Week 1 Conducting Quantitative Research in Nursing Discussion Which level of measurement would you preferred to utilize for a quantitative research? Defend

Week 1 Conducting Quantitative Research in Nursing Discussion Which level of measurement would you preferred to utilize for a quantitative research? Defend your answer. Chapter 19
Processes of Quantitative
Data Analysis
Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Tell whether the following statement is true or false:
Careful quantitative researchers lay out a data
analysis plan as they go along to guide the
progress.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
False
Careful researchers lay out a data analysis plan in
advance to guide that progress.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Quantitative Data
• Coded into numerical values
• Codes developed for legitimate data and missing
values
• Codebook contains decisions about coding and
variable naming.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
What are consistency checks?
A. Error-prone processes that requires verification
B. Values that lie outside the normal range of values
C. Codes that are not legitimate
D. Checks for internally consistent information
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
D
Data entry: error-proof process that requires
verification
Outliers: values that lie outside the normal range of
values
Wild codes: codes that are not legitimate
Consistency checks: checks for internally consistent
information
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Data Entry and Data Cleaning
• Data entry: error-proof process that requires
verification
• Data cleaning
– Outliers: values that lie outside the normal
range of values
– Wild codes: codes that are not legitimate
– Consistency checks: checks for internally
consistent information
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Missing Data Problems
• Missing values
– Amount of missing data
– How missing data are patterned
• Critical for intention to treat analysis
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Missing Values Patterns
• Missing completely at random: occurs when cases
with missing values are just a random subsample of
all cases in the sample
• Missing at random: occurs if missingness is related
to other variables but not related to the value of the
variable that has the missing values
• Missing not at random: pattern in which the value of
the variable that is missing is related to its
missingness
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Tell whether the following statement is true or false:
Listwise deletion is deleting variables with missing
values.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
False
Listwise deletion: deleting cases with missing
values
Pairwise deletion: deleting variables with missing
values
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Missing Values Strategies
• Deletion
– Listwise deletion: deleting cases with missing
values
– Pairwise deletion: deleting variables with
missing values
• Imputation
– Mean substitution: regression based estimation
of missing values
– Expectation maximization imputation
– Multiple imputation
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Data Transformation
• Reversing the coding of items
• Recoding the values of a variable
• Transforming data to meet statistical assumptions
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Assess Data Quality
• Evaluating scale reliability
• Examining distributions for anomalies or extreme
outliers that are legitimate values
• Analyzing the magnitude and direction of any biases
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Peripheral Analysis
• Determines whether pooling of participants is
warranted
• Tests for cohort effects of ordering effects
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Tell whether the following statement is true or false:
Table shells are fully laid-out tables with numbers in
them.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
False
Table shells: fully laid-out tables without numbers in
them
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Substantive Analysis Plan
• Reduce temptation to go on a “fishing expedition”
• Table shells: fully laid-out tables without numbers in
them
• Supplemental statistical analyses testing competing
hypotheses or doing sensitivity analyses can
facilitate interpretation.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Chapter 20
Clinical Significance and
Interpretation of
Quantitative Results
Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Tell whether the following statement is true or false:
Results of a study need to be evaluated with thought
to the aims of the study.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
True
The results need to be evaluated and interpreted,
giving thought to the aims of the study, its
theoretical basis, the body of related research
evidence, and limitations of the adopted research
methods.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Tell whether the following statement is true or false:
Methodologic decisions affect the inferences that can
be made between study results and the real clinical
world.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
True
Inference is central to interpretation. Methodologic
decisions made by researchers affect the inferences
that can be made about the correspondence
between study results and “truth in the real world.”
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Interpretation of Quantitative Research
Results
1. The credibility of the results
2. Precision of estimates of effects
3. Magnitude of effects
4. Underlying meaning of the results
5. Generalizability of results
6. Implications for future research, theory
development, and nursing practice
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Credibility of Quantitative Results
• Proxies
• Validity
• Bias
• Corroboration
• Precision and meaning of results
• Magnitude of effects and importance
• Interpreting hypothesized results
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Interpretation of Quantitative Research
Results (cont.)
• Inference is central to interpretation.
– Methodology affects inferences about the
correspondence between results and the “real
world.”
– A cautious outlook is appropriate.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Tell whether the following statement is true or false:
Credibility assessments involve a careful assessment
of validity threats and biases that could undermine
the accuracy of the results.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
True
Credibility assessments can involve a careful
assessment of study rigor through an analysis of
validity threats and biases that could undermine the
accuracy of the results.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Credibility Assessment
• Approaches include:
– Evaluating the degree of congruence between
abstract constructs and the proxies actually
– Careful assessment of study rigor
– Corroboration (replication) of results
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Clinical Significance
• Group-level results is often inferred on the basis of
such statistics as effect size indexes, confidence
intervals, and number-needed-to-treat.
• Individual results is discussed in terms of effects.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Clinical Significance (cont.)
• Benchmark: threshold that designates a meaningful
amount of change
• Ask whether:
– A change in the attribute is real
– A patient in a dysfunctional state returns to
normal functioning
– A patient has achieved a symptom state that is
acceptable to them
– The amount of change in an attribute can be
considered minimally important
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Clinical Significance (cont.)
• Minimal important change (MIC)
– Value for the amount of change score points that
an individual patient must achieve in order to be
credited with having a clinically important
change
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Clinical Significance (cont.)
Methods of establishing the MIC:
• A consensus panel
• An anchor-based approach
• A distribution-based method
– Bases the MIC on the distributional
characteristics of the sample
• Triangulation of approaches is increasingly common.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Clinical Significance (cont.)
MICs cannot be used to interpret:
– Group means
– Differences in means
• MICs can be used to interpret:
– If each person in a sample has or has not
achieved a change greater than the MIC
• Responder analysis compares the percentage
of responders in different study groups.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Group Level
Group-level clinical significance (practical significance)
• Involves using statistical information other than p
values to draw conclusions about the usefulness or
importance of research findings
• Most widely used statistics
– Effect size (ES) indexes
– Confidence intervals (CIs)
– Number-needed-to-treat (NNT)
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Individual Level
Individual-level clinical significance
• Efforts to come to conclusions about clinical
significance at the individual level can be directly
linked to EBP goals
– Benchmark
– The Reliable Change Index (RCI)
– J-T approach
– Patient acceptable symptom state (PASS)
– minimal important change (MIC)
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Critiquing Interpretations
• Review discussion section of research reports for
statements regarding:
– Limitations
– Sampling deficiencies
– Practical constraints
– Data quality problems
• Methodology section
– How limitations were considered in interpreting
the results
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Question
Which results are considered when interpreting the
results of a quantitative research study? (Select all
that apply.)
A. Magnitude of the effects
B. Underlying meaning of the results
C. Implication for nursing practice
D. Cost of the study
E. Credibility of the results
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins
Answer
A, B, C, E
The interpretation of quantitative research results (the
outcomes of the statistical analyses) typically
involves consideration of (1) the credibility of the
results; (2) precision of estimates of effects; (3)
magnitude of effects; (4) underlying meaning of the
results; (5) generalizability of results; and (6)
implications for future research, theory
development, and nursing practice. Cost of the
study is not generally considered relevant to the
quality of the results.
Copyright © 2017 Wolters Kluwer Health | Lippincott Williams & Wilkins

Purchase answer to see full
attachment