Unit 3: Ch. 13

Question 1

statistical results and interpretation

Answer 1

the statistical results of a study, in and of themselves, do not communicate enough meaning

statistical results must be interpreted to be of use to clinicians and other researchers

Question 2

aspects of interpretation (6)

Answer 2

1. credibility and accuracy of the results
2. precision of the estimates
3. magnitude of effects and importance
4. meaning of results
5. generalizability of results
6. implications of the results for practice, theory, and further research

Question 3

credibility and accuracy

Answer 3

credibility: is it believable?
accuracy: validity of the tool used to collect the data

Question 4

precision of the estimate of effects

Answer 4

precision is the exactness of results. Communicated through “p” values and confidence intervals (CI)

p value is the probability that the results are due to chance

• p value < 0.05 means that there’s less than 5% chance result happened by accident
• usually something less than 0.05

CI gives you a p value and a range in which most of the results will fall
-often give an OR

Question 5

magnitude of effects and importance

Answer 5

bigness of an effect; how big the effect is from the intervention (ex: did a tx cure a little bit or did it cure all cases?)

considered especially important in clinical decision making

clinically significant: important clinically

Question 6

generalizability of the results

Answer 6

whether or not you can apply what was found in the study to other groups

apply the results from the sample to the population

aka external validity

Question 7

implications of the results for practice, theory, and further research

Answer 7

theory implications, clinical implications, research implications

typically giving you ways the results can be used

Question 8

inference and interpretation

Answer 8

interpreting research results involves making a series of inferences
-inference: act of using logical reasoning to draw conclusions from limited information

we infer from study results the “truth in the real world”

Question 9

the interpretative, critique, or critical mindset

Answer 9

approach any research article with a critical and even skeptical mindset

show me! Expect researchers to provide strong evidence that their results are credible
-in other words, it’s true

don’t believe everything you read
-not all research articles are credible or true

Question 10

credibility of quantitative results (4)

Answer 10

1. proxies and interpretation
2. credibility and validity
3. credibility and bias
4. credibility and corroboration

Question 11

proxies and interpretation

Answer 11

proxy: when one thing is used to represent or substitute another thing
- ex: concepts like love, hate, anxiety require a proxy for measuring them –> come up with questionnaires to measure and describe these
- must be careful with reliability and validity

Question 12

credibility and validity

Answer 12

Validity: tool measures what it purports to measure

4 types of validity:

• statistical conclusion validity
• internal validity
• external validity
• construct validity

Question 13

statistical conclusion validity

Answer 13

accuracy of relationships between variables that have been determined via statistics

Question 14

internal validity

Answer 14

degree to which it can be inferred that the tx, and not a confounding factor, was responsible for the outcome

Question 15

external validity

Answer 15

aka generalizability

application of findings to others who are similar to the people in the sample

Question 16

construct validity

Answer 16

degree to which a tool really measures the construct under investigation

Question 17

credibility and bias

Answer 17

looking for things in articles that might slant results

Question 18

credibility and corroboration

Answer 18

authors were writing about looking for confirmation external to the article by comparing these results with results from other studies/articles

corroboration: looking for consistency across different studies and whether or not those studies found the same thing as the study you’re looking at

Question 19

corroboration

Answer 19

looking for consistency across different studies and whether or not those studies found the same thing as the study you’re looking at

Question 20

CONSORT guidelines

Answer 20

reporting guidelines developed so readers can evaluate methodologic decisions and outcomes

the consolidated standards of reporting trials (CONSORT) include a flow chart for documenting participant flow in a study

• CONSORT guidelines show you the final sample size in a study
• should be showing you where researchers got their samples from and why people were eliminated from the research

Question 21

meaning of results (5)

Answer 21

1. causation
2. going beyond the data
3. internal validity
4. generalizability
5. other situations regarding results

Question 22

meaning of results: causation

Answer 22

just because 2 variables were related and statistically significant doesn’t mean one caused the other

correlation ≠ causation

Question 23

meaning of results: internal validity

Answer 23

reader must be confident that the tx and not some other variable caused the outcome

Question 24

meaning of results: generalizability

Answer 24

reader must be confident that results can be applied to other populations and groups

Question 25

meaning of results: other situations regarding results

Answer 25

non-significant results provide no evidence of the truth or falsity of the findings

serendipitous results (“happy little accident”) - when the researcher finds things they didn’t expect to find

mixed results - when some hypotheses in the study are supported and others are not
-if some hypotheses are supported and others are not, you need to look at the methodology of the study –> may be something wrong with the way the study was conducted

Question 26

the null hypothesis in interpretation is…?

Answer 26

that the results are wrong and the evidence is flawed

Question 27

the research hypothesis is that the…?

Answer 27

evidence reflects the truth