codecountR: Counting Codes in a Text and Preparing Data for Analysis
Data analysis frequently requires coding, in particular when data are collected by interviews, by observations or even by questionnaires. Therefore, code counting and data preparation are necessary phases to carry out the analysis. Thus, the analysts will wish to count the codes inserted in a text (tokenization and counting of a list of pre-established codes) and to carry out the preparation of the data (feature scaling min-max normalization, Zscore, Box and Cox transformation, Yeo and Johnson transformation, median absolute deviation). For Box and Cox (1964) <https://www.jstor.org/stable/2984418> transformation, optimal Lambda is calculated by log-likelihood. The optimal lambda for Yeo and Johnson (2000) <doi:10.1093/biomet/87.4.954> transformation is calculated by correlation coefficient test inspired of Lye (1993) <doi:10.1139/l93-101>. Median absolute deviation is calculated on the basis of Leys et al (1993) <doi:10.1016/j.jesp.2013.03.013>. Package for educational purposes.
||Philippe Cohard [aut, cre]
||Philippe Cohard <p.cohard at laposte.net>
Please use the canonical form
to link to this page.