Chapter 4 Missing values

4.1 Inspecting missing values

A big percentage of PW_2 and LCA_CASE_WAGE_RATE_TO and YR_SOURCE_PUB_2 independently of the state. This means either that most employers have their employees in the same work site.
The employer’s postal code (LCA_CASE_EMPLOYER_POSTAL_CODE) is missing depending on the state. For exmaple, half of the applications from employers in Arkansas are missing the employer’s postal code

4.1.1 Missing values patterns

• The combination WAGE_RATE_TO, PW_2 and YR_SOURCE_PUB_2 is the most frequent. Each of these columns individually have approximately 80-90% of their values missing. This means that most of the employers specify the exact amount of salary and not a range.
  
• Less than 5% of the records are missing the employers postal code