How Bad Data is as good as missing data?

Accurate and reliable #data can bring context to research studies, help people understand trends, aid executives in knowing what’s working well for achieving organizational goals.

However, data discernment is crucial. Bad data can completely negate all the positive factors of trustworthy information.

Some glaring imperfections in data can be spotted right away. For example, those working closely with healthcare data might find various misspelled names or cases in which an entry appears two or more times in a list but should only be there once or date-related inaccuracies.

With experience, I blame the poor design of EMR which allows the user to save the data without validation. I have personally experienced EMR systems allowing users to define Immunization Name and dates as free text and as a result leaves room for errors.

What do data cleaning services do when the encounter bad data?

A lot of times, the data cleaners start drawing conclusions with what they have but this is always an assumption and will always leave for error.

Is this how we want to deal with the problem?

I am not comfortable doing manipulations based on assumptions and would rather go back to the source system to fix it.