Reading...
Play button
Play button
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;
303 Cards in this Set
- Front
- Back
|
Future trends for nursing research?
|
Incr focus on EBP(translational research)
Stronger evidence base thru replication and multisites incr on systematic reviews incr in local, small research multidisc collaboration exp dissemination of findings inr visibility incr focus on cultural/disparities |
|
|
Define systematic reviews, importance?
|
A cornerstone of EBP-
level I of pyramid best practice clinical guidelines rely on reviews |
|
|
Inductive Reasoning
|
developing generalizations from specific observations
|
|
|
Deductive Reasoning
|
developing specific predictions from general principles
|
|
|
Why is reasoning in and of itself limited?
|
b/c the validity of reasoning depends on the accuracy of the information with which one starts
|
|
|
What is a paradigm?
|
world-view, general perspective on the complexities of the real world
|
|
|
Name 4 paradigms
|
Ontologic
Epistemologic Axiologic Methodologic |
|
|
Ontologic
|
what is the nature of reality?
|
|
|
Epistemologic
|
what is the relatship b/w the inquirer and that being studied?
|
|
|
Axiologic
|
what is the role of values in the inquiry?
|
|
|
Methodologic
|
how should the inquirer obtain knowledge?
|
|
|
2 broad paradigms of discipline nursing inquiry
|
Positivist
Naturalistic |
|
|
Desrcibe positivist paradigm
|
rooted 19th century thought
modernism- rational and scientific nature is ordered and regular, objective reality exists independent of human observation |
|
|
within positivist paradigm, research activity is directed at....
|
understanding the underlying causes of natural phenomenon
|
|
|
Post-positivists
|
belief in reality, but recognize the impossibility of total objectivity
|
|
|
Desctibe naturalistic paradigm
|
Constructivist- Weber and Kant
post-modern thinking of deconstruction and reconstruction reality is not a fixed reality, rather a construction of the individuals participating in the research |
|
|
In naturalistic paradigm, how is knowledge thought to be maximized?
|
when the distance b/w the inquirer and study participants is minimized
|
|
|
Quantitative paradigm
|
Positivist- deductive- verification of researchers' predictions- seeks generalizations- focus on product
|
|
|
Qaalitative paradigm
|
Naturalistic- reality is multiple and subjective- inductive- insider knowledge, researcher part of process- seeks understanding- focus on product and process
|
|
|
Purpose of nursing research
|
answer ?s, solve problems relevant to nrsg profession
|
|
|
Basic research
|
to extend the base of knowledge in a discipline-
Ex perform in-depth experiment to better understand grieving without explicit nrsg applications in mind |
|
|
Applied research
|
focus on finding solutions to existing problems- greater immed utility for EBP
|
|
|
Specific purposes of Nrsg R
|
Identification and description of phenomena, exploration, explanation, prediction and control
|
|
|
Quantitative descriptions focus on?
|
prevalence, incidence, size, and measurable attributes
|
|
|
Qualitative descriptions focus on?
|
dimensions, variations, importance of phenomena
|
|
|
Exploratory research?
|
in addition to describing/observing phenomena, it investigated full nature of phenomen, how manifested, related factors
|
|
|
Explanatory research?
|
understand the underpinnings of specific natural phenom, explain systematic rltionships, often linked to theories
|
|
|
What are key objectives in nrsg studies, typically quantitative?
|
prediction and control- aimed to control outcome vs understand why
|
|
|
Rsrch purposes linked to EBP?
(greater potential for EBP impact) |
categorized as types of info clinicians need:
treatment, therapy/intervention, diagnosis/assessment, prognosis, prevent harm, etiology, meaning and processes |
|
|
what is purpose of meaning and processes research?
|
undestanding client's perspectives on health and illness can help design effective interventions, engage in health promotions
|
|
|
Define EBP?
|
conscientious use of current best evidence in making clinical decisions about pt care-
|
|
|
Features of EBP?
|
de-emphasizes decisions based on custom, authority, opinion, or ritual- emphasis is on identifying best avail R and integrating it with other factors
|
|
|
Misconceptions of EBP?
|
advocated of EBP do not minimize the import of clinical expertise- goal is to integrate R with other factors
|
|
|
Key ingredient in EBP?
|
effort to personalize evidence to fit specific pts needs in spec clinical situtation
|
|
|
Critics concern of EBP in nursing?
|
worry that advantages are exag
clinical judgements devalued insuff attention to role of qualitative R |
|
|
Cochrane Collab
|
cornerstone of EBP
UK- Archie Cochrane call for efforts in 70s for R summaries of clinical trials to be avail |
|
|
EBP lack of consensus around what?
