Data Quality on Model Validity

Imagine you are working with a logistic regression model that relies heavily on a single variable. This variable's sample data includes values such as 50.00, 100.00, and 40.00. Now, consider a scenario where a data quality issue arises, causing some values to lose their decimal points. For instance, a value of 100.00 might be mistakenly recorded as 10000. Would this error affect the validity of the model? Why or why not? What strategies would you employ to rectify the model?