Research Final Exam
purposive sampling
(aka judgemental or theoretical sampling) Researcher "handpicks" or selects certain cases thought to be best representatives to be included in the study Finds people with specific characteristics Ex. people who have had a traumatic amputation Used when population is very small and inclusion criteria is very specific Researcher makes a judgement regarding the type of subjects needed to provide the most useful information and picks them Commonly used in qualitative research Advantage- allowance for the researcher to hand pick the sample based on their knowledge Disadvantage- sampling bias
Quantitative research
(aka scientific inquiry) these methods emphasize measurement and numbers, testing of hypotheses, and statistical analysis of data Uses experiments, questionnaires, and surveys to advance nursing science Uses NUMERICAL data (age, weight, GPA, salary) Ex: how depressed would you say you have been on a scale of 1-10
Meta analysis
(sometimes referred to as quantitative synthesis) NOT a research design; it is a statistical method that generates a summary statistic by taking the results of several studies in a specific area and synthesizing findings to draw conclusions regarding the state of the science in the area of focus Instead of individual subjects being the unit of analysis, individual studies are the unit of analysis. Information is extracted about the strength of the relationship of the independent and dependent variables of each study. This information is quantified and an average score is computed across all studies.
systematic
) Individuals or elements of the population are selected from a list by taking every nth individual (1) Ex. every 10th person admitted to hospital x with a diagnosis of CHF (2) The nth is the sample interval and depends on the size and list of desired sample size. b) Each member does not have an independent chance of inclusion c) After the first individual is selected, the rest of them are automatically determined d) Considered random sampling if list is randomly ordered e) Data collected conveniently and efficiently
simple hypothesis
*A statement about the relationship between the variables that are being investigated Simple A statement explaining/ predicting a relationship between one independent and one dependent variable Easy to understand, test, and analyze
QI projects
-question on the cycle Health care quality (IOM) Safe, effective, patient-centered, timely, efficient, equitable The purpose of a QI project is often to effect immediate improvement in healthcare delivery Focuses on systems or processes or functions or a combination QI projects are site specific and are not intended to provide generalizable knowledge The intervention or protocol can change as it is being evaluated; to incorporate new ideas or insights QI is designed to sustain an improvement QI is a necessary, integral activity for a healthcare institution; research is NOT. Typically, no literature review or critical appraisal General principles: Customer focused Leadership (someone needs to recognize problem and do something about it) Involvement of people Process approach System approach to management Continual improvement Factual approach to decision making Mutually beneficial supplier relationships Publication of QI projects- for others to learn from one site's experiences... NOT generalizable
null hypothesis
A hypothesis stating that no relationship or difference exists between two variables (aka statistical hypothesis)
non-directional hypo
A hypothesis that does not stipulate in advance the direction and nature of the relationship between two variables Used when past research provides conflicting results or relationship is unknown
directional hypothesis
A hypothesis that makes a specific prediction about the direction of the relationship between two variables Clearer and indicate more knowledge about study
Iowa model
A model for evidence-based nursing practice Designed with the assumption that ideas for EBP start with a stimulus or trigger The selection of a topic can either be knowledge-focused or problem-focused Problem-focused: infection rates, fall data, how many CAUTI, benchmark → Need to fix, address Knowledge-focused: read about, hear about and then work on that Provides guidance for nurses and other clinicals in making decisions about day-to-day practices that affect patient outcomes Problem-solving steps A pragmatic multiphase change process with feedback loops Widely recognized for its applicability and ease of use by multidisciplinary healthcare teams A team is formed to develop, implement, and evaluate the practice change The team selects, reviews, critiques, and synthesizes available research evidence A practice change is piloted (project) A determination is made regarding appropriateness of adoption beyond the pilot On-going evaluation Dissemination-last step
complex hypothesis
A statement explaining/predicting relationships between two or more independent and dependent variables
Self-reports
Advantages = Inexpensive, can obtain diverse and large samples, anonymity, little researcher bias Disadvantages = Unable to clarify points, response rate, length is an issue, social desirability bias, reading level Social desirability bias: common when you ask someone a question face-to-face Surveillance bias: measuring people moving ahead in time and watch people more closely, try not to miss anything and watch closer than other group Finite set of questions, must be clearly written, closed and open-ended types
Nursing-middle
Appropriate for defining and refining substantive component of nursing science and practice Explicit and testable Organized, limited scope, contain fewer concepts/variables than grand theories More specific than grand theories Clearly defined variables with specified relationships and easier understood The purpose of middle range is to describe, explain, and predict phenomena Grand theories are explicit and testable Ex: Pender's health promotion model
creditbility
Are the findings faithful to the descriptions and interpretations of lived experience? Prolonged engagement, verbatim transcription, data saturation, member checking, search for disconfirming evidence (data triangulation) The findings are recognized by people in the situation as an accurate description of their own experience Uses multiple interviews with participants, and gives them the opportunity to review and amend the descriptions and themes that emerge from the data analysis To enhance credibility, the researcher's experience in collecting and analyzing data, often recorded as a diary or log, are included in the discussion of the findings
confirmability
Auditability Is the reader able to follow the decisions of the researcher? Requires readers to be able to follow the decision path of the researcher and arrive at the same or comparable, but not contradictory, findings, given the researcher's data, perspective, and situation Lots of details of methods Intercoder reliability checks The way a researcher would prove audibility give someone else the transcript to look it over- codes should be in about 85% agreement give codes to other researcher, both look at same transcript separately, then get together and look at it, should be in agreement
non-prob sampling
Chance plays no role in determination of sample - Limits ability to make generalizations about findings - Many nursing research studies use non- probability sampling because of difficulties in obtaining random access to populations - The sample chosen may not represent the larger population - Bias is always a threat!!