|
about what constitutes usable evidence for EBP- but findings from rigorous R are paramount- but what is "rigorous" and "best" evidence?
|
|
|
Level I of evidence hierarchy?
|
Systematic review of RCTs
Systematic review of non-random trials |
|
|
Level II evidence?
|
Single RCT, single non- RCT
|
|
|
Level III evidence?
|
Systematic review of correlational/observational studies
|
|
|
Level IV evidence?
|
single observational/correlational study
|
|
|
Level V evidence?
|
systematic review of descriptive/qualitative/physiologic studies
|
|
|
Level VII (final) evidence?
|
opinions of authorities, expert committees
|
|
|
Cons of evidence hierachy?
|
worth of evidence in level can vary considerably, not universal appropriate for certaiin research ?s,
|
|
|
What is "best evidence" then in nursing?
|
findings that are methodologically approp, rigorous, clinically relevant for ?s of efficact, safety, reliability, determinants of health, meaning of illness, nature of pts experiences
|
|
|
What enhances confidence in nrsg R?
|
compelling methods, confirmatory replication studies, systematically evaluated and synthesized evidenceh
|
|
|
Other sources for nrsg to answer clinical ?s when R limited?
|
benchmarking, chart review, QI and risk data, clinical expertise
|
|
|
A benefit of EBP movement when lack of evidence avail for clinical ?s
|
new resarch agenda can result
|
|
|
Best evidence in EBP states what about the evidence?
|
implies that most evidence on a clinical prob has been gathered, evaluated, and synthesized so that conclusions drawn about effective practices
|
|
|
"Bottom-line" of systematic review?
|
state-of-the-art summary of what the best evidence is at the time the review was written
|
|
|
Meta-analysis?
|
integrating quantitative R findings statistically
Redcues information! |
|
|
What is unit of analysis on meta-anal?
|
individual studies, NOT study participants
|
|
|
Meta-synthesis?
|
integrating qualitative R finings on a specific topic-
Amplifying and interpreting info! |
|
|
AHRQ?
|
another resource for systematic reviews
|
|
|
Preprocessed evidence hierarchy?
|
1. Clin practice guidelines
2. Synopses of system reviews 3. systematic reviews 4. synopses of single studies |
|
|
CATs?
|
critically appraised topics
quick summary of a clinical ? and appraisal of best evidence clinical bottom-line for practice- teaching tool for EBP |
|
|
Sources of pre-processed evidence?
|
CATs, clinical evidence.com, Evidence based nrsg journal,
|
|
|
Steps in EBP for clinician?
|
asking ? answerable with R
search/collect relevant evid appraise/synthesize evid integrate evid with own expertise, pt prefer, local context assess decision, intervention, advice |
|
|
Background clinical ?s
|
general, foundational, ?s about a clinical issue
ie/ what is cancer cachexia? found in textbooks |
|
|
Foreground clinical ?s
|
can be answered based on current best research evidence on diagnosing, assessing, treating or understaning meaning of their health problems/prognosis
The answer provides an opportunity for EBP! |
|
|
3 components of foreground, quantitative questions
|
population (characteristics)
intervention/exposure (of interest) outcomes (consequences of interest) |
|
|
3 components of qualitative R ?
|
population (characteristics)
situation (what conditions, experiences, circumstances are we interested in understanding) |
|
|
CURN project
|
Conduct and utilization of R in nursing project-
demonstrated that RU can be increased, but shed light on barriers to RU |
|
|
Barriers to RU
|
qual and nature of R
characteristics of nurses (Ed) organizational factors communication b/w R and clinicians |
|
|
Quality R is what?
|
addresses pressing clinical probs, replicated in a range of settings, methodologically strong (generalizability)
great collaboration b/w R and clinicians |
|
|
Appraisal of evidence does what?
|
considers validity and importance of findings, precision of estimates of effects, assoc costs/risks, clinical utility (part situations)
|
|
|
Integrating evidence is what?
|
pt preferences and values impt
avail of resources qual R can provide rich insights about how pts epxerience a problem, barriers to Tx compliance |
|
|
What are R concepts?
|
abstractions of aspects of human behavior- Ex pain, QOL, resilience
In Qual studies= phenomena |
|
|
What is a construct?
|
abstractions that are deliberately and systematically invented by Rs for a spec purpose
Ex- SELF-CARE |
|
|
What is a theory?
