CONTENTS

Table of Contents
Introduction

1. Descriptive Statistics in Evaluation
2. Subgroup Analysis
3. Variable - Are They Related?
4. Correlation
5. The t-Test of Difference Between Means

Supplements
Resources
Glossary

Wading Through the Data Swamp:
Program Evaluation 201

Distributions

Normal or Symmetric

A normal distribution is shaped like a bell curve. This type of distribution is symmetrical. The mean is right in the center of the bell, splitting both tails evenly.

Skewed Distributions

Those nasty outliers may turn your beautiful bell curve into a skewed asymmetrical curve. The scores may be concentrated at either end of the tail, resulting in a lopsided distribution.

Positively Skewed

These distributions are the most common in drug prevention. This type of skewed distribution has most values concentrated on the far left side, the lowest numbers. Fewer cases are concentrated on the higher end. For example, in Jack's program, marijuana use was positively skewed. Most kids didn't smoke pot, but the few that did caused the mean (1.7) to be higher than the mode and the median (both 0).

Negatively Skewed

These distributions have the most cases concentrated on the right side, the highest numbers. Fewer cases are concentrated on the higher end. The mean is lower than the median, which is lower than the mode.

Bimodal

A bimodal distribution resembles a two-humped camel. You sometimes see these types of distributions in self-reports of use. If I were you I would spare myself the headache and ask my evaluator about this one!

Uniform

This happens very rarely but it is when frequencies for all possible responses are the same.