cochrane data base
Cochrane is an international not-for-profit organization. The Cochrane library includes a collection of 6 different databases of critically appraised studies "Gold standard" database is the Cochrane Database of Systematic Reviews (CDSR) Systematic reviews are based on critical appraisal
Expressed as a coefficient
Coefficient of stability- the extent to which scores are consistent over time
Steps in QI
Collect and monitor data (1st step) Consider established benchmarks Done by one person or a team Analyze data Develop plan to address problem Implement and evaluate process improvement plan
convience
Collection of data from subjects or objects readily available or easily accessible to the researcher - no random selection Ex. people you work with, patients on the unit, people that come into the clinic, people you have access to, what they do in a mall where they randomly ask to take survey Advantages Easy to carry out Cost-efficient Disadvantages: Potential for bias Limited ability for results to be generalized (threats to external validity)
Q-sorts
Comparative rank ordering Characterizes opinions, attitudes, or judgements of individuals Yields normal distribution of responses, with fewer statements in the extreme Procedure Subjects presented with set of cards (40-100) with words or phrases 40-100 depending on research question They sort cards into piles according to some dimension (e.g. approve most... approve least) Told certain # cards must be in each pile so the subject is faced with forced choices Analyses: rank ordering, means, frequencies, etc.
nursing theory-grand
Complex and broad in scope Attempt to explain broad scopes and include numerous concepts that are not well-defined Not good guides for nursing practice Not grounded in empirical data Uses systems approach that focuses on human needs of protection or stress relief central to the discipline of nursing Ex. Neuman Systems model, Nightingale Environment Theory, Orem Self-Care Deficit theory, Roy's Adaptation theory
conceptual/operational
Conceptual general definition Operational definitions are clear-cut statements of how variables are measured Conceptual: defined as how subjects view themselves Operational: self-esteem is measured with the Rosenburg Scale
CINHAL
Cumulative Index of Nursing and Allied Health Literature and is the largest and most in-depth nursing and allied health databases. Cinahl receives content from journals, books, drug monographs, and dissertations. A "basic search" in CINAHL involves entering keywords in search field. Advanced search is available.
Research critique
Decide research's applicability to practice Determine usefulness of study Outlines strengths and weaknesses of the study Evaluate quality Evaluates research process Determines applicability of findings
Fundamentals of Evidence-Based Practice
EBP: "Clinical decisions" and "actions based on scientific research findings" mean that nurses engage in EBP. Relies on evidence from research and refers to clinicians making an effort to integrate research findings into clinical thinking and decision making It uses: Best available evidence Provider/clinician expertise Patient values and needs Characteristics: Problem solving approach that considers clinical experience Combines research evidence, knowledge, and theory Allows patient values to be heard
simple random
Every subject has an equal and independent chance of being chosen, randomized!! Selection of one individual in no way affects selection of another c) Randomly and blindly select individuals to participate (1) Random table- assign random numbers/pull names out of a hat (2) Helpful hint: whenever the researcher uses the phrase "randomly selected" it is simple random selection d) Time consuming e) Difficult, or impossible, to obtain accurate or complete sampling frame of every element in accessible population (1) Sampling frame- list of all subjects, objects, events, or units in the population Random selection vs. Random Assignment Random Selection: a) Type of selection in which each subject has an equal, independent chance of being selected. b) Refers to how subjects are chosen for a study Compare to Random Assignment: (1) Random ALLOCATION of subjects to either control or experimental group. (a) To provide SOME degree of randomization when random selection is not possible. (b) Ex. look at figure 7.1 on pg. 139
construct validity
Extent to which an instrument measures an intended hypothetical concept or construct Most valuable, but most difficult way to assess validity Can be gathered by hypothesis testing, multitrait-multimethod testing, known-groups method, and factor analysis Known groups: depression scale: psychiatric patients and primary care Convergent or divergent: administer 2 scales that are positively or negatively related to same group (strong and weak relationship) Ex: depression scale and self-esteem
Clinical significance
Findings that have meaning for patient care in the absence or presence of statistical significance
Phemenology
Focuses on the description and interpretation of people's lived experience Asks: what is the essence (elements or structural units) of a phenomenon as it is experienced by people, and what does it mean? xGives themes or patterns they identify and narrative Acknowledges people's physical ties to their world: "being in the world" - abstract ideas, meaning of things Researcher gains deep understanding of nature or meaning of daily "lived" experiences Rigorous, critical, systematic research method Views human beings as people who make choices based on meanings and values It is about the meaning of the participant (not nurse or researcher - bracketing)
non-experimental descriptive
Gather information about conditions, attitudes, or characteristics of individuals or groups of individuals Purpose is to describe the meaning of existing phenomena Studying whether one variable CAUSES the other Example: Evaluating the difference between blood pressures obtained at the upper arm and the forearm among adults.
