Lognormal distributions of user post lengths in Internet discussions - a consequence of the Weber-Fechner law?

EPJ Data Science

Table 1 Lognormal fit parameters to observed data

	Number of messages	μ	σ	Mode	Median	Mean
BBC, all messages	2,457,586	5.417	1.145	61	225	434
BBC, first messages	97,453	5.892	0.964	143	362	576
BBC, quotes	1,123,650	4.91	0.86	65	136	196
BBC, messages with quotes (best fit lognormal distribution is shifted by 189.6 bytes)	878,382	6.045	0.839	209	422	601
Poland, all messages	19,738	5.08	0.97	63	160	257
MySpace, all messages	18,874	4.278	1.042	24	72	124
YouTube, all messages	38,628	4.056	1.086	18	58	104
BBC, all messages, words	2,457,586	3.906	1.260	10.2	49.7	109.9
Poland, all messages, words	19,738	3.187	0.978	9.3	24.2	39.1

Best fit parameters of lognormal distributions and their characteristic measures for the analysed data. Except for the last two rows, the values are in bytes (characters).