|
systematic, abstract explanation of some aspect of reality- concepts are knitted together to desrcibe some aspect of the world
Both QUAL and QUANT |
|
|
QUANT def of theory
|
framework or conceptual model, predictions R makes- R then tests theory, rsults support or reject or modify theory
|
|
|
QUAL def of theory
|
Sensitizing frameworks- provide an impetus for study
helps to guide the inquiry and interpret info OR theory is the product of research Goal is to develop a theory that explains phenom as they exist, not as pre-conceived |
|
|
Independent variable
|
the presumed cause
Does a nursing intervention cause improvements in pt outcomes? |
|
|
Dependent variable
|
the presumed effect- the outcome of interest! depends on variability in the independent variable- what is of interest in studying/learning about- what is R interested in understanding, predictiing, or explaining?
|
|
|
When are Rs interested in a SINGLE concept or phenomena?
|
descriptive studies
|
|
|
Rs are studying what more commonly?
|
relationships- bond or connection b/w phenomena- both QUAL AND QUANT examine relationships
|
|
|
QUANT studies and relationships
|
interest in the relationship b/w the independ and dependent variables- more than, less than, etc....
|
|
|
Quant studies and relationship ?s
|
Does rltship exist?
what is direction of rltship?how strong is the rltship? what is nature of rltnship b/w variables? |
|
|
QUAL studies and relationships
|
not concerning with quantifying or testing causal rltships-
rather seek patterns of association, illuminate underlying meaning |
|
|
Major Steps in a QUANT study
|
1. Conceptual phase
2. Design and planning phase 3. Empirical phase 4. Analytic phase 5. Dissemination phase |
|
|
I Conceptual Phase
|
strong intellectual element
identify good R ?s review literature clinical fieldwork- update theoretical framework/concepts formulate hypothesis |
|
|
What is a hypothesis?
|
statement of the R's expectations about relationships between study variables- predicted answer to R ?
|
|
|
What is research ? then
|
asks how the concepts under investigation might be related
|
|
|
II Design/Planning phase
|
research design selection (architectural backbone of study)
intervention protocol ID the population sampling plan- data collection plan- ethical principles/safeguards Review/finalize plan |
|
|
III Empirical phase
|
collect data
prepare data for analysis/coding |
|
|
IV Analytic phase
|
statistical analysis
interpretating results ?s answered |
|
|
V Dissemination phase
|
communicate findings-research report
utilizing findings in practice |
|
|
Critical reading of research reports
|
evaluate R's major conceptual and methodologic decisions
|
|
|
Evaluating research problems- 4 points?
|
Significance of R prob
Researchability of R prob Feasibility of adressing prob Own interest in R prob |
|
|
What problems/examples are NOT amenable to research?
|
?s of moral/ethical nature
?s based on person's values points of view Ex should assisted suicide be legalized? |
|
|
Properties of a good hypothesis?
|
clear and concise wording
present tense predictions about reltnships in a population, not just particular sample |
|
|
Simple hypothesis
|
relationship b/w 1 IV and 1 DV
Ex "Children whose mothers participate in a nurse case management intervention for women exposed to violence (IV) will have fewer behavior probs (DV) than chilren whose mothers receive routine screening only." |
|
|
Complex hypothesis
|
prediction relationship b/w 2 or more IVs and/or 2 or more DVs
|
|
|
Independent variable
|
presumed cause, antecedent, precondition
|
|
|
Dependent variable
|
predicted effect, outcome, or consequence of IV
|
|
|
what type of studies are complex hypotheses common?
|
Studies that try to asssess the impact of a nursing intervention on multiple outcomes
|
|
|
Are hypotheses derived from theory more like directional or non-directional? why?
|
directional- b/c theories explain phenomena, thus providing a rationale for expecting variables to be related in certain ways
|
|
|
When are non-directional hypotheses more appropriate?
|
when there is no theory or related research, when related study findings are contradictory, when R's own experience leads to ambivalence
|
|
|
Cons of directional hypotheses?
|
R's may be intell committed to certain outcomes- may lead to bias
|
|
|
Pros directional hypotheses?
|
1. they clarify study's framework and demonstrate R's have thought critically about phenomena under study
2. may permit a more sensitive statistical test |
|
|
research hypotheses also known as ?
|
substantive, declarative, scientific
|
|
|
Null hypotheses also known as?
|
statistical hypotheses- logic of statistical inference
|
|
|
Null hypothesis defintion?
|
expressed as the expected absence of a relationship b/w IV and DVs
|
|
|
Example of null hypothesis?
|
"Pts age is unrelated to their risk for falling
"Older pts are just as likely as younger pts to fall" |
|
|
hypotheses are never _____, rather they are ______.
|
proved
accepted or supported |
|
|
central ? of written research review
|
What is the current state of evidence on this research problem?
|
|
|
steps in preparing a written research review
|
formulate ?
