HA535 Data Analytics for Health Care Managers - Units 1-4 Flashcards
What consists of the amalgamation of research evidence, experience, expertise, and patient preferences in the process of clinical patient care?
Evidence-based practice (EBP)
When is evidence-based practice prompted in a clinical care setting?
When patients ask questions about their care, diagnostics, certain therapies, drugs, etc.
What is the first step in EBP?
Asking clinical questions to familiarize the attending professional with the patient.
What consists of the most optimal outcome in the patient’s care process and overall quality of life?
Clinician collected experience
Patients describing personal values and past encounters
Clinically relevant research that has been proven with comprehensive methodology
What acronym is useful in acquiring a well built and comprehensive literature search to identify all potentially relevant scientific articles which are then evaluated for the compatibility and relevance to the patient’s case?
The PICO concept
Population/Patient/Problem, Intervention(s), Comparison, and Outcome.
What is clinical significance?
What is statistical significance?
Clinical significance requires a sizable change in the patient’s condition. It prioritizes judgement rather than statistics, such that relevant studies can be determined statistically significant but clinically insignificant.
Statistical significance is powered by a large number of observations and oftentimes produces only a trivial amount of noteworthy outcomes.
What is a collection of reports on the treatment of a single patient and have minimal statistical validity, given its lack of comparison between control groups and its general understanding of the question?
Case series/case reports
What types of studies should be reviewed when addressing diagnostic questions?
Gold standard or cross-sectional studies.
What types of studies should be reviewed when addressing therapy questions?
Controlled clinical trials, and then cohort studies in that order. Answers to questions such as cost and potential harms.
What types of studies should be reviewed when addressing prognosis and etiology of the disease?
Cohort studies, followed by case controls and case series.
Organize the following from the most general to the most isolated and concentrated ideas.
Case control studies
Randomized control trial
Animal Research
Meta-analysis
Case series/case reports
Systematic review
Cohort studies
Animal Research
Case series/case reports
Case control studies
Cohort studies
Randomized control trial
Systematic review
Meta-analysis
Food for thought - Primary literature should be the focus, while secondary sources offer assessments of original studies
What is the most specified of study designs, have thoroughly examined numerous validated studies and have combined the most statistically viable results to elaborate a cause-and-effect relationship between varying treatments and the resulting effects on patient conditions?
Meta-analysis
How is EBP used in areas other than clinical care?
Creation of policy and guidelines.
All in all, EBP generates a holistic approach towards treatment and recovery that balances general best practices with individualized care.
What are the two dominant research paradigms or methodologies used in human and social sciences?
Quantitative and qualitative research.
They are both modes of inquiry that use different methods to acquire answers to social phenomena.
There has been a growth in the mixing of quantitative and qualitative approaches in the twenty-first century, as researchers look to all available research techniques to address the research questions,
rather than promote a preconceived bias toward one methodology or another (Sechrest &
Sidani, 1995). This is referred to a a mixed methods design.
What is the difference between purists and pragmatists in regard to research design and implementation?
Purists advocate a mono-method, a single approach to research while pragmatists advocate to integrating multiple methods within a single study and the strengths of both methodologies can be utilized.
What is quantitative research often called?
Traditional, positivist, experimental, or empiricist paradigm.
To a positivist researcher, reality is objective and independent of the researcher. Research is
formal, value-free, and unbiased.
What is critical for all research?
Accuracy or validity, and consistency or reliability.
What type of research involves measuring subjects and reporting the results?
Quantitative research. Experiments test the cause and effect of the sample population.
What are the two quantitative methodologies?
Experiment and survey.
The process itself is deductive in nature, with a cause-effect approach
to the research. The researcher generalizes, leading to predictions, explanations, and
understandings
What allows the researcher to generalize the findings of a study to an entire population after the researcher has designed or used a data instrument to collect data?
Random sampling.
For those instruments used in other studies, the researcher will need permission and established instruments for data collection should have established validity and reliability.