Use of theories in research
Goal of theories Describe Explain Predict Increase knowledge Enhance understanding Improve care Guide research
Heterogeneity
Great Variability Characteristics of the group - can have diverse or similar groups of people
Ratio
Highest level of measurement Categories: different, ordered, and separated by a constant unit of measurement Has a true zero point, indicating absolutely none of a property Ratio units of measure include time, weight, and length Data can be manipulated
threats to internal validity
History Global events affect outcomes To prevent this: use control groups Maturation Consider biological or maturational processes Testing Use of pre-test Instrumentation Reliability of observation Use interrater reliability, 2 group design
Threats to internal validity
History Global events affect outcomes To prevent this: use control groups Maturation Consider biological or maturational processes Testing Use of pre-test Instrumentation Reliability of observation Use interrater reliability, 2 group design
Quota sampling
Identifies the strata of the population based on specific characteristics Two step (similar to stratified BUT NOT) Identify subgroups Use convenience sampling (instead of random like stratified) Ex. Identify freshmen, sophomores, juniors, seniors then go to senior seminar class ask for volunteers or go to freshman transition class and ask for volunteers Quota is computed proportionally or disproportionately to the population under study Ex. In US nurses 40% have AD, 50% have BSN, 3% have diploma, 5% MSN 2% have PhD If N=1000: 400 AD nurses, 500 BSN nurses, 30 diploma nurses, 50 MSN nurses, and 20 PhD nurses Send out letters until you meet quota and then be done
Type II
If it is false and you accept it Not as concerning when committed, because sometimes not enough people or instruments used were acceptable To reduce error: raise sample size
Type I error
If it is true and you reject it Concerning because you rejected a study that was true Where you set alpha level (.05) Alpha level is the degree of risk willing to take that the researcher committed a type 1 error To reduce error: raise level of significance
Non-significant
If p is greater than .05, non-significant, accept the null (there is no diff)
Significant
If p is less than .05, significant, reject the null (there is a difference) Indicates that the findings from an analysis are unlikely to be the result of chance
Beneficence
Imposes a duty on researchers to minimize harm and maximize benefits Right to freedom from harm and discomfort Right to protection and exploitation
Justice
Includes participant's right to fair treatment and their right to privacy Informed consent is obtained- subjects have adequate info, comprehend the info, and have the power of free choice
Respect for human dignity
Includes the right to self-determination and the right to full disclosure Right to self-determination- deciding voluntarily whether to participate Right to full disclosure- right to refuse with risks and benefits
Homogeneity
Internal consistency reliability Cronbach's alpha (coefficient alpha) More versatile than others because it can be used with instruments composed of items that can be scored with three or more possible values <0.8 is not reliable, >0.8 is reliable Variability (A little variability in which the scores in a distribution are spread out or dispersed).
Structured observation
Involved the use of formal instruments and protocols that dictate what to observe, how long to observe it and how to record the data Focus on observation of behaviors Concealment Method of recording observations: video recording Category sys Category systems → checklists Formal systems for systematically recording the incidence or frequency of prespecified behaviors or events Systems vary in their exhaustiveness Exhaustive systems: all behaviors of a specific type recorded and each behavior are assigned to one mutually exclusive category (e.g. handwashing) Non exhaustive system: specific behaviors, but not all behaviors are recorded (e.g. nurses positive communication behaviors in ICU)
Pearson Correlational coeffecients
Known as Pearson r- is computed with continuous measures Common correlational technique researchers use to examine the relationship between two variables. Parametric procedure using interval-ratio data If you have an Interval ratio (I/R) independent and dependent variable Use Pearson's R Do variables get measured in interval/ratio; is there a relationship? Exists in scaled (height, weight, BP, cholesterol level→ mL/dL (units)) Is there a relationship between two variables? Is someone's height and weight related? Is someone's glucose level related to weight?
