In this type of error, the measured value is biased due to a specific cause, appears when an error is introduced in the design of the work, either in the selection of individuals, in the information collected or in its analysis, so that a systematic difference is produced between the groups, not attributable to the factor under study.