Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
65 Cards in this Set
- Front
- Back
in quantitative studies, what is reliability?
|
It is the accuracy and consistency of information obtained from a study (methods used to measure research variables).
it is the degree of consistancy or dependibility with which an instrument measures an atribute. |
|
in quantitative studies, what is validity? (measurement)
|
the degree to which an instrument measures what it is intended to measure.
|
|
what is "Trustworthiness" in qualitative studies?
give an example of "triangulation" |
Credibility is the extent that research methods engender confidence in the truth. Could be enhanced by using triangulation
triangulation ex: ask not only the patient, but the realative what makes this person anxious. |
|
in qualitative, what is "dependibility?"
|
tells of the integrety in qualitative studies.
refers to the stability of data over time and and over conditions. analogous to "reliability" in quantitative ex: the answers do not change in a week. If you get different answers a week later, they are not dependable. |
|
in qualitative, what is Confirmability?
|
The degree to which study results are from characteristics of participants, not from researcher bias.
|
|
what is bias?
|
An influence that produces a distortion or error in study’s findings; could affect the quality of evidence in qualitative and quantitative studies.
|
|
what kind of stuff about participants could elicit bias from a researcher?
|
Participants’ behaviors, self-disclosures
|
|
what kind of bias can occur on the rersearcher's end?
|
Researcher subjectivity – distorted information in direction of expectations, communicate expectations to participants.
|
|
in bias, what is a "sample imbalance?"
|
occurs when the sample comprises only people with the same belief, poor retention.
|
|
in what ways can bias screw up a study?
|
Faulty methods of data collection – instruments or tools
Inadequate study design Flawed implementation (like, for example, the people doing the intervention may not be doing it the right way.) |
|
how can randomness help prevent bias?
|
some participants may fail to provide accurate information because of fatigue at data collection time.
may get tired after the 5th question, so shake the questions up and move them around. |
|
what is systematic bias?
|
measuring tool not calibrated
a weighing scale might be measuring body weight 2 lbs. heavier. |
|
what is "Research Control" and give an example:
|
Holding constant other influences on the dependent variable so true relationship between independent and dependent variables can be understood.
ex: make sure the temp of the room or the color of the walls is consistant. |
|
give an example of an extraneous (confounding) variable:
|
lasix taken during a weight study.
It confounds the data. |
|
in terms of sample selection, when is the process adhering to "randomness"
|
when each participant has an equal chance at random to be in a study
|
|
what is "Masking" or "Blinding?"
|
concealing information from participants, ( or even data collectors, care providers) to enhance objectivity
ex: giving a placebo |
|
what is "Reflexivity" in Qualitative Research?
|
Critically reflecting on the self, analyzing and making note of personal values that could affect data collection and interpretation.
|
|
what is "Generalizability" and which research (qual or quan) does it apply to?
|
quantitative research
extent to which findings can be applied to other groups and settings |
|
what is "Transferability", and which research (qual or quan) does it apply to?
IMPORTANT |
qualitative research – findings can be transferred to other settings
influenced by the amount of information about the context of studies (thick description, a rich and thorough description of the research context ) |
|
what are some features of a research design?
|
Intervention
Control over extraneous variables Masking Comparisons – between-subjects; within-subjectsof the same group Location – selecting and gaining access Timeframes – cross-sectional; longitudinal |
|
what is a cross sectional design?
|
one in which data are collected at one point in time.
|
|
what is a longitudinal study?
|
datais collected at more than one point in time
|
|
describe retrospective time frame:
|
data is collected on an outcome occurring in the present, and linking it retrospectively to antecedents occurring in the past
|
|
describe prospective time frame:
|
information collected about a presumed cause then the effect is measured
|
|
what is a pilot?
aka? |
(aka feasibility) Studies
A small-scale version designed to test the methods to be used in a larger, more rigorous study e.g. participant recruitment, appropriateness and quality of instruments, identification of confounding variables to be controlled, potential problems – attrition, intervention. |
|
IMPORTANT:
what is an Experimental or Randomized Clinical Trial (RCT) Design charicterized by? |
Characterized by the following:
Manipulation – intervention is administered to some subjects and withheld from others Control – a group of subjects does not receive the intervention Randomization – subjects are randomly assigned to a control or experimental condition |
|
what is an "After-only" or "post-test only" experimental design?
|
data on dependent variables are collected only once – after random assignment is completed & experimental treatment has been introduced.
|
|
what is a "Before-after" or "pretest-posttest" experimental design?
|
It is an experimental design in which data are collected from subjects both before and after the intervention.
appropriate for measuring change |
|
what happens in "factorial" expenmental design?
|
two or more variables are being maniplated simultaneously, permitting a seperate analysis of he main efects of the independent variables.
ex: looking at both hr and bp in a morphine pain study |
|
what happend in a "crossover" design
|
exposure of the same subjects to more than one experimental treatment
ex: having subjects in an obesity study go from the dietitian to weight watchers |
|
what are 2 strengths for experiments?
|
they are most powerful for hypothesis of cause and effect relationships
They yield the highest quality of evidence regarding intervention effects. |
|
what are some limitations that experiments have?
