Table 1. Comparison of wage survey methods by industry.

Survey name Construction industry wage survey Survey of wages for software engineers Survey of wages by occupation in small and medium-sized manufacturing Industries
Survey methods
Survey cycle Twice a year Once a year Twice a year
Survey target Construction workers Software engineers assigned to software projects (Basis: Article 38 of the Enforcement Decree of the Software Industry Promotion Act) Workers in small and medium-sized manufacturing industries
Survey content Actual wages by occupation in the construction industry Applied labor rate standards, Wage trend Wages by occupation type in small and medium-sized manufacturing Industries
Survey form Company status, Type of company, Wage status, Wage trend, etc. Company status, employee status, Software engineer salary, Wage trend. etc. Basic information about the company, Wages for each job of blue-collar workers
Survey method Mail survey, Online survey, On-site investigation by surveyors Mail survey, On-site investigation by surveyors Fax survey, Mail survey, On-site investigation by surveyors
Sampling method Quota sampling for considering region and contract amount Stratified sampling Stratified Systematic sampling
Population Construction workers currently working on construction sites at the time of the survey (target population) Out of a total of 15,293 companies in the entire population, companies that can represent the characteristics of the population were selected as the sample. Small and Medium-sized Manufacturing Enterprises are classified under the Korean Standard Industrial Classification (KSIC)
Alternative methods for non-response - - Weighting adjustment, Post-stratification
Computing method Wages for each job type - Standard error computing method, Variance estimation and coefficient of variation computing method
Outlier authentication algorithm IQR Review the data that has increased or decreased by more than 10% compared to the previous year Data that responded more than double the average or less than 0.7 times the average.
Handling outliers Weight decay Reconfirm respondents Modify