Creswell (1994) states that “during an experiment, a researcher makes
observations or obtains measures by using instruments at a pre-test and post-test stage”. Researchers create treatment conditions and develop a step-by-step procedure for
conducting the experiment.
Once a study outlines the selection of subjects, how many will participate in the experiment, what happens next?
The research states how the random sample will be selected.
What is a variable that is identified as treatment conditions or factors in an experiment?
Independent variable.
Examples (very small sample of all that exist)
- Biological events (such as food deprivation);
- Social environments;
- Hereditary factors;
- Previous training and experience; and
- Maturity.
What are the four qualitative research designs that are frequently found in human and social science research?
Ethnographies
Grounded theory
Case studies
Phenomenological studies
What are the responses or the criterion variables presumed to be “caused” or influenced by the independent variable?
Dependent variable.
What research is based on a constructivist or naturalist approach and began as a countermovement to the positivist paradigm where, to the researcher, reality is subjective and seen through the eyes of the participants of the study?
Qualitative research.
The process is inductive in nature, and patterns or theories are developed through the research process.
Match the definition to its study.
A Ethnographies
B Grounded theory
C Case studies
D Phenomenological studies
1 This is when the researcher explores a single phenomena that occurs during a defined time or activity and collects data.
2 This is when the researcher examines a human experience through detailed descriptions
3 This is when a researcher studies a cultural group in a natural setting during a specified period of time
4 This is when a researcher develops a theory through multiple stages of data collection and compares it with other theories found in the literature.
A3
B4
C1
D2
Identify the Perceived differences as Quantitative or Qualitative [Information]
1 Quantitative researchers view causal relationships among social phenomena from a
mechanistic perspective, while qualitative researchers assign human intentions a major role
in explaining causal relationships among social phenomena.
2 Quantitative researchers assume an objective social reality, whereas qualitative researchers
assume that social reality is constructed by the participants.
3 Quantitative researchers assume that social reality is relatively constant across time and
settings, whereas qualitative researchers assume that social reality is continuously
constructed in social situations.
4 Quantitative researchers take an objective, detached stance towards research participants
and their setting, whereas qualitative researchers become personally involved with research
participants.
5 Quantitative researchers study populations or samples that represent populations, while
qualitative researchers study cases.
6 Quantitative researchers study behavior and other observable phenomena, while
qualitative researchers study the meanings that individuals create and other internal
phenomena.
7 Quantitative researchers study human behavior in natural or contrived settings, while
qualitative researchers study human actions in natural settings.
8 Quantitative researchers use preconceived concepts and theories to determine what data
will be collected, while qualitative researchers discover concepts and theories after data have
been collected.
9 Quantitative researchers generate numerical data to represent the social environment,
while qualitative researchers generate verbal and pictorial data to represent the social
environment.
10 Quantitative researchers use statistical methods to analyze data, whereas qualitative
researchers use analytic induction to analyze data.
11 Quantitative researchers study human behavior in natural or contrived settings, while
qualitative researchers study human actions in natural settings.
12 Quantitative researchers use statistical inference procedures to generalize findings from a
sample to a defined population, whereas qualitative researchers generalize case findings by
searching for other similar cases.
13 Quantitative researchers prepare impersonal, objective reports of research findings, while
qualitative researchers prepare interpretative reports reflecting researchers’ constructions of
the data and an awareness that readers will form their own constructions from what is
reported (Gall, Borg, & Gall, 1996, p. 30).
Combining the Two Methods [Information]
Mixing of quantitative and qualitative methodologies is a possibility in any
given study.
Involve the use of observation to address research questions. Both methodologies
“describe their data, construct explanatory arguments from their data and speculate about
why the outcomes they observed happened as they did” (Sechrest & Sidani, 1995, p. 78)
- Use techniques that are relatively analogous at some level of specificity, such as
triangulation. - Try to discover meaning from the interpretation of data.
