The Subjective Precision of Computers: A Methodological Comparison with Human Coding in Content Analysis

Abstract

With the increased sophistication and speed of computers, fewer text content analysis projects utilize traditional human coding. This methodological case study was designed to compare results of human versus computer-assisted coding using the same data set. The two processes yielded significantly different results in a content analysis of newspaper coverage of a political campaign. In computer-assisted coding, several subjective steps must be taken to adapt the content to the program. Those decisions can be arbitrary and fall outside the concept of traditional intercoder reliability.