cognitive review Flashcards
Counted the number of times each event occurs.
FREQUENCY DISTRIBUTION
Spread or variation of the data/score
MEASURES OF VARIABILITY
Identify the meaning of a phenomenon, experience for an individual participant
Data gathered directly from informants through interview or observation
QUALITATIVE STUDY DESIGN
Stability within an instrument
How well the items measure same Variable and yield similar results?
Cronbach’s alpha > 0.7
INTERNAL CONSISTENCY
Scale with equal intervals between the numbers
No true 0 point
Interval
Properly calibrated equipment & instruments
MINIMIZING MEASUREMENT ERROR
Coding categories: “Cookbook”
Break down data to sort/label in categories
CODING & DATA REDUCTION
Not given all the individuals in the population equal chances of being selected.
Risk for selection bias
Limitation of generalizations about findings
NON-PROBABILITY SAMPLING
Going forward (usually a long period of time)
Prospective/Longitudinal Design
Measured scores are incorrect in the same direction
Affects average score: Bias
Sources of error
SYSTEMATIC ERROR
Access one person suitable for study - Find other people
“Referral sampling”
Very good methods for sensitive subjects/issues
Example:
Found one mother of children with cystic fibrosis who lives in her neighborhood & asked her if she knows any other mothers of children with cystic fibrosis.
SNOWBALL SAMPLING
Not predict the direction of difference and/or relationship between Variables
Example:
There is a difference in the mammography rates between women who receive and do not receive the education.
Non-Directional Hypothesis: 2-tailed
Undesirable variables that influence the study outcome
Not the variables that are actually of interest.
EXTRANEOUS (CONFOUNDING) VARIABLES
Scores vary in a random way
Inconsistent data collection procedures by collector(s)
Not affect the average score
Affects variability around average score
RANDOM ERROR
Who: research assistant, trained nurse…etc.
Where: home, clinic, conference room…etc.
When: before and/or after intervention…etc.
How: interview, survey, observation…etc.
How often: once, every 3 weeks…etc.
COMPONENTS
Scale with equal intervals between the numbers
Absolute 0 point
Ratio
Two variables are related across a group of participants
CORRELATION STATISTICS
Categorize a variable
Nominal (Categorical)
How to measure the Variable/Concept?
Make decision for type of data to be collected
OPERATIONAL DEFINITION
Stability between Raters
Correlating Rater score
Percentage or Cohen’s kappa > 0.85
INTER-RATER RELIABILITY
Already collected data set by someone else for some other purposes
Patient record, Public/Commercial databases of health data, Registries, Government sources of health data .
SECONDARY DATA
General meaning of Variable/Concept
Perceived Susceptibility: One’s opinion of chances of getting a condition
CONCEPTUAL DEFINITION
Commonly used to investigate and identify the cause of certain condition/disease with small sample size
Example:
Rare type of cancer
Hepatitis outbreaks
CASE CONTROL DESIGN
PICO Approach
Population
Intervention
Comparison/Control
Outcome
Measurement of a group over time
Independent variable manipulated at some point in time
Enables the researcher to:
- Determine effectiveness of intervention
- Quantify timing of effects
Example:
Test effectiveness of planned behavior intervention to increase adherence of community mental health professionals to a national suicide prevention guideline
TIME SERIES DESIGN
Made up of people who are easy to reach.
Example:
Patients from the hypertension clinics of 2 local hospitals to explore medication adherence among HTN patients.
CONVENIENCE SAMPLING
Proportion/Rate in one sample against another
Relationship between two nominal/categorical variables
Chi-Square
Provide a high level of control
measure something
Enable the researchers to:
- Examine the effectiveness of interventions
- Measure changes over time
- Explore the relationship between variables and outcomes
QUANTITATIVE STUDY DESIGNS
Stability of measurement over time
Same measurement tool with the same participants at different time
Correlation coefficient >/- 0.5
TEST-RETEST RELIABILITY
Lack of time
Negative attitude toward research & EBP
Organizational constraints
Lack of skill to search literature and interpret evidence
Perception that lack of authority to change practice
Cost
Individual judgment/patient input devalued
BARRIERS TO EBP
Best used for many variation in variable being measured
Strata/Layers
Divide population into smaller groups
Take random sample from each stratum/group
STRATIFIED RANDOM SAMPLING
Face validity
Panel of experts review the items of instrumen
CONTENT VALIDITY
Assess if results from the sample can occur in a large population
INFERENTIAL STATISTICS
Categorize variable with rankings
Ordinal (Ranking)
Words, not numbers
Feelings, experiences, perceptions
Transcribed verbatim & field notes
Need to be sorted & categorized contents
CONTENT ANALYSIS
Recurrent meaning that appear during data analysis
THEME IDENTIFICATION: FINDINGS
Longer Left tail
Concentrated majority of distribution on the Right side
Negative Skew
Characteristics that disqualify prospective subjects from inclusion in the study.
EXCLUSION CRITERIA
Predicts there is NO difference/relationship between variables/groups
Researcher wishes to disapprove this hypothesis
NULL HYPOTHESIS (Ho)
type of QUALITATIVE STUDY that
Study of people and cultures
Explore cultural phenomena
Example:
Experiences of Hmong refugee’s ER visit in U.S.
Perception of a “good death” among Japanese Americans
Sense of obligation of filial piety among elderly KA women
ETHNOGRAPHY
No standard/guideline for the analytic process
Results will be reported in words.
QUALITATIVE DATA ANALYSIS
Interviewer chooses the questions/techniques depending on the participant’s answer & the situation
Participant speaks freely without any set question order
Unstructured Interview
Data collected directly from the participants for own study
Most reliable data
Methodology section
PRIMARY DATA
Accuracy or correctness of the study results
Independent variable(s) made the differences/changes in Dependent variable(s).
