LIWC is the computer language analysis program which was developed around mid-1990 at U.S.A.. Previous langauge analysis program focused on content words in limited situation(e.g., psychotherapy), calculating emotional words, need and achievment words ...
LIWC is the computer language analysis program which was developed around mid-1990 at U.S.A.. Previous langauge analysis program focused on content words in limited situation(e.g., psychotherapy), calculating emotional words, need and achievment words. In contrast, LIWC focused on linguistic variables such as word length, articles, preposition, words per sentence based on about 3000 words that was used in daily life. In addtion, LIWC produce results exactly and very fast by computer word calculation. The previous research using this program suggest that only linguistic variables reflect various psychological variables but not content word variables. The process of developing Korean language analysis program can not be done just by borrowing English program, but there should be various consideration: First, Korean is concatenative langauge in which various morphemes are combined in a word. We should consider this kind of linguistic variation between two langauges. Also, there are various social and cultural differences between two languages except morphemic, syntactic differences. For example, Korean do not use first, second, and third pronouns often in a sentence, but delete them. Thus we should decide whether the program will recover them. Lastly, there are many homographs in various phrases, making us to analyze words more cautiously. Second, it is difficult to find Korean words corresponding one to one in English because semantic field and connotation are not same between two languages. Thus we should conduct research on semantic field and connotation in order to assign reliability on the variables. Third. LIWC is a very crude program in terms of its organization of software. Thus we developed a manager program that can connect various data, renewing existing data, and control its algorithm. Also we made the KLIWC as the public software that is installed on webspace, which can be accessed by any researcher simultaneously and fast. These processes provided basic stage for studing various psychological variables by langauge analysis, and provided basis to conduct research comparing cultural and linguistic factors between two langauges. The statistical reliability and validity test for this 1st version of KLIWC will be done in 2nd year of this project.