devising a search strategy conducting search retrieving relevant sources abstracting/encoding info critiquing studies analyzing the aggregated info preparing a written synthesis |
|
|
what type of sources should lit reviews be based on?
|
PRIMARY
|
|
|
ancestry approach for finding studies
|
tracking down earlier studies cited in a reference list of a report
|
|
|
descendancy approach to finding sources
|
using a pivotal study to search forward to subsequent studies that cited it
|
|
|
the term theory connotes ______?
|
an abstraction
use theory to refer to an abstract generalization that offers a systematic explanation about how phenomena are interrelated |
|
|
traditional definition of theory?
|
requires a theory to embody at least 2 concepts that are related in a manner that the theory purports to explain
|
|
|
less traditional def of theory?
|
less restrictive- refer to a broad characterization of a phenomenon
|
|
|
Descriptive theory?
|
account for/thoroughly desrcibe a single phenomenon;
summarize commonalities found in discrete observations |
|
|
Descriptive theory plays an impt role in _______ studies?
|
Qualitative
|
|
|
What are concepts?
|
basic bldg blocks of a theory
|
|
|
Logically interrelated deductive system=
|
relationships theory explores provides a mechanism for logically arriviing at new statements from the original propositions
|
|
|
Ex of these propositions
|
"is associated with" "varies directly with" "is contngent on"
|
|
|
Framework is what?
|
the overall conceptual underpinnings of a study- every study has a framework! - a theoretical rationale
|
|
|
In most nursing studies, the framework is not ______, rather it is ______.
|
Not an explicit theory or conceptual model;
framework is implicit,without being formally acknowledged or described |
|
|
In studies that lack a _____, it may be tough to figure out what the Rs thought was going on and why
|
conceptual framework
QUANTs are more guilty of this |
|
|
In most QUAL studies, conceptual frameworks are __________?
|
part of the research tradition in which the study is embedded
|
|
|
Ethnographers begin their theory groundwork _______?
|
within a theory of culture
|
|
|
Grounded theory researchers incorporate ________ into their framework and approach to look at phenomena
|
sociologic principles
|
|
|
?s asked and methods used to address ?s by QAUL Rs _________
|
reflect certain theoretical formulations
|
|
|
Theories and conceptual models cannot be ________?
|
proved
|
|
|
A theory is a scientist's ___________?
|
best effort to describe and explain phenomena
|
|
|
How are theories discredited/revised?
|
- new evidence or observations
- a new theory might integrate new observations with an existing theory to yield a more parsimonious or accurate explanation of a phenomenon |
|
|
Link between theories and ______ may cause supporters to be lost and theory falls into disfavor
|
values!
heories are invented by humans, values and ideals can change over time |
|
|
Theories and models are never considered _________?
|
final verified
always possibility that theory will be modified or discarded |
|
|
Theories are an impt resource for the development of ________?
|
nursing intervetions
provide a basis for predicting the occurrence of a phenomenon |
|
|
Grounded theories
|
QUAL Rs
data-driven explanation to account phenomena under study through inductive processes |
|
|
Nazi medical experiments
|
1930-40s, use of prisoners of war in testing the limits of human endurance and reaction to diesases/drugs
SUBJECTS COULD NOT REFUSE! |
|
|
Tuskegee Syphilis study
|
1932-1972 sponsored by US Public HS, 400 poor African-american men- Tx deliberately withheld to study course of disease
Public Health Nurse recruited |
|
|
Jewish Chronic Disease Hospital
|
1960s- elderly
injection of live cancer cells |
|
|
Willowbrook Study
|
researching Hepatitis on mentally retarded children- deliberately infected with virus
|
|
|
Nuremberg Code
|
1st intern'l recognized code of ethical standards- response to the Nazi trials
|
|
|
Declaration of Helsinki
|
Intl' 1964 by WMA- revised in 2000
|
|
|
Disciplines have devised own code of ethics
|
APA psychologist guidelines
ASA Code revised 1999 AMA updated recently |
|
|
Nursing ethics devised by
|
ANA- 1968 "The Nurse in research..."
1975- "Human rights guidelines for nurses in research..." Intl't Council of Nurses- Code of ehtics updated 2000 |
|
|
Conflict of interest for Nurse Researchers
|
expected behavior as nurses comes into conflict with the expected behaviour of Rs (ie deviating from a R protcol to assist a pt in need)
|
|
|
Beneficence
|
most fundamental
minimize harm, maximize benfits |
|
|
Benificence principles
|
Right to freedom from harm and discomfort(physical and psychological!)