- Use analytical techniques in an attempt to explain complex relationships in the social
science world (Dzurec & Abraham, 1993; Sechrest & Sidani, 1995). - Utilize techniques to verify their data.
- Use data reduction as an important part of the data analysis process.
- Promote the role of theory by testing theory (quantitative methods) or initiating and
building theory (qualitative methods). - Can use the same collection of data to get results (Onwuegbuzie & Leech, 2005, p. 380).
There has been a growth in the mixing of quantitative and qualitative approaches in the
twenty-first century, as researchers look to all available research techniques to address the
research questions, rather than promote a preconceived bias toward one methodology or
another
What is an exemplary manuscript and what does it include? (Strong Characteristics)
- One-third introduction and literature review that includes complex and important topics
facing scholars and policymakers and the state of current knowledge about those topics, - One-third research procedures and findings, which includes the sampling design, research
design, measurement, statistics, and the proper reporting of research evidence obtained,
and, - One-third discussion and implications, which encompasses synthesis and integration of
findings obtained from the present investigation within the context of existing knowledge
(Smart, 2005, p. 463).
What is a literature review?
A literature review is a review of the literature that informs the research project.
The literature review presents a theoretical background and a review of other studies that are
closely related to the study that the researcher is developing; makes connections to the
dialogue about the topic that may be appearing in literature; provides a framework for the
researcher’s study; and acts as a benchmark for findings that the research will reveal
Verification of Data
Both quantitative and qualitative data can be verified, using different techniques.
Quantitative research can be verified “through a control process and random sampling
techniques to maximize internal and external validity” (Onwuegbuzie & Leech, 2005, p. 380).
Qualitative verification methodologies include:
* Triangulation;
* Prolonged engagement;
* Persistent observation;
* Leaving an audit trail;
* Member checking;
* Weighing the evidence;
* Checking for representativeness of sources of data;
* Checking for researcher effect;
* Making contrasts and comparisons;
* Checking the meaning of outliers;
* Using extreme cases;
* Ruling out spurious relations;
* Replicating a finding;
* Assessing rival explanations;
* Looking for negative evidence;
* Obtaining feedback from informants;
* Peer debriefing;
* Clarifying researcher bias; and
* Thick description (Creswell, 1998; Onwuegbuzie & Leech, 2005, p. 380).
N/A
Typical Qualitative Research Project Format (Informational)
Vivar, McQueen, Whyte, and Armayor (2007) outline the steps that a researcher may take in
developing a qualitative research project. These steps include:
* Selecting an interesting topic;
* Conducting a literature review to identify what has already been written about the topic,
including
* The description of practice-based research, systematic reviews, literature reviews, and metaanalysis;
* Presenting the significance of the study by clarifying why the research is important, through
the writing of a proposal;
* Defining the concepts in the study;
* Finding a theoretical framework;
* Choosing a method of data collection, whether these be interviews, focus groups, and/or
observations;
* Planning the data collection, including running a pilot study to prevent unexpected
problems and to
* Identify modifications necessary in the full study;
* Describing the procedures of data analysis, with a justification of procedures to be used in
the study;
* Enhancing the quality of the study, including presenting an in-depth description of the
project, how and
* What decisions are made and any issues that occur throughout the study;
* Reporting the ethical issues; including potential risks;
* Illuminating the limitations of the study;
* Disseminating the findings, by sharing the results of the research with other professionals
through
* Presenting in professional academic journals, conferences, seminars, and other meetings
where
* Professionals gather in conversation about research;
* Planning the timeframe, or the time it takes to complete the study;
* Concluding, highlighting the major elements of the study; and, finally,
* Presenting the references, or the bibliography.
Typical Quantitative Research Project Format (Informational)
Creswell (1994) outlines a typical quantitative research project format, including the following
segments:
* Introduction:
* Context (Statement of the problem)
* Purpose of the study
* Research questions or objectives or hypothesis
* Theoretical perspective
* Definition of terms
* Delimitations and limitations of the study
* Significance of the study
* Review of the Literature:
* Methods:
* Research design
* Sample, population, or subjects
* Instrumentation and materials
* Variables in the study
* Data analysis
* Appendices:
* Instruments
What is the viewpoint of absence of sound theory?
