MN580 Midterm Exam with Complete Solutions
a few considerations for reliability and validity - ANS-one of the most important
attributions of a study is that the measurement instruments reliably and accurately
reflect the theory being examined
failure to do so threatens the study conclusions
evidence of reliability and validity is of paramount importance
action research - ANS-can obtain data through groups that we work with that are
actively engaged in designing interventions and identifying areas of priorities and
experiences of importance, using gathered data to guide action
can be patient-centered or community-centered but are not always timely and not
always aligned with evidence or funding
action that demonstrates the role of a knowledgable consumer of nursing research -
ANS-evaluating the credibility of research findings
advantages and disadvantages of crossover design for epilepsy study - ANS-A: fewer
subjects required, patients are their own control, fewer investigative centers
D: carryover effects, dropouts are problematic, cyclicity of disease, mid-study washout
(second baseline)
advantages and disadvantages of parallel design for epilepsy study - ANS-A: longer
term studies, dose-response relationships, ID of placebo response, less dependence on
natural hx of epilepsy
D: more research subjects required, more investigative centers
advantages vs. disadvantages in physiologic measurement - ANS-advantages:
objective, precise, sensitive, minimizes distortion
disadvantages: expensive, specialized training, environmental influences,
environmental variable control considerations
advantages vs. disadvantages of existing data - ANS-advantages: data already
collected, large sample sizes, larger samples=more sophisticated statistical analyses,
time and $$ savers
disdavantages: many datasets lack de-identification so are not sharable, researchers
are restricted to a set list of variables
advantages/disadvantages to using a new survey - ANS-have to spend several
weeks/months developing and pilot testing
advantages/disadvantages to using a previously developed survey - ANS-psychometric
properties are developed
has established reliability and validity
it's been pilot tested
,it's been used before
you can add your data to the larger data that has been collected from it
it's tough to find a survey that is going to measure exactly what you're interested in
measuring
may just add questions to beginning and end of it, which destroys the validity and
reliability of the instrument (so need to determine psychometrics on their own)
will have to request permission from the author to use the survey
after only nonequivalent control group design - ANS-experimental group-->intervention
is applied-->postintervention data are collected
control group-->data are collected
posttest only control group design
assumption is made that experimental and control groups are equivalent
ex: prenatal interventions that may affect outcomes
after-only experimental design - ANS-sample is selected from population
subjects are randomized
intervention group and control group: postintervention data are collected for both groups
useful when testing effects are expected to be a major problem or when outcomes
cannot be measured beforehand such as an experiment on post-surgical pain
management
appraising evidence with reliability and validity - ANS-was an appropriate method used
to test the reliability of the tool?
is the reliability of the tool adequate?
was an appropriate method used to test the validity of the instrument?
is the validity of the measurement tool adequate?
have the psychometrics been recalculated if a different patient population was studied
(i.e., appropriate for use in new population)?
have the strengths and weaknesses related to the reliability and validity of each
instrument been presented?
what kinds of threats to internal or external validity are presented by weaknesses in
reliability and/or validity?
are strengths and weaknesses of the reliability and validity appropriately addressed in
the discussion or limitations section of the report?
how do the reliability and validity affect the strength and quality of the evidence provided
by the study findings?
appraising findings from qualitative research - ANS-credibility
transferability
dependability
confirmability
appraising the evidence for data collection methods - ANS-are the measures clearly
identified?
is rationale for their selection provided?
,is the method used appropriate to the problem or clinical situation studied?
were data collectors adequately trained?
were data collection procedures consistent?
appraising the evidence for existing data - ANS-are the existing data used appropriately
for the problem being studied?
is new information being generated by examining these data?
is selection bias a consideration?
appraising the evidence for observational methods - ANS-who did the observing?
were observers trained to minimize bias?
was there an observational guide?
did the presence of the observers affect the behavior of the study participants?
was informed consent used?
appraising the evidence for questionnaire methods - ANS-is the measure described
adequately?
is there evidence that study participants were able to answer the questions?
are the items appropriately close- vs. open-ended?
appraising the evidence of physiologic measurement - ANS-is the instrument used
appropriate to the problem?
is the rationale provided for instrument selection?
is there a provision for evaluation of the accuracy of the instrument?
appraising the evidence of self-report interview methods - ANS-was an interview guide
used?
who were the interviewers and how were they trained?
is there clear indication that the subjects understood the task and the questions?
is there evidence of interviewer bias?
approaches in qualitative research - ANS-ethnography
narrative
phenomenological
grounded theory
case study
appropriate methods for randomization - ANS-can be done individually or in blocks
last digit of SS#, random numbers table (generated by computer)
choose a "block size" in which the number of subjects is proportional within each "block"
(1:1 or 2:1 or 3:1 within blocks of 2, 3, or 4 respectively)
assessing quality in qualitative research: tool - ANS-QARI software developed by
Joanna Briggs Institute, Australia
see quality appraisal checklist in slides
, attrition - ANS-drop outs, such as for treatment non-compliance, AEs, lack of benefit,
moved away, changed phone number
basic ethical principles - ANS-respect for persons
beneficence
justice
beneficence - ANS-do no harm
want to maximize benefits and minimize risk and adverse reactions
bias - ANS-systematic error that enters a clinical trial and distorts the data (as oppsed to
random error)
affects dependent variable being studied
blinding - ANS-by keeping the identity of treatment secret, investigator bias can be
minimized
but which recruitment method? - ANS-review literature and determine what's best for
specific type of research you are conducting
try multiple methods, esp. to reach rural areas
pilot study
case study - ANS-interested in a bounded situation--study a particular medical error,
procedure, case, etc. and understand what was involved in that
collect different sources to really understand that case
focus: organization, entity, individual, or event
sample size not relevant, more of a closed entitity
data collection: interviews, documents, observations
what are the factors that influenced a medical error in an ICU?
causal relationship - ANS-one thing causes another to happen
important to tell correlation from causation
characteristics of a hypothesis - ANS-relationship statement (between DV and IV--
implies a systematic relationship--direction of relationship is usually specified using less
than, more than, difference--causal or correlating relationship is usually specified)
testable (must be observed, measured, and analyzed--must predict an anticipated
outcome)
consistent with research findings and/or defined theory base
research problem-->lit review-->theoretical framework-->hypothesis
choice of method - ANS-depends upon problem you want to solve
classic randomized, controlled clinical trial (RCT) - ANS-sample is selected from the
population
baseline data are collected