Unit 6: Proportion Inference

Margin of error

the amount on either side of the statistic in a confidence interval. (the plus or minus number)

Critical value (defn. & symbol)

z*, the number of standard deviations we need to be ___% confident

statistic for a proportion

p-hat, my sample proportion, the % from my sample

parameter for a proportion

p, the true % from the entire population

point estimate

my p-hat, that is an estimate of p

Central Limit Theorem for proportions

As sample size increases, the sample proportions become approximately normal. Specifically when np and n(1-p) > 10.

Describe the sampling distribution of the sample proportions

It is centered at p, has a standard deviation of sqrt(p*1-p/n), and (if np and n(1-p) are > 10) is approx normal.

Null hypothesis

Ho. The statement about the population we accept to be true. Starts with p = ___.

Alternative hypothesis

Ha. The alternative hypothesis. The statement we are trying to see if we have evidence to believe. It says p < or > or ≠ ____.

test statistic

the z score that measures how far my statistic is from my parameter.

Conditions for inference (for proportions)

Random sample, n < 10% of the pop, and np > 10, n(1-p) > 10.

P-value

The probability of getting my observed statistic, or more extreme, if Ho is true.

significance level (alpha)

What we compare our p-value to, 0.05 is commonly used. If p-value is less than this value, it is considered unusual.