Right to protection from exploitation |
|
|
Respect for Human Dignity
|
Self-determination right, freedoom from coercion
Right to full disclosure |
|
|
3 primary principls of Belmont Report
|
Beneficence
Respect for Human Dignity Justice |
|
|
Justice
|
Right to fair treatment
Right to Privacy |
|
|
Vulnerable groups?
|
may be incapable of fully informed consent (ie MR)
OR high-risk for side effects (ie pregnancy) |
|
|
Who are vulnerable groups?
|
Children
Mental or emotional disabled Severely ill or physically disable Terminally ill Institutionalized people Pregnant women |
|
|
Reliability refers to what?
|
accuracy and consistency of info- associated with the methods used - ie thermometer faulty readings
|
|
|
Statistical reliability refers to what?
|
probability that same results would be obtained with a completely new sample- reflective of a wider group
|
|
|
Validity of a study?
|
soundness of study's evidence
unbiased, cogent, well-grounded findings- quality of evidence regarding the effect of the IV on the DV |
|
|
what do Rs strive for?
|
solid conceptual definitons of R variables and valid methods to operationalize them
|
|
|
Triangulation enhances what?
|
credbility in a QUAL or QUANT study
|
|
|
what is triangulation?
|
use of multiple resources to draw conclusions constituting the truth-
|
|
|
Ex of triangulation in a QUAL study?
|
multiple means of data collection
Ex in-depth discussions with study participants as well as watching their behaviors in natural settings |
|
|
Ex of triangulation in a QUANT study?
|
multiple operational definitions of a DV to determine whether predicted effects are consistent
|
|
|
Integration of both QUANT and QUAL data in a single study
|
triangulate across paradigms
|
|
|
What is bias in general?
|
an influence that produces a distortion or error in the study results
|
|
|
what are some causes of bias?
|
participants mispresenting selves
R subjectivity sample imbalances faulty methods of data collection inadequate study design flawed implementation despite well-designed |
|
|
Random bias
|
fail to provide accurate info due to fatigue at time of data collection
|
|
|
Systematic bias
|
bias is consistent or uniform
Ex faulty scale measureing weights |
|
|
Minimize systematic bias to strengthen ________?
|
rigor
|
|
|
One approach to minimize systematic bias?
|
triangulation- multiple sources of info can help counterbalance biases or identify them
|
|
|
Masking
|
used in QUANT studies to prevent awareness biases
concealing info from participants or research agents |
|
|
Open vs closed study
|
masked vs unmaskedb
|
|
|
Generalizability criterion for what studies?
|
QUANT only
|
|
|
How do Rs enhance generalizability?
|
design studies strong in validity and reliability
little point in wondering whether results are generable if they are not accurate or valid! |
|
|
Comparison with other studies
|
study structure that supplements comparison types- ie between 2 groups, comaparison of one group's status at 2 or more points in time
|
|
|
Cross-sectional designs
|
collection of data once!
describe status of a phenomena or relatships among phenomena in a fixed point in time |
|
|
Longitudinal designs
|
data are collected at more than one point in time over an extended period
enhance research control- used by QUANT Rs |
|
|
Data collection in longitudinal designs?
|
collect from one group multiple times, OR different samples
Ex Trend studies |
|
|
Panel studies
|
type of longitudinal
same people are used to supply data at 2 or more points in time- face problems of attrition |
|
|
Retrospective research design?
|
collecting data on an outcome occurring in present and linking it to antecedents or determinants occurring in the past
|
|
|
Prospeective research design?
|
iinfo is collected about a presumed cause or antecedent and then subsequently the effect or outcome is measured
EX intervention studies yield better-quality evidence |
|
|
broad categories of R ?s that are relevant to EBP in nursing?
|
?s of interventions, DX assessments, prognosis, harm, meaning and processes
|
|
|
Counterfactual
|
what would have happened to the same people exposed to a causal factor if they simultaneously were not exposed to the causal factor
|
|
|
criteria for causality?
|
Temporal- cause must precede effect in time
Empirical relationship Cannot be explained as caused by a third variable |
|
|
Experiment, true experiment, experimental study same as ?
|
Randomized control trial, randomized clinical trial
|
|
|
Quasi-experiment, quasi-experimental same as?
|
controlled trial without randomization
|
|
|
Non-experimental, correlational same as?
|
Observational
|
|
|
Retrospective study same as?
|
case-control study
|
|
|
prospective non-experimental study same as?
|
cohort study
|
|
|
group under a condition (experimental or control group/condition) same as?
|
Arm (intervention or control arm)
|
|
|
Researchers are _____ in an experimental, RCT?
|
active agents
NOT passive observers! |
|
|
true RCT/experimental has what properties? 3
|
manipulation (intervention)
control (counterfactual) randomization (assigned to control or experimental randomly) |
|
|
Manipulation affects what variable? How?
