1/59
Looks like no tags are added yet.
What is a sample in a research study?
The group of participants included in the study
What does the Sample section of a study describe?
The participants in the study
What participant characteristics should you consider when evaluating a sample?
Age, gender/sex, health conditions, disease severity, etc.
What are inclusion/exclusion criteria?
Requirements that determine who is eligible or ineligible to participate in the study
What is selection bias?
When the way participants were recruited/selected may have influenced the results
What question should you ask to check for selection bias?
Could the way the participants were recruited/selected have influenced the results?
Why does sample size matter?
It affects how precise the results are; researchers should explain how many participants were needed and why
What should researchers explain about their sample size?
How the required sample size was determined (the rationale for the number of participants needed)
Does a larger sample size automatically mean a better study?
No. More participants does not automatically mean better; the number needs a rationale
What is representativeness of a sample?
Whether participants are similar to the population of interest or the patients you encounter in clinical practice
Which type of validity does representativeness relate to?
External validity
What question should you ask to evaluate representativeness?
Are the participants similar to the population of interest or the patients I encounter in clinical practice?
Study A: 10/20 patients improved (50%). Study B: 200/400 improved (50%). Which statement BEST explains how sample size affects interpretation?
Study B generally provides a more precise estimate, but representativeness depends on participant selection
Why is 'Study B automatically has greater external validity because it has more participants' incorrect?
External validity depends on how well the sample reflects the population (selection), not on sample size alone
Why is 'both studies are equally precise because outcomes are identical' incorrect?
The percentages match, but the larger sample (Study B) gives a more precise estimate
Why is 'Study A is more reliable because fewer participants introduce less variability' incorrect?
A smaller sample gives a less precise estimate; fewer participants does not make results more reliable
What is statistical power?
The probability that a study will detect a statistically significant difference if a true difference of a specified size exists
What is the purpose of a power calculation?
To determine the sample size needed to detect a clinically meaningful difference between groups
In which section of an article should researchers report how the required sample size was determined?
The Methods section
Where do you look to see how researchers calculated power?
The Methods section
What power level is commonly set in most research studies?
0.8 or 80%
What does 80% power mean in terms of repeating a study?
If an effective treatment were tested 100 times, a statistically significant effect would occur in about 80 of them
What does 80% power say about the chance of finding vs missing a significant difference?
80% probability of finding a significant difference; 20% chance of missing one
Power is related to which statistical value?
The p-value
What is the p-value in this context?
The value determined to be statistically significant
What should researchers consider before doing a power calculation?
What differences between groups would be clinically meaningful
What is data collection?
The systematic process of gathering the information needed to answer the research question
What three questions should you ask when evaluating data collection?
What data was collected? What was the setting? How was the data collected?
When evaluating data collection, what does 'what data was collected' refer to?
The specific outcome variables the researchers are measuring
When evaluating data collection, what does 'setting' refer to?
Where the data was collected
When evaluating data collection, what does 'how was the data collected' refer to?
The method used, such as a questionnaire, survey, or measurement
What is validity of a data collection tool?
The tool is appropriate for, and measures, what it is intended to measure
What is reliability of a data collection tool?
The degree to which the tool produces consistent results under appropriate conditions
What is the difference between validity and reliability?
Validity: does it measure what it is intended to measure? Reliability: does it measure the same thing consistently multiple times?
What is potential bias during data collection?
Systematic error that can influence study findings
Give an example of how bias can occur during data collection.
The people collecting the information are part of the research study and know about it
What is contamination of study groups?
Mixing bits of the new intervention/information into the comparison (control) group, blurring the lines between groups
Why do research articles mention reliability terms?
They are expected when a validated tool is used in the research design
What is inter-rater reliability?
The degree of consistency or agreement between two or more individuals measuring or rating the same characteristic
Which type of reliability looks at different people measuring the same thing?
Inter-rater reliability
Example of inter-rater reliability?
Two respiratory therapists independently reviewing the same thing and agreeing
What is test-retest reliability?
Consistency when the same tool is given to the same participants at different times
What must be true about the characteristic for test-retest reliability to work?
It must be relatively stable (not changed between tests); practice effect is a concern
What is internal consistency reliability?
How consistently different questions/items within the same tool assess the same characteristic
Example of internal consistency?
A survey that asks differently worded questions to measure the same characteristic
Researchers develop a COPD self-management confidence questionnaire and check whether the different questions produce appropriately related responses. Which reliability?
Internal consistency
Why is inter-rater NOT the answer for the COPD confidence questionnaire scenario?
Inter-rater involves two or more people rating the same thing; the scenario compares items within one tool
Why is test-retest NOT the answer for the COPD confidence questionnaire scenario?
Test-retest gives the same tool to the same participants at different times; the scenario compares questions within the tool
Why is external validity NOT a type of reliability?
External validity is how well the sample reflects the population (generalizability), not consistency of measurement
What is sensitivity?
How accurately a test identifies individuals who HAVE a disease/condition (identifying disease)
What does high sensitivity mean?
Fewer false negatives (fewer missed cases)
What is specificity?
How accurately a test identifies individuals who do NOT have a disease/condition (excluding disease)
What does high specificity mean?
Fewer false positives
What is an ideal diagnostic test?
One with both high sensitivity and high specificity, minimizing both false negatives and false positives
What are the trade-offs of a test with high sensitivity and low specificity?
Fewer missed cases (fewer false negatives) but more false positives
What are the trade-offs of a test with low sensitivity and high specificity?
Fewer false positives but more missed cases (more false negatives)
What is a false negative?
The patient has the condition but the test is negative
What is a false positive?
The patient does not have the condition but the test is positive
Why can a D-dimer result in false positives?
Other things besides the condition of interest can affect it
How should sensitivity and specificity be interpreted?
Never alone; always together with the patient's physical appearance and other diagnostic results