STATS NOTES

T-Test

A statistical test used to compare the means of two groups.

Z-Test

A statistical test used to compare proportions; based on success and totals.

Chi-Squared Test

A statistical test that compares observed frequencies to expected frequencies.

Two Sample T-Test

A T-test used to compare two independent groups.

T-Test for Slope

A T-test used to check for a linear relationship.

T-Test Matched Pairs

A T-test used when natural pairs or the same subjects are measured twice.

T-Test Format

The formula is Mean + Z score (SD / ext{SQUARE ROOT} of N).

Chi-Squared Format

Involves a table of frequencies.

Confidence Interval Including Zero

Indicates no significant difference.

Confidence Interval Not Including Zero

Indicates a significant difference exists.

Point Estimate for Mean Difference

Calculated as the sum of absolute values divided by 2.

Sampling Means Spread

The spread always gets smaller with increased sample size.

Probability of Defective Items

For 3 out of 60 defective, the probability is calculated as (0.01)³(0.99)^57.

Type 1 Error

Rejecting the null hypothesis (H0) when it is actually true.

Type 2 Error

Failing to reject the null hypothesis (H0) when the alternative hypothesis (HA) is true.

Observational Studies

Cannot conclude causation.

95 Percent Confidence Statement

A statement indicating confidence about the true mean.

Z-Test Success and Totals

The format for performing a Z-Test based on successes and totals.

Table of Frequencies

Used in Chi-Squared tests to compare expected and observed data.

Natural Pairing

Refers to subjects that are matched naturally for paired t-tests.

Linear Relationship

A relationship that can be described with a straight line, assessed by slope tests.