Download NSG 6691 Final Review: Key Concepts in Nursing Research and Methodology and more Exams Nursing in PDF only on Docsity!
NSG 6691 FINAL REVIEW EXAM QUESTIONS WITH
ACCURATE ANSWERS
Levels of evidence (highest to lowest too) - ANSWER 1. systematic review or meta-analysis of randomized controlled trials (RTC) Evidence-based clinical practice guidelines based on systematic reviews
- well-designed RCT
- Controlled trial without randomization (quasi-experimental study)
- single nonexperimental study (case-control, correlation, cohort studies)
- systematic reviews of descriptive or quaLitative studies
- Single description or quaLitative study
- opinion of authorities and/or reports of expert committees
critical appraisal - ANSWER appraisal by a nurse who is a knowledgeable consumer of research, and who can appraise research evidence and use existing standards to determine the merit and readiness of research use in clinical practice.
(To use research you have to understand how it works.)
PICOT (name each letter and define) - ANSWER -Population (subjects participating)
-Intervention (the portion of health care that is of interest: therapeutic, diagnostic, prevention, organizational.)
-Comparison (standard of care, non-interventional)
-Outcome (effective outcome)
-Time
Independent Variable - ANSWER manipulated item, X, has or causes the effect, thing that is changed to get different results
Dependent Variable - ANSWER not manipulated, Y, the result of the changes, the outcome, the way researchers explain
9 - ANSWER How many steps in conducting a literature review
- determine question
- identify key terms
- conduct computer research on data bases (min. 2)
- review abstracts and narrow down
- retrieve relevant source
- store or print relevant sources
- preliminary readings of these sources
- critically read sources
- synthesize critical summaries - ANSWER name 9 steps of literature review
Internal validity - ANSWER degree to which it can be inferred that the experimental treatment resulted in the observed effects (rather than uncontrolled condition)
other factors of what could have caused results are ruled out
Cronbach's alpha - ANSWER test of internal consistency
operational definition - ANSWER what instruments or how a researcher measures a variable
level of significance - ANSWER "p value" or "alpha value"
level of significance - ANSWER probability of making a type 1 error
(min level .05, -usual, can be .01 if really confident in your research) setting at. increases risk of type 2 error
qualitative research design - ANSWER helps understand a phenomena; explanatory, descriptive, and inductive in nature; words not #'s to explain a phenomena. Includes: review of lit, study design, study setting and sample, approaches for data collection and analysis, study findings, and conclusions with implications for practice and research. Works to understand human health experiences. 4 kinds: phenomenology, grounded theory, ethnography, and case study
quantitative research design - ANSWER provides a blue print, way to systematically test research questions and hypotheses,
gives a way to remain in control of a study; involves independent and dependent variables and manipulation of these;
must have concern for internal and external validity. Tests relationships, differences, and cause and effect interactions among and between variables that are tested with hypotheses or research questions
correlational study - ANSWER type of nonexperimental research design that examines the relationship between two or more variables. (pg. 204) Not testing cause and effect; tests if one changes, will the other also have a change, studies the relationship strength
assumed causes then moves forward in time to the presumed effect. (pg. 206) same as longitudinal. Threats are internal validity or bias r/t mortality, instrumentation, and testing
randomized controlled trial - ANSWER a research study using a true experimental design. (pg. 184) may be call "clinical trial". Considered GOLD STANDARD for providing information about cause-and-effect relationships. Level 2 evidence. These ARE randomized, controlled, and manipulated
retrospective study - ANSWER nonexperimental research design that begins with the phenomena of interest (DV) in the present and examines its relationship to another variable (IV) in the past. (pg. 207) like expost facto. DV has already been affected my IV. Study done after the fact. Research hypothesizes results. Increased confidence in these results bc of higher level of control
systematic review - ANSWER process where investigators find all relevant studies, published and unpublished, on topic or question, at least two members of the review team independently assess the quality of each study, include or exclude studies based on pre-established criteria, statistically combine the results of individual studies, and present a balanced and impartial evidence summary of the findings that represents a "state of the science" conclusion about the evidence supporting benefits and risk of a given health care practice. (pg. 21) Summation and eval of group of research studies that test a similar research question, if more than on study called meta-analysis. (pg.220) Sum of quant studies with similar research questions/ inclusion and exclusion criteria
secondary analysis - ANSWER a form of research in which the research takes the previously collected and analyzed data and reanalyzes the data for a secondary purpose. (pg.212) Can experimental or nonexperimental. Cannot ne used to find causality but can provide information. (pg. 283) uses this type of research to study records or data to figure out problems that need solving
non-probability - ANSWER Which samplings lacks randomization
Convenience sampling - ANSWER (nonprob) easy to find sample, HIGH bias risk, questionable representation of sample, uses most available persons for research
quota sampling - ANSWER (nonprob) moderately easy to find sample, unknown source of bias, that affects external validity, somewhat representative of population of interest. (pg.238) a percentage represents a whole
purposive sampling - ANSWER (nonprob) moderately easy to find sample, bias increased, heterogeneity of population, conscious bias, lacks ability to represent the whole sample bc sample not handpicked. (Pg. 238) subjects are chosen bc they are thought to be typical of the population. Used to study a highly unusual group (ex: rare disease)
Simple random sampling - ANSWER (prob) BEST rep of sample, time consuming to find sample, careful and controlled process, must meet certain criteria
stratified random sampling - ANSWER time consuming to find sample, low bias, good rep of sample. (pg. 241) population is divided into strata or subgroups, then divided into homogeneous, then number of elements from each subset are randomly selected on the basis of their proportion in the population. Stratified by age, gender, religion, etc
multistage or cluster sampling - ANSWER (prob) less or more depending on strata, as risk for more bias that other 2, less rep of sample than others. (pg. 242) Take samples from a large group and make smaller groups in a second sampling stage
type 1 error - ANSWER rejects null when it is actually true
type 1 error - ANSWER accepts there is a difference when there is actually no difference btw groups, results are not significant when you think they are
type 1 error - ANSWER false postitive
type 2 error - ANSWER accepts null when it is actually false
informed consent - ANSWER voluntary participation; tells of risks and benefits, must obtain before research conducted or results uses if prospective studies; 8th grade reading level
benefience - ANSWER benefits outweigh risks; obligation to act to benefit others and to maximize possible benefits
justice - ANSWER fairness; principle that human subjects should be treated fairly
respect for persons - ANSWER the principle that persons have the right to self-determination and to treatment as autonomous agents, they have the freedom to participate or not participate in a study
confidentiality - ANSWER assurance that a research participant's identity cannot be linked to the information that was provided to the researcher. Breaches of this that are harmful to subject are: sexual preference, income, drug use, prejudice, or personality variables. Data analyzed as group info so confidentiality is not breached
generalizability - ANSWER inferences that data are representative of similar phenomena in a population beyond the studies sample
3 types of non-probability sampling - ANSWER Convenience, quota, purposive
3 types of probability sampling - ANSWER simple random, stratified random, multistage or cluster
