AP Statistics Chapter 10 Review

What happens when Degrees of Freedom increases?

Its shape gets taller and shorter and starts to resemble the standard normal distribution.

How do you calculate degrees of freedom for one sample?

n-1

Why is t* bigger than z*?

Because its distribution is wider and shorter in height.

What type of variable is large counts for?

Categorical data

What type of variable is large sample for?

Quantitative data

What happens when the sample size is below 30?

You must create a dot plot then check to see if there is strong skewness or a strong outlier.

How do you find the standard error of the sample mean x-bar?

SE = Sx
x sqrt(n)

When doing the state step, you MUST:

Define all variables and say what you’re doing.

In the do step, you MUST:

State what procedure you’re doing (EX: Construct a one sample t-interval for M)

What is the formula for determining sample size given a confidence level, standard deviation, and margin of error?

z*o/sqrt(n) < ME

How many students need to be surveyed to estimate the mean number of tests send during school hours with 99% confidence and a margin of error of at most 2 texts? Standard deviation is 9 for population

First do Invnorm(0.99/2+0.5, o:1, m:0) = 2.578

Then solve for square root of n

135 students are needed.

We compare populations or treatments by doing inference about…

M1-M2

If you have two samples and are determining the conditions, be sure to use _____

BOTH

Remember when you have to make a box plot, be sure to put

LABELS!!!!!

Whenever treatments are used (AKA if individuals are given two treatments), be sure to find their mean score difference and use t-interval (NOT a 2 sample one)

ok.

When doing a conclusion about a Udiff, be sure to ____

estimate their mean difference

When doing conditions for Udiff be sure to sound like ____

you’re doing conditions for a single sample.

When doing a t-test or a two-sample t interval, be sure to mention the following:

Upper and lower boundary AND DEGREES OF FREEDOM.

How do you interpret confidence level?

If we took many many samples of ______ and each time constructed a C% confidence interval in this same way, about C% of the resulting intervals would capture the _____ (EX: the true difference in mean earning for females and males)

REMEMBER: When you are given the standard deviation, you MUST plug that into the _____

standard error formula and get that standard deviation.

When you are finding Udiff, what should you check for normality if the sample size is under 30?

Check the distribution of both sample distributions

What is the test statistic formula?

(x-bar - M)/ Sx/sqrt(n)

How do you interpret standard error of the mean?

If many samples of size ____ are taken, the sample mean will typically vary from the population mean by _____