TextQuest - Software


Content Analysis

The main application of TextQuest is the content analysis. Search patterns for a category system can be words, any part of it, word sequences and word co-occurrences. Word co-occurrences are strings - up to six - that must occur all within a text unit. Distance and order can be set by parameters for each word co-occurence. Joker symbols like ? and * can also be used. Ambiguous and/or negated search patterns can be coded interactively at the screen, the coding process can be controlled by several files:

  • file of text units with potential ambiguous search patterns
  • file of uncoded text units
  • file of text units with negations
  • complete control of the coding process

TextQuest works with category labels that forces you to document the categories, but you benefit from this because the category labels are used in interactive coding and in some output files. While you are coding interactive, you see the categories with their labels and can code with a mouse click.



Negated search patterns are recognised, if there are indicators for negation before and/or after the search pattern. These indicators are stored in files and can be changed. Also multiple negation is detected. The differences between automatic and interactive coding can be computed by the ICRC (interactive coding reliability coefficient). Setups (syntax files/scripts) are generated for SAS, SPSS, SimStat and ConClus (cluster analysis) and make statistical analyses much easier, because all variables have labels (taken from the category labels). The file of the uncoded text units consists of all text units that do not contain a single search pattern; this file can be used for the extension of the category system.