Appendix A: Criteria for Establishing Levels of Evidence of Effectiveness

The following descriptions are intentionally brief. For a more rigorous definition of the criteria, refer to the parent document Reducing Tobacco Use Among Youth: Community-Based Approaches.

Strong Level of Evidence

Consistent results of strong or medium effect from:

• At least three studies with experimental or quasi-experimental designs and

• The use of at least two different methodologies

  OR

• Two studies with experimental or quasi-experimental designs and

• At least three case studies

Medium Level of Evidence

Consistent positive results from:

• At least two studies with experimental or quasi-experimental designs and

• The use of at least two different methodologies

  OR

• One study with experimental or quasi-experimental design and

• At least three case studies

Suggestive but Insufficient Evidence

Research or practice evidence that:

• Is based on a plausible rationale or on previous research and

• Is being demonstrated in well-designed studies or programs currently in process

• Minimally demonstrates that the intervention being tested is linked to a positive effect.

Substantial Evidence of Ineffectiveness

Research and practice evidence demonstrating that a prevention approach is not effective. The criterion for inclusion in this category is a statistically significant negative effect in a majority of competently done studies, including at least two quantitative studies with sample sizes sufficient to test for the significance of the effect.