INTERNAL VALIDITY
Example:
Every 3rd nurse is randomly selected from the mailing list of the ANA
SYSTEMATIC SAMPLING
Present exactly the same questions in the same order in each interview.
Structured Interview
Identify a problem
Develop research question
Literature review/Investigate the current practice
Synthesize evidence/findings
Develop recommendations
Pilot recommended interventions and changes
Assess the effectiveness of the interventions & changes
STEPS OF EBP
Not concerned with measurement strategies
Analyze each participant’s own words
QUALITATIVE DATA
Congruent the characteristics of target population &sample
Affects generalizability of results back to the target population from which they were chosen
REPRESENTATIVENESS
Self determination Privacy and dignity Anonymity and confidentiality Fair treatment Protection from discomfort and harm
RIGHTS OF STUDY PARTICIPANTS
Predicts only one direction of change/relationship between Variables
Example:
There will be increased mammography rates among women who receive the education compared to those who do not receive the education.
Directional Hypothesis: 1-tail
- Do not require IRB review
- Must meet the criteria for Exempt Review
- Studies must pose NO risk (i.e. using public data)
Exempt Review
Rejection of null hypothesis
ALTERNATIVE/RESEARCH HYPOTHESIS (H1)
Combines Quantitative and Qualitative measurements
Involves the measurable state of a phenomenon and the individual’s subjective response to it
Use quantitative data analysis to compliment as the qualitative data analysis
MIXED METHODS STUDY
No true relationship between the variables
No true difference between two groups
Statistically testing
NULL HYPOTHESIS (Ho) TESTING
Describe a phenomenon with objective precision
Identified Variables
Statistical analysis for results
Measured Variables in a reliable and valid way
QUANTITATIVE STUDY
Non-invasive procedure
Minimal risk
Expedited Review
Assess/Estimate the prevalence of a particular disease in a certain population
Commonly used for public health planning
Examine one variable in different groups that are similar in all other characteristics.
CROSS-SECTIONAL DESIGN
To gain more information about phenomena occurring in its natural setting
No manipulation of Variables
Used to observe, document, and describe.
Does not enable the determination of cause and effect
DESCRIPTIVE STUDY
Natural approach to research. Unstructured data Interview, Observation, Focus group Summary of subjective conclusion Understanding the meaning of an event/phenomenon from the individual’s perspective
QUALITATIVE STUDY
Positive statement to predict the relationship/difference of the variables: Outcomes of study
RESEARCH/ALTERNATIVE HYPOTHESIS
type of QUALITATIVE STUDY that
Capturing the “lived experience” of a phenomenon
Example:
What is the lived experience for the treatment of a malignant brain tumor in adult males?
What is the lived experience of caregiver for a patient with Alzheimer’s disease (AD)?
PHENOMENOLOGY
Determine the strength and direction of any relationship between two variables
CORRELATION DESIGN
No concealment
Aware of being observed
Direct Observation
Given all individuals equal chances of being selected for the sample.
Must utilizes random selection process/procedure
PROBABILITY (RANDOM) SAMPLING
Very similar to Experimental design
NO random assignment to intervention/control groups
QUASI-EXPERIMENTAL DESIGN
SAMPLE SIZE
Power/Power Analysis
How large of a sample is needed to detect adetect a difference in the outcome variable?
QUANTITATIVE STUDY
Bell-shaped curve: Symmetry
Basic assumption for most statistical tests
Normal Distribution
Longer Right tail
Concentrated majority of distribution on the Left side
Skewed Distribution
Positive Skew
Possibility of generalizability of the study results to:
Other population
Other environmental conditions
EXTERNAL VALIDITY
Most commonly used for Qualitative study
INTERVIEWS
Do not meet criteria for Exempt/Expedited review
Full Review
Strongest design: provide strongest evidence for practice
Allows for conclusion of cause and effect
Manipulated Independent variable (Intervention)
Measurement of Dependent variable (Outcome)
Random Assignment of participants to groups
EXPERIMENTAL DESIGN
Judgmental sampling:
Researcher knows enough about the population and its elements to handpick the sample
Selected participants deliberately based on intentions of the researcher and purpose of the study
Chosen for specific purposes
PURPOSIVE SAMPLING
Researcher uses some knowledge of the population to build some representatives into the sampling plan.
Selected participants based on pre-specified quotas regarding demographics, attitudes, behaviors, some other criteria
Arbitrarily choose participants from each group.
QUOTA SAMPLING
Outcome has already occurred
“look back” for events that occurred in the past.
TIME ELEMENT DESIGN
Retrospective Design
Quantify participant’s level of responses
Asked to express degree of agreement/disagreement
Each degree has numerical value
Calculate a total numerical value from all participants
Likert Scale
Example:
Total of 16 nurse managers are randomly selected from 8 hospitals in 2 randomly selected clusters in the State of Illinois.
CLUSTER SAMPLING
Congruent characteristics of the population and sample
Required characteristics that the prospective subjects must have to be included in the study
ENROLLMENT: SELECTION CRITERIA
INCLUSION CRITERIA
Most commonly used data collection tools
Complete/Distribute by face-to-face, mail-out, online
open or close ended questions
SURVEY/QUESTIONNAIRE
Describe the pattern of scores among participants
MEASURES OF CENTRAL TENDENCY
SAMPLE SIZE
Redundancy and Saturation
When is the point at which no new information is being generated
QUALITATIVE STUDY
Completed with each participant
Detailed Information: extent and sufficient
Comprehension
INFORMED CONSENT
Unaware of being observed
Debriefing
Indirect Observation: Concealment
