CDC Home

# Test Hypotheses

### Purpose

The t-test and chi-square statistics are used to test statistical hypotheses about population parameters.  For example, you might wish to determine whether the mean intake of calcium among males in a population is different from that among females.  By using various statistics, it is possible to test whether these differences of statistically significant (i.e. the differences are not due solely to random error).  This module will demonstrate the use of these statistics in NHANES data analysis.

### Task 1: Use the T-Test Statistic

The t-test is used to test the null hypothesis that the means or proportions of two population subgroups are equal or, equivalently, that the difference between two means or proportions equals zero.  It is appropriate in cases where a small number (<30) of degrees of freedom are available, which is the case for the NHANES sample. This tutorial shows how to calculate a t-statistic for a sub-population only in SUDAAN.  Doing the same calculation in SAS requires more advanced programming (see the Continuous Tutorial for more information).

### Task 2: Generate Confidence Intervals

A confidence interval (CI) gives a range of plausible values of a population parameter, such as a population mean or percent.  CIs reflect the uncertainty of an estimate for a variable that is computed from a probability sample of the population rather than a census. NHANES surveys have numerous variables on demographic and health characteristics of the U.S. non-institutionalized population, such as age, body mass index, and food and nutrient intakes.  This tutorial shows how to generate a confidence interval only in SUDAAN.  Doing the same calculation in SAS requires more advanced programming.

### Task 3: Perform Chi-Square Test

The chi-square test is used to test the association between two variables cross-classified in a two-way table and the homogeneity of their association.