The Comparability of Treatment and Control Groups at Randomization

This report analyzes the treatment and control groups in the National Long-Term Care Channeling Demonstration and concludes that the randomization procedure resulted in groups that are very similar on observable characteristics. Even for site level comparisons, where larger differences were expected because of smaller sample sizes, the number of statistically significant differences was no larger than would be expected by chance. These findings imply that the Channeling control group provided a reliable measure of what would have happened to the treatment group in the absence of the Channeling intervention. Additionally they imply that other investigations which rely on screening data to assess other possible sources of non-comparability of data (e.g., non-comparability of baseline assessment data or differential sample attrition) would not be confounded by differences between groups at randomization. [30 PDF pages]