Some researchers believe modern research lacks sound theory and does not significantly
contribute to the field of knowledge. Smart (2005), for example, asserts that “reliance on
strong theories is a rare attribute of higher education research manuscripts.” He states that
“the focus seems to be more toward development of our own distinct intellectual heritage
and less toward reliance on the intellectual traditions of the more mature and advanced
academic disciplines” (p. 468).
He mentions three benefits to including sound theory in any research. Sound theories:
- Provide a roadmap by identifying the important variables to be included in studies and the
hypothesized relationships among those variables; - Incorporate extant empirical findings within a logical and consistent framework; and,
- Offer a coherent potential answer to important issues facing the academic community (p.
467).
What skills are required of a practicing evidence based practitioner?
See photo.
What test can you do to differentiate between continuous and discrete variables?
think number of occasions of hives vs the time taken for hives to subside.
Use the “mid-way” test.
If, , for every pair of values of variable, a value exactly mid-way between them is meaningful, the variable is continuous.
What can descriptive statistics be broadly put under in two categories?
Sorting/grouping (frequency distribution tables, like a stem and leaf diagram) and illustrations/visual displays
Summary statistics
What are the most common tools used for visual display of discrete and continuous variables?
What about comparing variables?
Frequency diagrams, bar charts (for discrete variables) and histograms (for continuous variables)
Composite bar charts (see image)
Pie charts help show how total quantity is divided among its constituent variables.
Scatter diagrams can be used to illustrate the relationship between two variables. (see image)
Why is the median the best measure of central tendency from among the mean, median, and the mode?
In a symmetric distribution, all three are the same, whereas in skewed data the median and mean are not the same; lie more toward the skew, with the mean lying further to the skew compared with the median. See picture for example.
What is a measure of dispersion that are it becomes larger, the further the individual units are from the mean?
How is it calculated?
Variance.
Subtracting each individual value from the mean and then adding them and finding the average would give a clue to the total dispersion, but the problem is that since the deviations are a mix of negative and positive values, the final total becomes zero.
To calculate variance, this problem is overcome by adding squares of the deviations. So variance would be the sum of squares of the variation divided by the total number in the population (for a sample we use “n-1”). If we were to divide by n alone, we would always underestimate variance, and if we divided by n-2, we would always over estimate. N-1 is best for variance calculation.
To get a more realistic value of the average dispersion, we take the square root of the variance, which is called the “standard deviation.”
Standard deviation is the spread of a group of numbers from the mean.
The variance measures the average degree to which each point differs from the mean.
While standard deviation is the square root of the variance, variance is the average of the squared difference of each data point from the mean.
Example:
For a class of students, the scores are 85, 90, 88, 92, and 82. To calculate the variance and standard deviation, find the mean: (85 + 90 + 88 + 92 + 82) / 5 = 87.4. Then, calculate the variance: [(85-87.4)^2 + (90-87.4)^2 + (88-87.4)^2 + (92-87.4)^2 + (82-87.4)^2] / 5 = 64.8 / 5 = 12.96. Finally, calculate the standard deviation, which is the square root of the variance: Standard Deviation = √12.96 ≈ 3.60.
What is a box plot?
It is a composite representation that portrays the mean, median, and mode, and the outliers.
What measures the symmetry of distribution?
For example, distribution curves can be symmetric where they look the same on both sides of a central point.
Also, what is kurtosis? This involves “tails”.
The concept of skewness and when a distro curve is not symmetrical, it is considered skewed.
Kurtosis is a representation of outliers. Distributions with high kurtosis tend to have heavy tails indicating a larger number of outliers, whereas distributions with low kurtosis have light tails, indicating lesser outliers.
There are formulas to calculate both skewness and kurtosis.
