Primer 16 - Biostats, Confidence Interval Flashcards
What percent is involved in two standard deviation in a normal bell distribution?
It is 95 percent.
What percent is involved in three standard deviations in a normal bell distribution?
It is 99.7 percent.
How is Standard error of the mean (SEM) calculated?
Standard deviation over the square root of a sample size.
How do we decrease Standard Error of the mean and why?
By increasing the sample size: the equation for SEM is standard deviation over the square root of the sample size.
What is confidence interval (CI)?
It is a type of interval estimate of a population parameter. It is observed interval (i.e., it is calculated from the observation), in principle different from sample to sample, that frequently includes the value of an unobserved parameter of interest if the experiment is repeated.
How is Confidence Interval (CI) calculated?
CI: Range from [mean - Z(SEM)] to [mean + Z(SEM)].
Z= a constant that is specific to the confidence interval that you are trying to calculate (you don’t have to calculate it).
If you are looking for a 95% CI, Z is 1.96.
If you are looking for a 99% CI, Z is 2.57.
What happens when the Confidence Interval crosses zero?
That there is no difference between the groups. That the study intervention did not affect the outcome: accept the null hypothesis.
In a study of USMLE scores at a particular medial school, the mean score is 230 and the standard deviation is 20. Knowing that the sample size is 100, calculate the 95% confidence interval.
SEM: standard deviation / (Square root of sample size).
CI: mean -[Z(SEM)] to mean +[Z(SEM)]
For 95%, Z= 1.96.
SEM: 20/square root of 100= 2
CI: 230-[1.96(2)] to 230+[1.96(2)] = 226.08 to 233.92.
In a study of USMLE scores at a particular medial school, the mean score is 230 and the standard deviation is 20. Knowing that the sample size is 14, calculate the 95% confidence interval.
SEM: standard deviation / (Square root of sample size).
CI: mean -[Z(SEM)] to mean +[Z(SEM)]
For 95%, Z= 1.96.
SEM: 20 / square root of 14 = 5.35.
CI: 230 -[1.96(5.35)] to 230+[1.96(5.35)] = 219.5 to 240.5
What is the Student’s t-test used for?
A way to calculate the mean of only two groups when the study data are continuous variables for each of the groups.
What is an ANOVA test?
A way to calculate the mean of three groups or more when the study data are continuous variables for each of the groups.
What is a Chi Squared (X^2) used for?
Can be used to determine the percentages or proportions of categorical outcomes, (not mean values which is for t-test or ANOVA) into two or more groups.
What is the Correlation coefficient?
Designated by a small r, looks at two variables in your study and determines the relationship between the two.
If r = 1: Perfect correlation b/w the 2 variables.
If r = 0: No relationship b/w the 2 variables.
If r > 0: Positive correlation.
If r
What is an example of a positive correlation and a negative correlation coefficient?
Positive: As you smoke more, the chances of getting cancer increase.
Negative: As you do more exercise, the incidence of coronary artery disease decreases.
What is coefficient of determination?
Designated as a small r ^2, it is a determination of goodness of fit; it reflects the portion of variance and why that is due to the variance in X. Again, it a measure of how two variables are related to each other.
