EBP Flashcards
3 types of research designs
experimental
quasi-experimental
non-experimental
experimental research design must have
randomization, control, and manipulation
sampling
methods used to control bias
measurement
levels of variables
determining the significance of findings
p value
confidence intervals
essential components of experimental designs
randomly select subjects for the target population
randomly assign subjects to groups
control
manipulation
true experimental designs
randomized clinical control trials (RCT)
before and after designs
six types of true experimental designs commonly reported in the scientific literature
- two group, pre-test, post-test (baseline then test after intervention)
- two group, post test only (lack of baseline limits study)
- solomon four group (combo of 1&2)
- multiple experimental groups
- Factorial
- Crossover designs:
quasi experimental
involve manipulation of the independent variable but do not meet all essential components of experimental designs (lacks either randomization or control group)
3 common quasi experimental designs
- nonequivalent control group pretest-posttest
- time series designs
- preexperimental designs
times series design
a quasi experimental design where one group is measured prior to administering the intervention and then multiple times after the intervention
preexperimental design
a quasi experimental design with posttest only that involves manipulation of the independent variable but lacks control for extraneous variables
one group posttest only design
often seen as a pilot setting
no randomization, no pretest
Nonexperimental: 3 descriptive designs
used for the purpose of
- exploratory: describing a phenomenon in detail
- comparative: explaining relationships and differences among variables
- survey: predicting relationships and differences among variables
nonexperimental: 4 correlational designs
- covary (change in one variable is associated with change in another variable)
- descriptive-correlational designs (explain the relationship among the variables or groups using a nondirectional hypothesis)
- predictive correlational designs (researchers hypothesize which variables are predictors or outcomes)
- model testing correlational design (test a hypothesized theoretical model)
data planning and piloting is
key component in all research studies
quantitative data
collecting numbers questionnaires scales (likert scales, visual analog scales) physiological indicators (BP) Issues in quantitative data collection
exploratory/descriptive quantitative research design
answers “what” questions; describes frequency of occurrence
questionnaires, scales
correlational quantitative research design
examines relationships among variables
questionnaire, scales, biophysiological
quasi-experimental quantitative research design
examines why certain effects occur
questionnaire, scales, biophysiological
experimental/clinical trial quantitative research design
examines causes of certain effects
questionnaire, scales, biophysiological
Likert scales
ordinal level scale containing seven points on an agree or disagree continuum
Visual Analog scale
ratio level scale of a 100 mm line anchored on each end with words or symbols
nominal data
lowest level, categorized in groups
ordinal data
values continuum, intervals not equal
interval data
values continuum, intervals equal, no zero
ratio data
highest level, numeric values begin with absolute zero and have equal intervals
data must be collected in a very consistent manner
instruments should be administered in same order for all subjects, in the same context and setting, using the same set of directions
interrater reliability
the extent to which two or more individual raters agree
Random error
error that occurs by chance during measurement
systematic error
error that occurs in the same way with each measurement
validity defined and 3 types
the degree that an instrument measures what it is supposed to measure
- content
- criterion-related
- construct validity
Reliability
obtainment of consistent measurements over time
an instrument can be reliable but not valid. (scale measures your weight reliably, but not your anxiety attached to your weight)
content validity
test to ensure that the instrument measures the content
measured two ways: face validity and content validity (experts perform evaluation)
concept being studied must be clearly defined to ensure the selected instrument fits
criterion-related validity
degree to which the observed score and the true are related.
tested two ways:
concurrent validity and predictive
concurrent validity
new instrument is administered at the same time as an instrument known to be valid. Scores of the two instruments are compared. Strong positive correlations indicate good validity.
Predictive validity
new instrument is given at two different times and scores are correlated. Strong positive correlations indicate good validity
Construct validity
to what extent does it measure the theoretical construct or trait measured 6 ways: hypothesis testing convergent divergent multitrait-multimethod known groups factor analysis
hypothesis testing
hypothesis derived from theories are tested with the new instrument
convergent
new instrument is administered at the same time as in instrument known to be valid. Scores of the two instruments are compared. Strong, positive correlations indicate good validity.
divergent
new instrument is administered at the same time as an instrument measuring the opposite of the concept. Scores of the two instruments are compared. Strong NEGATIVE correlations indicate good validity
multitrait-multimethod
new instrument, established instrument of same concept, and established instrument of opposite concept are given at the same time. Strong + and - correlations indicate good validity
Known groups
new instrument is administered to individuals known to be high or low on the characteristic being measured.
Factor analysis
statistical approach to identify items that group together
correlation coefficient 1.00=
perfect relability
correlation coefficient 0.00=
absence of reliability
what is the acceptable reliability coefficient for well established instruments?
0.80
what is acceptable reliability coefficient for newly developed instruments?
0.70
what three attributes are important when testing instruments for reliability?
stability
equivalence
internal consistency (homogeneity)
what are the seven ways instruments are tested for reliability? pitsick
parallel or alternate form interrater test-retest split half item to total correlation Cronbach's alpha Kuder-Richardson coefficient
which of the following terms are associated with qualitative sampling? snowball random purposive subjects
snowball
Which of the following are techniques for maintaining scientific rigor in qualitative study? achieving saturation thick description peer debriefing generalizing to populations
qualitative research does not have the purpose to generalize to a broad population. Aim is to use research somehow. Therefore, the answer is peer debriefing, thick description, and achieving saturation
which of the following are criteria for establishing trustworthiness? Credibility reliability confirmability inferability
credibility and confirmability
which of the following are techniques used by qualitative researchers? peer review member checks data saturation audit trails
all