|
manipulates the IV by giving TX to some subjects and not others- then observes varied IV effect on DV
|
|
|
The counterfactual is the ________?
|
control group-
|
|
|
How is systematic bias minimized or eliminated?
|
randomization- random assignment
|
|
|
Randomization does not guarantee ________?
|
that the groups will be equal
|
|
|
Cluster randomization
|
randomly assigning clusters of individuals to diff TX groups vs individuals- may enhance R feasibility
|
|
|
Stratification
|
subgroups of study participants are allocated equally to TX conditions
|
|
|
After-only, posttet-only designs
|
experimental design- data on V variables are collected only once- after randomization and intervention
|
|
|
Before-after design, pretest-posttest design
|
collection of baseline data and data collection at multiple post-intervention points- EX cooling blanket
|
|
|
Cross-over design
|
within groups- sujects are exposed to all treatments, but are randomly assigned to diff orderings of treatments- subjects serve as own controls
|
|
|
Randomized block design
|
between groups- random assignment to groups within diff levels of a blocking variable that is not under experimental control EX gender
|
|
|
Main effects
|
effects resulting from experimentally manipulated variables
|
|
|
Interaction effects
|
effects resulting from combining treatments
|
|
|
Factorial experiments
|
subjects are assigned at random to a specific combination of conditions- 4 cells
|
|
|
Factors in a factorial design are _______?
|
Two IVs
|
|
|
Randomized block design?
|
similar to fatorial- Two factors (IVs) BUT one factor is not experimentally manipulated- the 2nd factor is a stratifying variable
different subjects and diff treatments |
|
|
In a randomized block design, the variable which you cannot manipulate (ie gender) is called what?
|
blocking or stratifying variable
|
|
|
What type of effets can be examined in factorial and randomized blocking designs?
|
interaction effects
|
|
|
Mininum # of subjects per blocking or factorial cell?
|
20
so 2 x 2 design = 80 subjects minimum |
|
|
Cross-over design
|
exposure of same subjects to more than one experimental treatment- within-subjects design
|
|
|
Cross-over design problems?
|
carry-over effects- may be influenced in 2nd condition by their experience in the 1st condition
Drug studies rarely use this design |
|
|
Experimental limitations?
|
artificiality
focus on only a handful of variables while holding all else constant (reductionist) Hawthorne effect |
|
|
hawthorne effect?
|
placebo effect caused by people's expectations
|
|
|
Quasi-experiments
|
intervention without randomization
not as powerful in identifying causal connections b/w interventions and outcomes |
|
|
most frequently used quasi-experimental design?
|
non-equivalent control group before-after design
experimental and comparison groups are not equivalent |
|
|
Quasi-experimental befor-efter design strength?
|
baseline data exists to determine whether pts in 2 hospitals had similar satisfaction initially
|
|
|
In quasi-experimental design the control group is referred to as _______?
|
comparison group- the grouo against which outcomes in the treatment group are evaluated
|
|
|
Time series design
|
no comparison group
information on the DV is collected over a period of time before and after the treatment- used in single-subject experiments |
|
|
disadvantage of time series design
|
absence of a comparison group-
far from yielding an ideal counterfactual |
|
|
A good variation of time-series design would be what?
|
combo of time series and non-equivalent control group (quasi-experimental)
|
|
|
strength of quasi-experimental design?
|
practical- some R control when full exp rigor is not possible-
acceptable to more people- ie pple do not want to give up their TX |
|
|
Non-experimental, observational R design =
|
when R do not intervene by manipulatiing the IV (ie birth wt)
|
|
|
Most nursing studies are ______?
|
non-experimental!
|
|
|
correlational research
|
examine relationships b/w variables when cannot manipulate potential cause
|
|
|
Correlation def and ex
|
assoc b/w 2 variables, a tendency for variation in one variable to be related to variation in another EX Ht and Wt
|
|
|
Ex of correlational research design
|
Retrospective (case-control studies)
Prospective non-experimental (cohort design) |
|
|
Weaknesses of correlational research
|
weak ability to reveal causal relationships
selection bias- preexisting groups |
|
|
Strengths correlational research
|
explain problems not amenable to experiementation
efficient means to collect a large amt of data look at many variables at once undertaken in the next phase of developing an evidence base for a causal connection |
|
|
Validity is a matter of d_____, not a______
|
degree vs absolute
|
|
|
Statistical conslusion validity
|
validity of inferences that there truly is an empirical relationship, or correlation b/w the presumed cause and effect
|
|
|
Internal validity
|
the validity that it IS the IV rather than other factors that caused the outcome
R's job to rule out plausibility that something other than presumed cause can account for the observed relationship |
|
|
Construct validity
|
whether the measures of the DV are good operationalizations of the constructs for which they are intended
|
|
|
External validity
|
the generalizability of causal inferences
critical concern for EBP! |
|
|
Threats to validity are -
|
reasons that an inference could be wrong
|
|
|
controlling intrinsic External validity
|
randomization
cross-over (subjects serve as own control) Homogeneity Blocking/stratification Matching |
|
|
problems to matching
|
Rs must know in advance what relevant confounding variables are- cumbersome with more than 2 variables to match
|
|
|
statistical power
|
the ability to detect true relationships among variables
large sample increases stat power!! |
|
|
Statistically, results are clearer when differences b/w groups or conditions being compared are large or small?
