7-9-1 Tests

7-9 Tests

The following is a list of tests, and a description of what each one tests for.

 

Test Name

Description

 

 

 

Z Test

The Z Test provides a variety of different tests based on standard

 

 

deviation based tests. They make it possible to test whether or not

 

 

a sample accurately represents the population when the standard

 

 

deviation of a population (such as the entire population of a

 

 

country) is known from previous tests.

 

 

 

 

1-Sample Z Test

Tests a single sample mean against the known mean of the null

 

 

hypothesis when the population standard deviation is known.

 

 

The normal distribution is used for the 1-sample Z test.

 

2-Sample Z Test

Tests the difference between two means when the standard

 

 

deviations of the two populations are known.

 

 

The normal distribution is used for the 2-sample Z test.

 

1-Prop Z Test

Tests a single sample proportion against the known proportion of

 

 

the null hypothesis.

 

 

The normal distribution is used for the 1-Prop Z test.

 

2-Prop Z Test

Tests the difference between two sample proportions.

 

 

The normal distribution is used for the 2-prop Z test.

t Test

Used instead of the Z Test when the population standard deviation

 

 

is unknown.

 

1-Sample t Test

Tests a single sample mean against the known mean of the null

 

 

hypothesis when the population standard deviation is unknown.

 

 

The t distribution is used for the 1-sample t test.

 

2-Sample t Test

Tests the difference between two means when the standard

 

 

deviations of the two populations are unknown.

 

 

The t distribution is used for the 2-sample t test.

 

Linear Regression t Test

Tests the linear relationship between the paired variables (x, y). The

 

 

method of least squares is used to determine a and b, which are the

 

 

coefficients of the regression formula y = a + bx. The p-value is the

 

 

probability of the sample regression slope (b) provided that the null

 

 

hypothesis is true, = 0.

 

 

The t distribution is used for the linear regression t test.

 

 

 

χ2 Test

Tests the independence of two categorical variables arranged in

 

 

matrix form. The χ2 test for independence compares the observed

 

 

matrix to the expected theoretical matrix.

 

 

The χ2 distribution is used for the χ2 test.

2-Sample F Test

Tests the ratio between sample variances of two independent

 

 

random samples.

 

 

The F distribution is used for the 2-sample F test.

20060301