Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
research
midterm
| Question | Answer |
|---|---|
| Definition of nursing research | Sysm inquiry to answer a questions about nursing |
| Definition of evidence-based practice | Taking best evidence w pt. prefereces and values |
| What is the impact of nursing research vs. the impact of EBP? | To generate new knowledge, take it, and translate it into practice. |
| Deductive vs. inductive reasoning | Deductive: top-down approach Inductive: bottom-up approach |
| What is a PICO question? | Elements of a good researchable question. |
| Know how to identify each component of the PICO question | Population, intervention/exposure, comparison/control, outcome, sometimes T (time). |
| What is sampling? | Subset of population to conduct study on |
| Why is sampling important? | we want it to be representative |
| Target population | who we want to apply to |
| Accessible population | who we have access to |
| sample population | the actual individuals we experiment on |
| sampling error | the difference between sample statistic and true population parameter (ultimately what we want to know) |
| what affects sampling error | sample size- inc size= dec error heterogeneity- variance. as it goes up, error increases |
| point estimate | single best guess, dont know how precise |
| how do we know how precise our point estimate is | how wide the confidence interval is tells us how precise, we want it more narrow |
| if two intervals dont over lap? | we can be confident they're different |
| if two intervals overlap? | we can be confident they're not different |
| data collection methods | Biophys. elements self-report and surveys |
| what is reliability | consistency |
| what is validity | accuracy, true miss |
| types of validity | content, criterion, construct |
| content validity | does the content match the construst were trying to get |
| criterion validity | correlation |
| construct validity | is our measurement associated with things it should be associated with |
| types of bias from structures observation and surverys | social desirability, recall bias, response bias, aquestant bias, extreme bias |
| social desireablility | changing behavior that is more socially acceptable |
| recall bias | remembering about the past (look this up) |
| response bias | are the people who responded to the survey different than the people who did not |
| aquesant bias | they want to agree w everything |
| extreme bias | everything really great or everything terrible |
| what is a study design | guides our research process (the blue print) |
| quantitative vs qualitative | numbers vs non-numeric |
| hierarchy of evidence | arranges interval validity can any effect be attributed to the independent variable, is the cause the effect? |
| primary hierarchy of evidence | conducting the research for the OG research purpose |
| what primary study design generates the highest level of interval validity? | true experimental design |
| secondary hierarchy of evidence | using info from existing studies |
| Three required properties of true experimental design | Only generate high level of evident if they are designed well |
| definition of binding | concealment of what group a participant is in from as many parties of the experiment as possible |
| Definition of allocation concealment | Hiding the sorting of upcoming assignments so info can’t be exploited |
| definition of intervention fidelity | Following the study protocol, treating everyone the same |
| Definition of intention to treat analysis | Analyzing participants in the group they were assigned to no matter how much they adhered to the protocol |
| independent variables | cause, the predictor |
| dependent variables | the outcome, the effect |
| What makes a quasi-experimental design different than an experimental design? | Quasi don’t have randomization, sorting into groups some other way. Without randomization we may have non-equivalent groups and then worry about confounding |
| definition of confounding | Anything that affects the independent or dependent variables in anyway |
| definition of bias | any deviation from the truth that can cause false conclusions |
| Strengths and weakness of quasi-experimental designs | More practical Less expensive More generalizable Main limitation: cant test the cause and effect relationship |
| Difference between observational and experimental/quasi-experimental designs | There’s no active manipulation with In quasi, we tell them what Tx they’re going to receive |
| cohort experimental design | we measure exposure first and see if they develop the outcome, following people for long people at a time, the establish temporality, expensive, time consuming, not good for rare outcomes |
| case-control experimental design | generate an odds ratio, less expensive and more efficient, good for rare outcomes because we start with the outcome |
| cross sectional experimental design | prevalence, how many people have this currently, cheap and easy. No idea when it started just knowing they’re occurring together |
| Definition of internal validity | Cause and effect relationship Strengths of inferences from the study |
| Definition of external validity | generalizability |
| What is a systematic review? | Qualitative synthesis |
| What is a meta-analysis? | Synthesis in numbers |
| Levels of measurement | nominal, ordinal, interval, ratio |
| nominal | categorized, race, ethnicity |
| ordinal | ranks |
| interval | categorized rank and evenly spaced: degrees in Fahrenheit, IQ not real zero |
| ratio | all of the things of interval plus a true zero, height, age |
| confidence interval | a range of values, calculated from sample data, that likely contains the true population parameter |
| type one error | false positive, rejecting a null hypothesis when its actually true |
| type two error | false negative, fail to reject a null that’s actually false. Say there’s no difference when a difference exists |
| Statistical significance | Doesn’t always translate into clinical significance |
| p-value | .05 or less: 5% chance of it happening |
| Emergent design | Initial plan of research cant be too tightly prescribes |
| Reflexivity | Researcher needs to be conscious of bias, values, experiences they bring |
| Purposive sampling | We’re selecting individuals that can purposely inform the |
| Data saturation | Do date collection until no new information is being found |
| Triangulation | Using multiple sources to provide a corroborating |
| Memoing | RESEARCHER RECORDING IDEAS |
| Bracketing | Researcher setting aside personal interpretations and bias |
| Coding | Aggregating test into themes |
| Phenomenology | understanding the essence of experience |
| grounded theory | generate some sort of illustration in a figure of a process or interaction involving individuals |
Created by:
ago24