|
LARGE!
|
|
|
Precision
|
maximizing stat power
accurate (reliable) measuring tools, controls over extraneous variables, powerful stat methods |
|
|
Intervention fidelity (treatimplment fildelity)
|
extent to which the implementation of an intervention is faithful to its plan- strength of an intervention can be undermined if the intervetion is not as powerful in reality as it is on paper
|
|
|
Threats to internal validity
|
1. temporal ambiguity (diff to determine in correlational studies what came first- chix vs egg?)
2. selection- nonrandomized or self-selection of subjects into no TX group 3. history (external events taking place concurrently with IV that can affect DV 4. maturation- passage of time, subjects change 5. mortality/attrition 6. testing/instrumentation (effects of taking a pre-test on posttest performance) |
|
|
Studies at high threat to internal validity
|
Quasi-experimental, correlational
|
|
|
Example of threat to construct validity
|
hawthorne effect (reactivity to the study situation)
Use masking to eliminate, outcome measures less susceptible to reactivity (ie hospital records) |
|
|
Impt of construct validity
|
constructs are the means for linking the operations used in study to mechanisms for translating the resulting evidence into practice
|
|
|
Interaction effects in a factorial design?
|
when 2 TX are simulataneously manipulated- ? is whether the effects of TX A hold (are comparable) for all levels of TX B?
|
|
|
Threats to External Validity
(relationship moderators) |
1. Interactions b/w relationships and people
2. Interaction b/w causal effects and TX variation- diff results obtained from same TX b/c of who is administering TX |
|
|
Practical/Pragmatic clinical trials
|
attempt to maximize external validity with smallest negative effect on internal validity
|
|
|
Rationale for MIXED METHOD STUDIES
|
triangulation of QUANT and QUAL data
1 complementarity 2 incrmentality 3 enhanced validity arriving at covergence in tapping a construct |
|
|
Pragmatism
|
it is the R ? that should drive the inquiry, the ? is more impt than the methods used
|
|
|
Clincial Trials
|
studies designed to assess clinical interventions- methods are medical research based, many nurse R's are using vocab
|
|
|
Phase I Clinical Trial
|
after develop of drug, to establish safety and tolerance and optimal dose-
small-scale, without a control group |
|
|
Phase II Clinical Trial
|
seeking prelim evidence of efficacy of TX as designed in Phase I- ascertain feasibility of more rigorous testing- look for signs of poss sidefx- refinements needed
|
|
|
Phase III Clintcal Trial
|
full experimental test- RCT-
determine efficacy- large, heterogenous subjects- |
|
|
Phase IV Clinical Trial
|
effectiveness of TX in general population- emphasis on external/internal validity-tightly controlled- risks relevance
|
|
|
Research that does not involve interventions
|
Outcomes research, survey research, secondary analysis,
|
|
|
Outcomes Research
|
designed to document the effectiveness of health care services/nrsg and plays role in policy research- global asssessment nrsg
|
|
|
Outcomes research qualities
|
complex and multidisciplinary
evolving area |
|
|
Survey Research
|
prevalence, distribution, interrelations of variables within a population
|
|
|
Secondary Analyses
|
use of data gathered in a previous study to test new hypotheses or explore new relationships- collect more data than analyzed- efficient and economical
|
|
|
QUANT secondary analysis
|
variables and relationships among variables previously unanalyzed may be examined- DV could become IV and visa versa
|
|
|
QUAL secondary analysis
|
to exploit rich data sets- more voluminous
|
|
|
types of QUAL secondary analysis
|
analytic expression
retrospective interpretation cross-validation |
|
|
Sampling
|
process of selecting a proportion of the population to represent the entire population so inferences about population can be made
|
|
|
a sample
|
subset of population elements
|
|
|
an element
|
most basic unit about which information is collected (ie humans)
|
|
|
Sampling Frame
|
through which resources can you access them- your accessible population
|
|
|
Key consideration in assessing a sample in a QUANT study is ____________?
