Figure 1From: SUPPORT Tools for evidence-informed health Policymaking (STP) 17: Dealing with insufficient research evidenceTwo problems with classifying results as 'statistically non-significant' or 'negative'. The blue dots in the Figure above indicate the estimated effect for each study and the horizontal lines indicate the 95% confidence intervals. A 95% confidence interval means that we can be 95% confident that the true size of the effect is between the lower and upper confidence limit (the two ends of the horizontal lines). Conversely, there is a 5% chance that the true effect is outside this range.Back to article page