Statistics Flashcards
What does p < 0.05 mean?
It means that there is less than a 5% chance that the observed difference occurred by chance
Describe Type I error
Occurs when you incorrectly reject the null hypothesis (find a difference when one does not really exist - false positive)
Describe Type II error
Occurs when you incorrectly accept the null hypothesis (find no difference when one really exists - false negative)
Probability of a type II error is termed “beta”
Describe confidence intervals
Convey the magnitude of an observed difference
A 95% confidence interval provides a 95% probability that the parameter is contained within the interval
- if the interval contains zero (no difference) it is not clinically significant
What is Sensitivity?
The probability that if a disease is present, the test will be positive
Sensitivity = true positives / total with disease
What is Specificity?
The probability that if the disease is absent, the test will be negative
Specificity = true negatives / total without disease
What is positive predictive value?
The probability that if the test is positive, the disease is actually present
PPV = true positives / all positives
What is negative predictive value?
The probability that if the test is negative, the disease is absent
NPV = true negatives / all negatives
What is the difference between prevalence and incidence?
Prevalence is the total number of people with a disease
Incidence is the frequency of new cases of a disease
Describe the normal distribution curve
68% within 1 standard deviation
95% within 2 standard deviations
99% within 3 standard deviations
What type of hypothesis testing to use for each type of data?
Continuous data
- different data points for same subject = paired t-test
- different subjects = unpaired t-test
Categorical data - chi square
Ordinal data - wilcoxon
What is standard error?
= standard deviation / square root of sample size
Describes the precision of the population mean
