Chapter 13 - Rigor and Interpretation in Quantitative Research

Question 1

What are the aspects of interpretation?

Answer 1

1. The CREDIBILITY and accuracy of the results
2. The PRECISION of the parameter estimates
3. The MAGNITUDE of the effects and importance of the results
4. The MEANING of the results, especially with regard to causality
5. The GENERALIZABILITY of the results
6. The IMPLICATIONS of the results for nursing practice, theory development, or further research

Question 2

What is an inference?

Answer 2

the act of drawing conclusions base on limited info, using logical reasoning

• several inferences are used to interpret research findings
• -> We infer from study results “truth in the real world.”

Question 3

Consolidated Standards of Reporting Trials (CONSORT) Guidelines

Answer 3

Reporting guidelines have been developed so that readers can better evaluate methodologic decisions and outcomes.

–>a flow chart for documenting participant flow in a study

Question 4

The Interpretative Mindset

Answer 4

Approach the task of interpretation with a ________—and even skeptical—mindset.

-Test the “null hypothesis” in interpretation that the results are wrong against the “research hypothesis” that they are flawed

____ __!!! Expect researchers to provide strong evidence that their results are credible—i.e., that the “null hypothesis” has no merit.

Question 5

CREDIBILITY: Proxies and Interpretation

Answer 5

Constructs are linked to the actual strategies/outcomes in a series of approximations that can be evaluated. (how did we get our sample population? –> delineation, identification, selection, attrition)

The better the proxies, the more credible the results are likely to be.

–>how plausible is it that the actual sample reflects the recruited sample, accessible population, or the target population?

Question 6

CREDIBILITY: Credibility and Validity

Answer 6

• ->in interpretation, researchers must search the evidence that the desired inferences are, in fact, valid
• may need to consider alternative hypotheses

1. Construct Validity
2. Statistical Conclusion Validity
3. External Validity
4. Internal Validity

Question 7

Construct Validity

Answer 7

involves inferences from the particulars of the study to higher-order constructs
–>do the specified eligibility criteria adequately capture the population construct? (who our research is actually trying to assess)

Question 8

Statistical Conclusion Validity

Answer 8

the extent to which correct inferences can be made about the existence of “real” relationships between key variables
-do power analysis at the outset to estimate how large of a sample is needed

Question 9

External Validity

Answer 9

the generalizability of the results

–>who would dit be safe to generalize the results to?

Question 10

Internal Validity

Answer 10

the extent to which a causal inference can be made
–>were those in the intervention group more likely/less likely to drop out of the study than those in the control group? (attrition)

Question 11

CREDIBLITY: Credibility and Bias

Answer 11

Not so much a question on if there are bias present - but rather what types of bias are there and how extensive/sizable/systematic the bias are?

Question 12

CREDIBILITY: Credibility and Corroboration

Answer 12

Corroboration: evidence that confirms or supports a statement/theory
-can come from internal or external sources –> REPLICATION is important in both cases

Question 13

What is used to determine precision of results?

Answer 13

p values are incomplete

CIs communication info about whether study findings are precise

Question 14

Meaning of Results: Interpreting Hypothesized Results

Answer 14

• correlation does NOT prove causation
• avoid the temptation to go beyond the data to explain the results
• there is always the possibility that the observed relationships occurred by chance (even when you receive expected results, ALWAYS test them)

Question 15

Meaning of Results: Interpreting Nonsignificant Results

Answer 15

• the real reason for failing to reject a null hypotheses is often hard to discern
• acceptance of a false null hypothesis could be caused by poor internal validity, an anomalous sample, weak statistical procedures, or unreliable measures

***Nonsignificant results provide NO evidence of the truth or the falsity of the hypothesis

-interpreting nonsignificant results can be aided by considering factors like sample size and effect size estimates

Question 16

Meaning of Results: Interpreting Unhypothesized Significant Results

Answer 16

Can occur in two situations:

1. exploring relationships not considered during study design
2. obtaining results opposite to those hypothesized

Question 17

Meaning of Results: Interpreting Mixed Results

Answer 17

• ->some hypotheses supported, some not
• may be caused by differences in validity or reliability of the measures
• may suggest a theory needs to be qualified

Question 18

Meaning of Results: Generalizability of the Results

Answer 18

researchers seek evidence that can be used by others

–> will the intervention “work” in other settings?

Question 19

Meaning of Results: Implications of the Results

Answer 19

once you have reached conclusions about the credibility, precision, importance, meaning, and generalizability of the results - you are ready to draw inferences about the implications

–>what is the next step? how should the results be used by nurses in their practice?