Quasi-experimental (without randomization)
Like experimental, they dictate manipulation of the independent variable Outcome is "predicted" by theory and research Like a true experimental with researcher controlled intervention, but does not have a control group or randomization/randomized trials Controlled intervention aka manipulating the independent variable Missing something a TRUE experimental design has No or little control over extraneous variables Not the best design to use
Components of articles' limitations and discussion sections
Limitations- confounding variables, sample deficiencies, data collection procedures, unreliable measures, and design problems Ex: if the time of day a treatment is administered can influence the level of subjects' response Found in the Discussion section Discussion- nontechnical interpretation and explanation of findings; explains what the results mean in relation to the purpose of the study
Likert scale
Looking for testing knowledge, comfort level, values/beliefs Numbers 0-7, saying if strongly agree or disagree If have scale that is strongly agree, agree, neutral- 20-100 would be range of possible scores because you total the items Possible likert scale score would be the total Also called Summative scales, require subjects to respond to a series of statements to express a viewpoint; subjects read each statement and select an appropriate ranked responses Response choices commonly address agreement, evaluation or frequency Bias: Halo effect means that someone answers same for everything (Likert scale common)
Nominal
Lowest level of measurement Numbers classify subjects or objects into 2 or more categories If there are only 2 categories, then the variable of interest is said to be dichotomous in nature. Example: gender b/c it only has two categories of male and female
non-experimental-correlational
Measuring relationships or associations Investigates relationship between or among two or more variables Without trying to manipulate or control any of them Describes, Tests, and Predicts relationships
Parametric testing
Most powerful, used with ratio and interval data, data must be normally distributed Require assumptions to be met for statistical findings to be valid. General assumptions include the dependent variable being measured interval-ratio scale that is normally distributed in the population and groups being mutually exclusive (independent of each other) If randomization is used, the assumption of the independent is met
experimental
Must be prospective, best causality. They offer the greatest amount of control
concurrent
New instrument administered around same time that data are collected on the criterion Concern involves comparing results obtained from one sample with a diff criterion sample results
Criterion-related validity
Newer instrument compared with older instrument Measure of how well an instrument measuring a particular concept compares with a criterion (provides more quantitative evidence) Correlation coefficient for each set of scores used to determine the validity coefficient High validity coefficient indicates high criterion-related validity
control group
No intervention is used on this group / gets no treatment "Usual Care" or standard or normal procedures are continued An alternative intervention may be used (music therapy, humor therapy) It might just be another therapy that the researchers are not interested in studying Control might receive the treatment the experimental group received but NOT until AFTER the data has been collected Placebo is used Control is achieved through use of control groups
Gender (1 for males, 2 for females), blood type, religion (1 for Catholic, 2 for Protestant, 3 for Jewish) Categories can't have "more" or "less". For example: you can't be "more" Catholic than someone else
Nominal
Ex: A researcher interested in patient satisfaction may classify 20 subjects as 1 (very satisfied); another 15 as 2 (somewhat satisfied); and another 25 as 3 (not satisfied). 1 = 0-$20,999 2=$21,000-$30,999 3=$31,000-$40,999
Ordinal
PICOT
Patient/population, intervention, comparison, outcome, time
Metaparadigim
Person Refers to the individuals, families, communities, and other groups who are participants in nursing. Environment Refers to the person's significant others and physical surroundings, as well as the settings in which nursing occurs, which ranges from the person's home to clinical agencies to society as a whole. Also refers to all of the local, regional, national, and worldwide cultural, social, political, and economic conditions that are associated with the person's health. Health Refers to the person's states of well being at the time that nursing occurs Nursing Refers to the definition of nursing, the action taken by nurses on behalf of the person, and the goals or outcomes of nursing actions (the nursing process).
Grounded theory
Phenomenology: Focuses on the description and interpretation of people's lived experience Asks: what is the essence (elements or structural units) of a phenomenon as it is experienced by people, and what does it mean? xGives themes or patterns they identify and narrative Acknowledges people's physical ties to their world: "being in the world" - abstract ideas, meaning of things Researcher gains deep understanding of nature or meaning of daily "lived" experiences Rigorous, critical, systematic research method Views human beings as people who make choices based on meanings and values It is about the meaning of the participant (not nurse or researcher - bracketing) Grounded theory Discovery of theory from systematic-obtained data; grounded from sociology Based on symbolic interaction Focuses on the nature of social interaction among individual Basic principles central to symbolic interaction include the fact that human beings act in relation to one another, take one another's acts into account as they themselves act, and provide meaning to specific symbols in their lives-Figure or illustration showing us basic social process All communication interactions are social, has to do with words, interactions deal with the meaning of the person whose expressing it Characteristics: Research problem/question is discovered as the process that resolves it Should not force data with preconceived problem Researcher maintains an "open mind" Data collection: Participant observation, informal interviewing, and formal interview three main sources of data Memo'ing Theoretical sampling Process of data collection for generating theory, whereby the researcher jointly collects, codes, and analyzes data and then decides what data to collect next in order to develop the grounded theory Conceptualizing data Analyzing data Data collection and data analysis occur simultaneously Researcher does interviews, codes data/transcript, then does another interview Cannot continue on to next interview unless they analyze the previous interview Uses constant comparison method to develop and refine theoretically relevant categories Focuses on understanding a central concern or theory/core variable A basic social process explains how people come to resolve the problem or concern Avoiding contamination
Interval
Possess all characteristics of nominal and ordinal, in addition to having equal interval size based on actual unit of measure Used in many surveys NO true zero point Helpful to compute averages of scores (reports scores) Some psychosocial instruments yield interval data
primary/secondary source
Primary source is written by the person(s) who developed the theory or conducted the research Secondary Source is a brief description of a study, written by a person(s) other than the original researcher, usually critiquing the work
PARiHS
Promoting Action on Research Implementation in Health Services Developed in an attempt to reflect the complexities of bringing evidence to practice Successful implementation is represented as a function of the nature and type of evidence, the qualities of the context in which the evidence is being introduced, and the way the process is facilitated Elements: Evidence (research, clinical and patient experience, local data) 4 different types Context (culture, leadership, evaluation) Environment to setting; conductive to change Facilitation (purpose, role, skill, and attributes) Process of enabling the implementation of evidence into practice Use of teams and process oriented approach Technical and practical support
Idea of confidentiality
Protecting data that are gathered or learned from patients by not disclosing information without those patients' permission Anonymity- researchers cannot link participants to their data Vulnerable subjects- participants who require special protections Ex. children, terminally ill, institutionalized, mentally or emotionally disabled IRBs approved (Institutional Review Board) - allows permission to be granted to conduct the study
Odds ratio
Question: which of 2 groups of individuals has a higher risk of suffering an adverse outcome? → Odds ratio The odds= the proportion of people with an adverse outcome relative to those without it Ex: the odds of... Odds ratio is compute to compare the odds of an adverse outcome for two groups being compared (e.g., men vs. women, experimentals vs. controls) The numerator is the odds in the intervention arm The denominator is the odds in the control or placebo arm= Odds Ratio So if the outcome is the same in both groups the ratio will be 1, which implies there is no difference between the two arms of the study. However, If the OR is >1 the control is better than the intervention If the OR is < 1 the intervention is better than the control
Ratings scale (structured observation)
Ratings are descriptive continuum, typically bipolar Ratings can occur: At specific intervals (observing environmental cleanliness q 80 in hallway of a unit) Upon the occurrence of certain events (shift change) After an observational session (global ratings) random
Ex: temperature Celsius, distance, HR, BP Fahrenheit temperature is not an example b/c 0 degrees fahrenheit does not represent absolutely no heat.
