
not yet been measured
● Retrospective: looks at data that already exists
When appraising a study, what should you pay close attention to?
When appraising a study pay close attention to descriptions on selection and
assignment process of participants and other factors like hypotheses, intervention»_space;>assessment, & data analysis«
