Plausibility alone is not enough to invalidate/nullify survey findings though some nutrition clusters and users interpret  and take actions differently.

Dear colleague, 

There is not a one-size-fits-all threshold for when survey findings should be considered invalid. Rather, it depends on the pre-determined validation process for the setting, contextual factors, and the consensus reached by those doing the validating.

For a given setting, there may already exist specified thresholds for survey validation. For example, some technical working groups (TWGs) will not validate a survey that is scored as “Problematic” by the ENA Plausibility Check. Others will be more interested in specific quality aspects such as digit preference or if the sampling was appropriately conducted. Alternatively, a survey can be scored as “Good” by the ENA Plausibility Check but if data was falsified or tampered with the survey can still be considered invalid.  

In some cases, ENA can give a low score to a Plausibility Check parameter based on something that is a true observation of the population. For example, a small WHZ SD for a population that is highly homogenous. Any ENA Plausibility Check parameter that is given points should be explored and explained for the context. This can help the validating party understand if it is a quality issue or something else.

Lastly, it depends on those who are validating the surveys. Ultimately, validation requires a conversation and discussion among the group of individuals who will reach a consensus on whether to validate a given survey or not.