Text length and the thematic concentration of text

The impact of text length very often biases results of stylometric indices which are based on rank-frequency distribution (e.g. type-token ratio, repeat rate, entropy). The aim of the article is to observe the relation between text size and thematic concentration indicators (TC, STC). The corpus consists of 1471 English texts of various genres. The obtained results show that thematic concentration is independent of text length in the interval . Given that the analysis corroborates the findings of the previous research in Czech language, TC and STC seem to be reliable stylometric indicators applicable to text analyses of different languages.