Ratio
Difference between research, QI, and EBP
Research- focuses on questions where the answer is not known. It generates new knowledge Not a routine part of clinical practice EBP- implements evidence in practice Integral to ongoing improvement of quality care (not management) QI (EBQI)- evaluates how well practice is working Integral part of ongoing management of system for delivering clinical care *PSDA determines QI from research article Plan: plan the change and observation Do: try the change on a small scale Study: analyze the data and determine what was learned Act: refine the change based on what was learned, and repeat the testing *Both EBP and QI promote use of the best evidence with the best processes so best practice outcomes are improved
manipulation/intervention
Researcher must be able to manipulate the action of the independent variable Researcher directly determines what form of an independent variable will be able to be manipulated Example of variables that CAN be manipulated: type of intervention, teaching methods, educational programs Example of variables that CANNOT be manipulated: gender, religion, ethnicity, religious preferences
randomization
Researcher must be able to select subjects randomly from the population Then randomly assign them to the control and experimental groups Based on probability alone
systematic review
Rigorous process of collecting and reviewing the literature to answer a specific clinical question Methods associated with systematic reviews are carried out in a methodological and organized manner Broad questions answered by a systematic review: What is the state of the science in a particular area? What are the methodological gaps in current studies? What is the impact of interventions on patient outcomes from these studies? What should be the focus of the next research study?
Roger's
Rogers' Diffusion of Innovations is a framework of change and work as an excellent guide for practitioners. It consists of five variables that determine the rate of adoption of innovations: Attributes of innovations Innovation is the new idea, practice such as training and instruction to the individuals Type of innovation- Decision Communication channel Is diffused through certain channels and then adopted Sharing information top to bottom, bottom to top etc. Who are the key players? Nature of the Social System Members of that society engaged to accomplish a common goal (people and the organization) Extent of change Agents' Promotion Variables determining the rate of adoption of innovations Optional Collective Authority Rate of Adoption of new Innovations Innovators-out-of-the-box thinkers Always make the first move 2.5% Early adopters-opinion leaders Hold back until innovators make the first move Champions of change 13.5% Early majority-follow the lead of the early adopters 34% Late majority-wait to see how innovations is progressing Cautious 34% Laggards-individual stepped in tradition Difficulty with change 16% Five stages in the decision innovation process Knowledge Persuasion Relative advantage, compatibility, complexity, trialability, observability Decision Implementation Confirmation
stratified sampling
Selecting a sample to identify subgroups in the population that are represented in the sample b)Achieves greater degrees of representation within population (1) Ex. all acute care hospitals in PA but divide the list into 4 regions, select one large and one small hospital from each region and then chose subjects randomly from selected hospitals c) Eliminates bias's like geographical influences in sample d) Associated with the lowest sampling error e) Populations can be stratified by age, gender, ethnicity, socioeconomic status, diagnosis, type of care, type of institution, geographic location
sampling
Selection of sample members guided by desire for information-rich sources; process of selecting individuals for a study in such a way that individuals represent the larger group from which they were selected "Representativeness" is not a key issue We don't expect to generalize, we just talk about this specific population chosen Random selection is not considered productive/NOT a characteristic of qualitative research
surveys
Self administered, may be mailed or conducted face-to-face Series of questions posed to a group of subjects; surveys describe, explain and compare or explain knowledge, attitudes and behavior
Hierarchy of evidence
Several different approaches to grading evidence or use of a rating scale have been proposed. Although there is no general consensus among researchers and/or associations about what constitutes different grades and/or levels of evidence, most hierarchies rank systematic reviews and randomized clinical trials as the strongest, and expert opinion of authorities and/or reports from expert committees ranked as weakest evidence Systematic reviews and meta-analysis - LEVEL I (STRONGEST) Randomized clinical trials - LEVEL II Well-designed RCTs w/o randomization - LEVEL III Well-designed case- control/cohort studies - LEVEL IV Systematic reviews of descriptive/qualitative - LEVEL V Expert opinions/reports from expert committees - LEVEL VII (WEAKEST)
Time interval between measurements is important
Should be no longer than 4 weeks If time is too short- good memory from first test may affect the second test scores
Ordinal
Specified the order of items being measured without specifying how far apart they are Classifies categories incrementally and rank order the specific areas of the category Ranked highest to lowest and most to least Ex: A researcher interested in patient satisfaction may classify 20 subjects as 1 (very satisfied); another 15 as 2 (somewhat satisfied); and another 25 as 3 (not satisfied). Does not indicate intervals between ranks Shows which subject reported a higher level but does not indicate how much higher Data is put into ranks, specifying which term was 1st, 2nd, 3rd Ex for income: 1 = 0-$20,999 2=$21,000-$30,999 3=$31,000-$40,999 Like nominal, ordinal is not manipulated mathematically
Evaluation of Biophysiologic measures:
Strong on accuracy, objectivity, validity and precision (may need to re-calibrate equipment) May be cost-effective for nurse researchers But, caution may be required for their use and advanced skills may be needed for interpretation
Quesitonnaires
Structured, self-administered surveys, in writing Closed-ended: ask subjects to select an answer from among several choices (male vs female)* Open-ended: Useful in early stages or at end of questionnaire, code answers if analyzed by computers Questions: specific, neutral, represent on concept, finite set of questions, must be clearly written/ simple language, delimit any reference to time
