Data processing is an important part of computer-assisted reporting. Several major steps are involved in data processing. These include general database design, data matrix design, data coding, and data editing. Experienced data handlers call this data cleaning. Data cleaning must take place before any serious publication-bound analysis occurs. Most researchers know from experience that, regardless of the level of care used in entering data, some errors occur. The point of cleaning data is to eliminate as many of those errors as possible. For large databases with hundreds of fields, they are essential to explain the contents of the data matrix. For individuals building their own databases with heavy dependence on codes, a layout summary and a codebook are just as essential as they are for databases already in electronic form that are obtained from public sources. Variable measurement is essential to building any database. Variables must be evaluated in terms of their attributes.