Confidence Intervals
What does a confidence interval estimate?
What is a population parameter?
The null hypothesis is typically a statement of this.
What is no effect or no difference?
The value we estimate using a sample.
What is a parameter?
Interpret a 90% confidence interval from 12.1 to 15.3.
What is we are 90% confident the true parameter lies between 12.1 and 15.3?
Formula for standard error of a sample mean.
A 95% confidence interval means this.
What is we expect 95% of such intervals to contain the true parameter?
The probability of observing a result as extreme or more extreme than the one observed, assuming the null is true.
What is the p-value?
The statistic that estimates the population proportion.
What is p^?
If a 95% CI for a mean is 10 ± 2, this is the margin of error.
What is 2?
The z* value for a 95% confidence level.
What is approximately 1.96?
Increasing the confidence level does this to the interval.
What is makes it wider?
If the p-value is less than α, we do this.
What is reject the null hypothesis?
The number that tells us how many standard deviations away our result is from the null.
What is the test statistic (z or t)?
If the null value is outside the confidence interval, we do this in a significance test.
What is reject the null hypothesis?
You survey 100 people and find p^=0.6. Give the standard error.
What is square root of (.6)(.4)/100=0.049
This is required before calculating a confidence interval for a mean.
What is the population is approximately normal or the sample size is large (n ≥ 30)?
The two types of errors in significance testing.
What are Type I and Type II errors?
The term for variability of a statistic due to random sampling.
What is sampling variability?
What does a confidence interval not tell us?
What is the probability the parameter is in the interval (it either is or isn’t)?
What affects margin of error besides confidence level?
What is sample size and standard deviation?
The formula for a confidence interval for a population proportion.
What is p^ +- z square root of p^(1-p^)/n
The steps in conducting a significance test (list at least three).
What are: State hypotheses, check conditions, calculate test statistic, find p-value, make a conclusion?
The margin of error does NOT account for this type of error.
What is bias?
Misinterpreting a p-value of 0.04 as "There's a 4% chance the null is true" is incorrect because...
What is the p-value assumes the null is true and measures likelihood of data, not of the hypothesis?
True or false: A larger sample size results in a smaller standard error.
What is true?