ethnography
Study and description of a particular culture of a particular group of people (oldest form of qualitative research) Takes place in 3 stages: pre-fieldwork, fieldwork, and post-fieldwork (live among the people) Culture is the way a group of people live - the patterns of activity and the symbolic structures that give such activity significance Inferred from the group's words, actions, and products Relies on extensive, labor-intensive fieldwork Assumption: culture guides the way people structure their experiences Seeks an emic perspective of the culture Insider's view Relies on wide range of data sources (interviews, observations, documents; some may be quantitative) Participant observations is a particularly important sources Can't just observe, must go into field and live/experience the study Product: an in-depth holistic portrait of the culture under study 3 stages: Pre-fieldwork: choose a group to study and learn more specific and detailed info about culture Sort out practical considerations, such as language training, funding and residence Fieldwork phase 1: Participant observation Make contacts, develop trust, establish consistent role Gate keepers: introduces you to what you're looking for; keeps the contacts flowing Become familiar with daily routines But do not want to intrude Fieldwork phase 2: Work closely with small number of informants Begin to identify emerging themes Fieldwork phase 3: Raise more sensitive questions Obtain increasingly large volume of information Double-check and monitor field information Post fieldwork: Review findings and begin writing up findings Findings written in authoritative and neutral voice Presentation of findings: rich in detail and well ordered
Frequency distribution
Systematic arrangement of numeric values on a variable from lowest to highest and a count of the number of times (and/percentage) each value was obtained Can be described in terms of: Shape, Central tendency, and Variability
ANOVA
Test the difference between more than two means (3 or more group means)* Difference between ANOVA and T-test is: the number of groups being compared T-test (2 groups), ANOVA (3+ groups) Sorts out the variability of an outcome variable into two components: variability due to the independent variable and variability due to all other sources Variation between groups is contrasted with variation within groups to yield a F ratio statistic One-way ANOVA (e.g., three groups) One independent variable with several levels Ex: "nursing speciality" can have three levels: pediatric, community-health and medical-surgical nursing Multifactor (e.g., two-way) ANOVA Represents two independent variables with several levels Repeated measures ANOVA (RM-ANOVA): within subjects Is there a difference in mean scores between group A, B, and C? Ex: music therapy, guided imagery, meditation on anxiety levels 3 difference intervention groups checking mean scores on anxiety Result of ANOVA- you get an overall F-statistic Yes or no to a difference* If yes: perform post-hoc test (aka multiple comparisons) to see where the difference is (if p is less than .05) If no: stop there
Chi square
Test the difference in proportions in categories within a contingency table (not ratio or interval data) Compares observed frequencies (actual number in each category) in each cell with expected frequencies(based on theory or past experience)- the frequencies expected if there was no relationship Nonparametric statistic; appropriate when variables are measured on a nominal scale and the researcher counts the number of items in each category Categorical independent and dependent variables use a chi square Looks at numbers, counts, frequencies (examines expected vs. observed data) Exists as words (treated, untreated) Is there a dependency between one variable and the other? Ex: usual discharge teaching and a group that received follow up telephone support 2 categorical variables
T-test
Tests the difference between to means-if the sample means are far enough apart, the t-test will yield significant difference, allowing the researcher to conclude that the two groups do not have the same mean Parametric test T-test for independent groups: between-subjects tests Ex: means for men vs. women T-test for dependent (paired) groups: within-subjects tests Ex: means for patients before and after surgery T-tests are only used if you have 2 groups* Is there a difference in mean scores between group A and group B? Group A and group B have to be categorical-between men and women, treated and untreated Then must decide: are you doing independent t-test or paired sample t-test? Independent: the scores are independent of each other (not influencing each other) Cannot be in one group if you are in the other Ex: Men vs. women, diabetes patients vs. non-diabetes patients Paired: some relationship or influence from one group to the next Ex: pre and post test on the same people (most common) Test a group, and then test them again at a later time Husband and wife (some relationship)
EBP
The conscientious, explicit, and judicious use of theory-derived, research-based information in making decisions about care delivery to individuals or groups of patients and in consideration of individual needs and preferences Derived from the principle that healthcare professionals should not center their practice on tradition or experience, but instead on scientific research findings
equivalence
The degree of similarity between alternative forms of an instrument or between multiple raters/observers using an instrument Most relevant for structured observations* Uses a checklist (should match- using interrater reliability) Assessed by comparing agreement between observations or rating of two or more observers (interobserver/interrater reliability) Example: Cohen's kappa (if greater than 0.75, indicates good agreement)
Stability
The extent to which scores are similar on two separate administrations of an instrument; concerned with consistency of repeated measurements Evaluated by test-retest reliability- same instrument, same subject, same conditions, on two occasions Things that can change are not good examples Fatigue, mood
fittingness
The extent to which study findings fit the data; that is, findings should be truly grounded in the lived experience under study and reflect the typical and atypical elements of that experience Uses direct quotes from the study (and thick descriptions) Ultimate test: the affirmation of the existence and meaning of the lived experience by the reader
mode
