Chapter 17 Flashcards
What is required to achieve valid and significant results?
An adequate number of study participants.
How do you pick the correct number of study participants?
Based on statistical estimations about how many people are required to answer the study question with a specified level of certainty.
What happens if too many participants are recruited?
Resources are wasted, including the time of both the researcher and the participants.
What happens if too few participants are recruited?
The whole study will be worthless, because the sample will not have enough statistical power to answer the study question.
Which population samples usually yield a sample mean close to the true population value?
Large samples
What is a confidence interval?
An estimate of how close to the population value (like the mean) a sample of a particular size is expected to be.
When the sample size is small, then it is represented by a _____(narrow/wide) confidence interval. When the sample size is large, that it is represented by a _____(narrow/wide) confidence interval.
Wide; narrow (greater certainty)
What is the confidence interval calculated based on?
1) The number of individuals included in the sample
2) The mean age of those individuals
3) The standard deviation of their ages
Larger sample size is make it more likely that a study will yield:
Statistically significant results
What tool should be used early in the study design process to identify an appropriate goal for the number of participants will need to be recruited for the study?
A sample size calculator/Sample size estimator
When the level of certainty about inputs is low, it is wise to:
Err on the side of a larger sample size.
What is another way to check for sample size requirements?
To work backwards from the number of participants likely to be recruited to see whether a study population of that size will provide adequate statistical power for the study design.
Power is related to the ability of a statistical test to do what?
Detect significant differences in a population when differences really do exist.
What is a type one error?
A study population yields a significant statistical test result even though a significant difference or association does not actually exist in the source population. (Significant when NOT significant) (a=5%)
What is a type two error?
A study population find no significant result even though there is a significant difference association in the source population (Not significant when IS significant) (B=20%)