|
Representativeness
|
|
|
Probability sampling
|
random selection of elements
|
|
|
Non-probability sampling
|
Non-random selection of elements- every element usually does not have a chance of inclusion!
|
|
|
Strata
|
mutually exclusive segment of a population EX- gender, age group
|
|
|
Sampling bias
|
systematic over or under representation of some segment of the population in terms of a characteristic relevant to the R ?
|
|
|
Sampling bias is a function of population
|
homogeneity
|
|
|
One straightforward way to increase generalizability of a study is to select study participants from ________?
|
multiple sites- hospitals, nrsg homes, communities
|
|
|
3 methods of Non-probability sampling
|
1. Convenience sampling worst(Snowball sampling)
2. Quota sampling 3. Purposive sampling |
|
|
Purposive Sampling
|
judgmental- based on R's knowledge about the pop can be used to hand-pick sample members- subjective, when want sample of experts
|
|
|
Quota sampling
|
divides population into homogenous strat to ensure representation of subgroups in the sample; within each stratum, sujects are sampled by convenience!
|
|
|
Systematic sampling
|
selection of every nth case from a list- essentially a sample of convenience-
ccan be applied to lists that have been stratified |
|
|
Probability sampling
|
all elements in population have equal probability of being selected
|
|
|
Types of probability sampling
|
1. Simple random sampling- sampling frame
2. stratified random sampling 3. cluster sampling 4. systematic sampling |
|
|
QUAL sampling style?
|
Purposive- may begin with volunteer informants and supplemented with new ones via snowballing- eventually evolve to purposive-selecting cases that benefit the study
|
|
|
Types of purposive sampling
MV HS EC IS |
Maximum variation
Homogenous sampling Exteme deviant case sampling Intensity sampling |
|
|
Phenomenological sampling
|
small, 10 or less
all participants must have lived experience |
|
|
Grounded Theory sampling
|
20-30 pple
goal is to select informants who can best contribute to the evolving theory |
|
|
Ethnography sampling
|
initially a "big net" approach
mingle with as many members of culture as possible 40s- rely on key informants of culture have to decide on whom AND what to sample |
|
|
Another name for convenience sampling
|
accidental
|
|
|
Cluster sampling
|
multistage sampling- successive selection of random samples from larger to smaller units by simple random or stratified random methods
|
|
|
Rigor
|
how much I can rely on of results- influenced by validity of study
|
|
|
What effects Rigor?
|
validity
|
|
|
Validity
|
how well-founded, defensible is inference made from rsrch findings?
|
|
|
Internal validity
|
confidence in results- how confident can say IV influenced DV?
|
|
|
External validity
|
generalizability
|
|
|
Solomon 4 group
|
4 groups, all randomized- main purpose is to control
2 control groups- 1 obs before 2 TX groups - 1 obs before |
|
|
Factorial design
|
many diff IVs, interaction effects!
EX Factor A type of exercise Factor B time of exercise DV= pulse |
|
|
Why is repeated measures design chosen?
|
R doesn't have money or time for mult groups
|
|
|
Problem with repeated measures?
|
cross-over effect/carry-over
Ex GRE test preps |
|
|
Quasi-experimental studies-
3 types? |
1. non-equivalent control group (pre-experimenal)
2. Time series design 3. Times series non-equivalent control group |
|
|
Quasi experiement good and bad?
|
practical- partial control
BUT cause n effect less clear- internal validity threatened |
|
|
Again, what is Indepen Variable?
|
what MAY influence dependent, what researcher can manipulate around
|
|
|
Steps in a QUANT design?
4 |
1. conceptual
2. design and planning 3. empirical 4. dissemination |
|
|
Steps in QUAL design?
|
Less linear
planning, data collection stratagies, gathering and analyzing data, disseminating |
|
|
QUANT studies statements of purpose
|
ID variables, population, suggests nature of inquiry through verbs
Ex To test, to compare, to evaluate "relationship" term= expect to see correlation, multiple regression etc.... |
|
|
QUAL statements of purpose
|
"this study intends" "this study reports"
IDs central phenomenon, ids research theory, indicates group of interest, suggests nature of inquiry thru verbs |
|
|
"analyze the relationship" indicates what type of research/design?
|
QUANT, observational
|
|
|
all extraneous variables are m_____ and m_______?
|
mediating, moderating
|
|
|
Moderator variables
|
directly affect strength/direction
EX stress |
|
|
Mediating variables?
|
interven b/w DV and IV
|
|
|
Grounded theory R?s are
|
process ?s
|
|
|
Phenomenology R?s are
|
meaning ?s
|
|
|
Etnography R ?s are
|
cultural description ?s
|