The most frequently occurring score in a distribution Appropriate for nominal data Useful mainly as gross descriptor, especially of nominal measures Ex: 2,3,3,3,4,5,6,7,8,9 Mode = 3
median
The point in a distribution above which and below which 50% of cases fall Useful mainly as descriptor of typical value when distribution is skewed (Ex: household income) Skewed distribution: has outlying observation in either direction, above or below the mean Symmetrical distribution: the mean, median, and mode are all the same It is not sensitive to extreme scores Ex: 2, 3, 3, 3, 4, 5, 6, 7, 8, 9, Median = 4.5
range
The simplest measure of dispersion It calculated by subtracting the lowest score from the highest score. It is considered an unstable measure b/c it is based on only two values in the distribution and is very sensitive to outliers
Nursing-borrowed/shared
Theories are taken from other disciplines and applied to nursing questions and research problems Useful if the fit and relevance to nursing are clarified Ex: A researcher expanded the theory of planned behavior to help explain the influence of parents on adolescent behaviors and describe its application to adolescent sexual risk behaviors. Ex. Transtheoretical Theory
descriptive statistics
Used to describe, organize, and synthesize/summarize data Telling us about our data Includes measures of central tendency and dispersion Parameters: descriptor for a population Numerical characteristic of a population (mean, SD) Statistics: descriptive index from a sample
inferential statistics
Used to make inferences about the population based on sample data Infer to the population Generalize findings from a sample to population Make inferences concerning a parameter Used to answer the research question or hypothesis testing Two parts: estimation of parameter, hypothesis testing Based on laws of probability
Non-parametric testing
Used with nominal and ordinal data, not based on population parameters, used with small samples Make no assumptions about the shape of the distribution Less powerful than parametric tests
visual analogue scales
Useful for subjective phenomena (pain, fatigue, anxiety)-measures symptoms Unidimensional Only quantifies intensity Measures millimeters from low end to the subject's mark Use a line, but 0-100(more precision in numbers)- all of the interval markings are shown Subjects are asked to mark a point on the line indicating the amount of the phenomenon is scored by measuring the millimeters from the low end of the scale to the subjects mark
network sampling
Useful when formal lists are limited Researchers asks subjects to help them get in touch with other potential subjects They don't have this disease but they persuade the people that do have it to participate in the study Obvious biases exist because subjects are not independent of one another and because they volunteer to participate Used more in qualitative studies
snowball sampling
Useful when list of individuals who share particular characteristic cannot be obtained Members identified to be part of group identify other members of the group Useful for studies where the criteria for inclusion for a specific trait is difficult to find. Volunteer to be in the study
Semantic differential scales
Uses pairs of opposite adjectives, measures attitude and/or belief, assigns value to each space on scale, avoids bias by random ordering Set of scales that use pairs of adjectives that reflect opposite feelings; it measures attitude, beliefs or both Semantic=use of words; differential= opposite Different from Likert: only two extremes are labeled and the continuum is based on agree/disagree but rather on opposite adjectives that express the respondent's feeling Horizontal line and anchors with opposite words at the end (easy or difficult) Response to emotion/perceptions about something
content validity
Usually used in questionnaire development, or interview schedule/guide Extent to which an instrument or test measures an intended content area Constructed using concepts from the literature to reflect the range of dimensions of the variable being measured Determined by panel of experts Evaluate all items or questions used Appropriateness for use in a proposed study population
Biophysical measures
Variables measured with physiological instruments The method of measurement must closely fit the conceptual definition of the variable or concept Scoring and interpretation guidelines Instruments are typically developed, tested, and refined to measure a particular variable or concept (ex: scale) Accomplished with various monitoring equipment to measure vital signs or other physiological data Yields quantitative data EX: maximal inspiratory pressure is used to indicate inspiratory muscle strength, which provides an index of performance for patients with COPD
Determine generalizability
Were they able to be generalized to the general population? Extent to which research findings can be generalized beyond the given research situation to other settings and subjects; also called external validity
cluster
a) Mainly know it is probability sampling b) Groups, not individuals, are randomly selected c) Used for convenience when populations are large d) Done when there are multiple master lists at different places e) Two step process f) Start off with a large population and choose another subgroup (1) Clusters (2) Selection of individuals from those clusters by random selection g) Ex. 5 out of the 20 sites are chosen and then randomly selected from the sampling frame (aka "master list")
Threats to external validity
all threats limit generalizability Interaction of selection and treatment Samples confined to certain types of subjects Research group and control group are different on income levels and researchers are measuring attitude on incorporating more fruits into daily diet (higher income influences access to fruits) Hard to generalize because it is such a specific characteristic that defines the sample The number of subjects who were invited to participate and refused should be reported in order to ascertain threats to external validity Interaction of setting and treatment Bias when members of different settings agree to participate Ex. hospital and rehab centers Ex. authoritarian PT at one site and more gentle PT at another site Interaction of history and treatment Concerns about ability to generalize results to different periods of time in the past or future Did things in the past have an effect on the outcome Ex. cholesterol reduction versus now and 20 years ago Need to consider the period in history during which the study was conducted and the time it is now.
