TABLE 3.

LIWC word category differences describing changes in dream experiences between individuals with new‐onset insomnia, pre‐existing insomnia and good sleepers during the COVID‐19 pandemic

			New‐onset insomnia dreams (n = 369)	Pre‐existing insomnia dreams (n = 240)	Good sleeper dreams (n = 302)	New‐onset and pre‐existing insomnia	New‐onset insomnia and good sleepers	Pre‐existing insomnia and good sleepers
Word type	Kruskal–Wallis chi squared test	Adj. p‐value	Mrank	Mrank	Mrank	Post hoc Dunn test (Z‐value, adj. p‐value)	Post hoc Dunn test (Z‐value, adj. p‐value)	Post hoc Dunn test (Z‐value, adj. p‐value)
Negative emotion	34.03	< 0.001***	13.80	11.14	6.33	−1.72, p = 0.086	−5.78, p < 0.001***	−3.54, p < 0.001***
Affect words	23.20	< 0.001***	14.34	12.22	7.58	−1.07, p = 0.284	−4.71, p < 0.001***	−3.20, p = 0.002**
Leisure	20.01	< 0.001***	12.94	11.26	16.02	−1.17, p = 0.241	3.42, p < 0.001***	4.20, p < 0.001***
Anxiety	12.37	0.021*	3.87	2.61	1.21	−1.52, p = 0.128	−3.52, p = 0.001**	−1.70, p = 0.128
Death	10.33	0.031*	1.02	0.60	0.46	−1.56, p = 0.169	−3.20, p = 0.004**	−1.38, p = 0.169

Note: This table displays mean rank differences of the significant LIWC word category differences describing changes in dreaming between groups.

^*** p < 0.001.

^** p < 0.01.

^* p < 0.05, adj. p‐value = adjusted p‐value using the Benjamini and Hochberg (1995) correction for multiple comparisons, where critical value cut‐off scores were converted to adjusted p‐values using the method described by Benjamini, Heller, Yekutieli (2009). For Dunn's test post hoc comparisons, corrections were applied within each word category, not all word categories. Only results that reached statistical significance are displayed in this table. Please refer to Table S1 for 30 LIWC category results.