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

Module Content Objectives Introduction Reported Use of Marijuana Missing Data Analysis of Marijuana Use Subgroup Analysis Standard Deviation What Did Jack Learn From This Analysis? Lessons Learned: What Jack Should Have Asked His Evaluator How Can Jack Use This Statistical Information To Improve His Program? How Can Jack Use This Statistical Information To Promote the Results of His Program? Module 2 Quiz

Missing Data

Uh oh, did you say 48 kids at posttest? We're MISSING SCORES? That's right. Two of the kids did not answer the posttest question about marijuana use. So now what?

The conservative approach would be to exclude those kids from the analysis of the data for the marijuana question. Exclude both the pretest scores for which they both answered 0 days and the missing posttest responses. Therefore, the number of kids (N) will change from 50 to 48. We must use the new N value when calculating our statistics.

Don't try to GUESS based on the participant's pretest score. This approach doesn't work if you are looking at outcome data. However, if you are missing descriptive data, such as age or grade level, it is quite common to plug in the mean.

Jack should consider himself very lucky! He only had a few missing scores. Missing data caused from attrition and loss to followup is very common. (Our case would be considered loss to followup rather than attrition. The participants just failed to answer one particular question. They answered other questions on the posttest and were in the program until the end.)