RESEARCH DESIGN and ETHICS Flashcards
a true experimental design in which subjects are randomly assigned to at least two comparison groups
randomized controlled trial
list the three essential characteristics of an experiment
- independent variable must be manipulated by researcher
- a control group must be incorporated into the design
- subjects must be randomly assigned to groups
experimental trials that examine the effect of interventions on patient or community populations; either therapeutic or preventative
clinical trials
pre-clinical research
in vitro
animal research
design drug delivery or surgical methods
if successful seek approval to begin human trials
phase 1
purpose is determining safety
small number of subjects
data on dosage, timing, side effects
phase II
demonstrate effectiveness
small samples several years
should be better than current protocol to continue
phase 3
compares new therapy to standard treatment or procedure
- randomized, controlled, blinded trial
large scale, many subjects
- if successful, seek FDA approval
phase 4
- patients followed after FDA approval
-evaluate effects in other populations - if problems are found, drug may be taken off market
aka one way design
one independent variable with any number of levels
- may have one or more dependent variables
-compares 2 or more groups randomly assigned
single factor experiment independent groups
factorial designs incorporates two or more independent variables; looks for main effects and the interaction effect of variables a and b together
multi factor designs for independent groups
example of a multi factor design for independent groups
analyzed by 2 or 3 way anova
one group, repeated measurements “within subjects” design
- each subject is its own control
- controls for individual differences because most factors remain constant throughout trials
repeated measures design
one independent variable (with levels), one group
- each subject is exposed to every level of the independent variable
-crossover design
one way repeated measure design
this is an example of:
each subject has eyesight measured with contacts and glasses and without any aid
one way repeated measures design
groups are selected on the basis of whether or not they have the disorder under study; PURPOSE IS TO DETERMINE WHY CASES HAVE THE DISEASE AND CONTROLS DO NOT
case-control study
two groups are chosen, one which has been exposed to a clinical intervention or agent, by chance or choice; ESTABLISH RELATIONSHIP BETWEEN EXPOSURE AND RISK OF DEVELOPING A PARTICULAR DISEASE
cohort study
description of new, interesting, and unique clinical cases as a means of sharing special information to professional colleagues
case studies
DESCRIPTIVE RESEARCH DESIGNS
description of developmental change and sequencing of behaviors over time
developmental studies
observation of different age or developmental groups at one point in time
cross-sectional study
called “Self reports” meaning that the researcher does not directly observe the behaviors but only records the respondents report of them
interviews and questionnaires
can be used under any classification of research, but they are most commonly used as methods for collecting descriptive data
surveys
these are the objectives that delineate what the researcher is trying to find out
formulation of guiding questions
detailed outline listing each item of information needed to answer guiding questions
questionnaire outline
find out if there is a similar questionnaire out there that can be used with modification; doesn’t necessarily have to be in the same field.
review of existing surveys
write a series of questions that address each components of the guiding questions; questions can be grouped by category or topic
preliminary draft
after draft is ready, pass out questionnaire to about 5-10 people to see if they understand it
pilot testing and revisions
verify the questionnaire measures what it is intended to measure
validity
should be looking friendly with uncluttered pages, nice print, looks good. includes some type of introduction or cover letter to describe the purpose of the study with instructions
final draft
must have an accessible population identified; anonymous; administered in person, phone, mail; may have to follow up to get better response
administration of survey
allow the respondent to express themseles
open ended questions
ask respondents to select an answer from among several choices provided to them
closed ended questions
a series of statements is presented expressing a viewpoint and respondents are asked to choose an appropriately ranked response
likert scales
statements are presented that reflect increasing intensity of the characteristic being measured
guttman scale
how are systematic reviews different from classical literature reviews
clinical decision vs updating the field
what are the “subjects” in a systematic review?
studies - papers
examining data from papers (similar topic or hypothesis) to put forth a new conclusion
meta analysis
how are studies in a systematic review evaluated for validity or methodological quality?
stats- peer review- rating scales
science is built on a foundation of trust and honesty
scientific integrity
list underlying motives for scientific fraud
- career/ academic pressure
- belief that you know the answer and take shortcuts to get there
- some experiments yield data that are not precisely reproducible
what is not misconduct
proposing hypotheses that ultimately turn out to be false
offering conflicting interpretations
making erroneous observations and analyses
usually done by senior members of the scientific community
peer reviews
an individual exploits, or appears to exploit their position for personal gain or for the profit of a member of their immediate family or household
conflict of interest
editors delaying publication of competitor. reviewers not being sufficiently critical of a friends manuscript/grant
scientific conflict of interest
types of intellectual property and ownership
trade secrets
trademarks
copyrights
patents
protect expression/ presentation of idea; not idea itself
copyrights
blastomere separation and somatic cell nuclear transfer to create a copied organism
reproductive cloning
one member must be a veterinarian
- practicing scientsis with experience in animal research
non scientist
non institutional affiliated individual
institutional animal care and use committees
list components of belmont report
respect for persons: dignitiy and autonomy
beneficence: max benefits, min risks
justice: distributed benefits and burdens of research
reviews all research proposals relating to human research determines risk to benefit ratio
IRB institutional review board
