Unfair comparison

Dissimilar measurement

Look out for intervention comparisons where what happened was measured differently in the comparison groups.

Sometimes, outcomes were measured or detected differently in two intervention comparison groups in a study. When this is the case, it is hard to know how much it affected the study results.

For example, agricultural researchers measuring the outcomes in a comparison different of types of agricultural support services might believe that one of the interventions is better. If they know which farmers were given that type of support, they may be more likely to think that those farmers had better outcomes.

One way of keeping this from happening is not to let the people checking outcomes know which people got which intervention (to “blind” them) although this is often not feasible – particularly for socio-economic interventions.

REMEMBER: Think about whether outcomes were measured the same way in the groups that were compared.