|
Artificiality
they focus on some variables while holding others constant (use for paper.) Clinical studies – difficult to determine if experimental group received the treatment Hawthorne effect - you feel special so your response to the treatment is happiness |
|
what is a "Quasi-Experiment?"
|
An intervention in the absence of randomization.
|
|
what are some strengths of quasi-experiments?
|
they are practical
there is an increase in people who are not willing to relinquish control over their treatment condition |
|
what are some limitations to quasi-experiments?
|
the results are less conclusive
|
|
what are the 3 types of Nonexperimental rsearch?
|
correletaional
descriptive descriptive correlational |
|
what is corelational research?
|
an association between two variables is observed
|
|
what is descriptive research?
|
used to observe, describe, document aspects of a situation as it occurs
|
|
what is Descriptive correlational?
|
describes relationships among variables rather than to infer cause-and-effect relationships
|
|
what are some Strengths of Correlational Research?
|
It is an efficient way of collecting data about a problem e.g. health histories and eating habits.
It can be used in studies when an experimental design cannot be used. It is realistic for solving problems. |
|
what are some Limitations of Correlational Research?
|
It is weak in determining causal relationships.
It is susceptible to faulty interpretations because of working with preexisting, nonrandom groups who are self-selected |
|
what types of things do descriptive studies focus on?
|
prevalence, nature, intensity of health-related conditions and behaviors, critical in developing effective interventions.
|
|
how could info from a qualitative spawn a quantitative study?
|
Qualitative studies sometimes suggest a causal relationship which could be the focus of quantitative studies.
|
|
how could retrospective studies spawn prospetive studies?
|
Retrospective case-controlled studies may pave the way for prospective studies (best for examining causal relationships between prognosis, etiology and harm)
|
|
what are the 4 types of validity?
|
statistical conclsion validity
internal validity construct validity external validity |
|
describe Statistical conclusion validity:
|
there truly is an empirical relationship or correlation between the presumed cause and effect.
|
|
describe internal validity
|
It is validity of inferences:
given the existence of an empirical relationship, it is the independent variable that caused the outcome. |
|
describe construct validity:
|
inferences from aspects of a study (treatment, outcomes, people, settings) they represent.
|
|
describe xternal validity:
|
– inferences about observed relationships will hold over variations in persons, settings, time, or outcomes.
|
|
what are some things to help control confounding subject charicteristics?
|
randomization
crossover homogenity blocking or stratification matching statistical control |
|
what does randomization do to help in controlling confounding subject characteristics?
|
it equalizes groups (this is the most effective form)
|
|
what does crossover do to help in controlling confounding subject characteristics?
|
subjects serve as their own
|
|
what does homogenity do to help in controlling confounding subject characteristics?
|
you use only subjects who are homogeneous with regards to confounding variables, so findings would only be generalized to type of subjects who participated in the study.
|
|
where is blocking or stratification used in helping to control confounding subject characteristics?
|
commonly used in quasi-experimental or correlation studies. Random assignment to treatment conditions.
|
|
what is "matching" in controlling confounding subject characteristics?
|
using knowledge of subject characteristics to create comparable groups.
You get people who are similar. |
|
what is "Statistical control" in controlling confounding subject charateristics?
|
Analysis of covariance controls by statistically removing the effect of extraneous variables on the dependent variable
ex: removing the effect of age when examining the effect of training program on the heart rate of elderly persons. |
|
which one requires that the researcher knows the variables that need to be measured and controlled?
|
Homogeneity
blocking design analysis of covariance |
|
what are some things you look at in the subjects with implementation of treatment?
|
Extent to which intervention is similar from one subject to the next
Manipulation check - checking to make sure the way of treatment is correct Treatment adherence - check to see if the subjects are adhering |
|
what is internal validity?
|
Extent to which it is possible to make an inference that the independent variable (IV) is truly causing or influencing the dependent variable (DV)
make aure that the relation between the IV and DV is not the effect of a confounding variable. we must be vigilant when doing research! |
|
what are some threats to internal validity?
|
Temporal ambiguity - make sure that the cause preceeds the efect
Selection History - of pt had chf prior to a heart study Maturation - changes in subjects dou to age, maturation Mortality, attrition Testing and instrumentation - for example, the instrument is not calibrated, or different people use it differently. |
|
what aresome threats to construct validity?
|
Reactivity to study situations - some people all of a sudden feel "special."
Researcher expectancies - we may beso commitedthat we only want to see the good. Novelty effects Compensatory effects - sometimes the subjects may feel guilty of depriving the placebo group Treatment diffusion or contamination |
|
what is a good way to remember internal vs external validity?
|
internal - whithin the study
external - between different studies |
|
what are some enhancements to external validity?
|
Representativeness of exemplars (sample, setting)
Replication (multi-site) - to apply the findings to different settings. everything (instruments, etc..) must be the same. |
|
what are some threats to external validity?
|
Ways in which relationships might interact with or be moderated by variations in people, settings, time, and conditions.
Interactions between relationships and people. Interaction between causal effects and treatment variation. - if they think something is working they will work harder |
|
so basically, what do you do when you critiquing guidlines for study validity?
|
you make judgments about the extent threats to validity were minimized – or at least assessed and considered when interpreting results.
|