DV
also called criterion or outcome variable is the variable that is observed for change or reaction after the treatment is applied
Qual research
an approach to structuring knowledge that uses methods of inquiry that emphasize verbal descriptions and the meaning of the experience for the individual. Mainly used when little to no info is available on the topic These methods emphasize understanding of phenomena from the individual's perspective Uses NARRATIVE data (verbal, quotes from participants) Ex: Have you felt sad or depressed at all lately? Uses participant observation, in-depth interviews, case studies, ethnographies, and narrative analyses Emphasizes understanding of phenomena from individual perspective Used in research subjective experiences Has interviews and observations No hypothesis
Principles of ethical behavior
benefiecence, respect for human dignity, justice
Research dissemenation
communicating research findings Publications- reaches larger audience Disadvantage: the delay in receiving feedback Page 304-305? Posters - engages in active dialogue w/ other researchers Reaches limited number of people with limited amount of info Oral presentations Advantages- quick dissemination and audience feedback Disadvantage- reaches limited number of people
patients who watch 3 videos or 6 videos will have lower blood sugar, increased weight loss, and exercise
complex hypothesis
Ex: instrument measuring stress that has been validated in a healthy normal population will not provide valid results when used to measure stress in patients Ex. subjects complete 2 depression surveys & scores correlate with each other
concurrent validity
Reliability:
consistency of instrument or test; examined in every study where a instrument is used in and retested every time Expressed as a reliability coefficient ® Range from 0.00-1.00; a completely reliable test has a coefficient of 1.00, a completely unreliable test has a coefficient of 0.00
subjects randomly assigned to standard care plus informational video intervention information will have higher satisfaction ratings than the other group.
directional hypothesis
external validity
extent to which study results can be generalized from the study sample to other subjects, populations (wide population), measuring instruments, and setting Subject selection: those who are in the study, decline, or drop out (attrition)
patients with diabetes who watched 3 education videos will have lower blood sugar results than those patients who received usual care
hypothesis simple
meta-synthesis
intentional and coherent approach to analyzing data across qualitative studies. It is a process that enables researchers to identify a specific research question and then search for, select, appraise, summarize, and combine qualitative evidence to address the research question Used with qualitative studies
Ex: age, education years, temperature Temperature is a good example because a temperature of zero is not absolute. When it is 0 degrees, there is no absence of temperature.
interval
What is the relationship between X and Y" - don't know if it is positive or negative
non-directional hypo
Ex. college admission exams (NLN, PAX, TEAS) GRE scores to admit students to grad school High HDL, low LDL, increase lipid and increase HTN→ predictive of MI
predictive validity -Ability to predict future events, behaviors, or outcomes
rare disease people will tell others with that same disease to participate in the study Someone volunteers to be in the study, usually from someone in the study telling someone else to do it (snowballing) or one woman volunteered and gave names of other women to contact.
snowball sampling
Ex: stopping smoking cessation program and ask have you had a cigarette since we last met (and you did)-but lie and say you didn't, not true but answer you give them
social desirability bias
3 aspects of reliability:
stability, homogenity, equivalence, heterogenity
mean
the average" Equals the sum of all scores divided by the total number of scores Ex: 2, 3, 3, 3, 4, 5, 6, 7, 8, 9 Mean = 5 Sensitive to outliers that pull direction of the mean Most stables and widely used indicator of
Accuracy of an instrument or test Ex: an instrument may have excellent reliability even though it may not measure what it claims to measure (accuracy)
validity
IV
variable that is observed, introduced, or manipulated to determine the effect it has on another variable. It can also be classified as experimental or predictor variable
From no pain (beginning of line) to worst pain you've ever felt (end of line), mark an X on the line that depicts your pain
visual analgoue scale
