. 2017 Aug 24;12(8):e0182949. doi: 10.1371/journal.pone.0182949

Table 2. Intraclass correlations (ICC) for each PIL category in the three characteristic dimensions: form, thematic content, temporal focus.

PIL category	Percentage weighting (SD)	session level N = 38 sessions		level of utterances N = 8717
		ICC	95% CI	ICC	99% CI
Form
Repeating, paraphrasing, summarizing	19.18 (7.84)	.80	[.67, .89]	.70	[.68, .71]
Drawing attention to a behavioral and/or cognitive pattern	9.46 (3.61)	.52	[.28, .70]	.65	[.63, .66]
Highlighting discrepancies	3.33 (1.79)	.44^V	[.12, .67]	.54^V	[.52, .56]
Implicitly indicating a parallel	.67 (1.08)	.76	[.63, .86]	.64	[.63, .65]
Parallel without relationship context	.60 (.61)	.24^V	[.06, .45]	.36^V	[.35, .38]
Parallel regarding behavior towards oneself	.12 (.44)	.88	[.80, .93]	.58^V	[.56, .59]
Parallel regarding role reversal	.14 (.44)	.84	[.72, .91]	.68	[.67, .69]
Parallel regarding others and significant carers	.63 (1.04)	.92	[.86, .95]	.70	[.68, .71]
Parallel regarding relationships	1.54 (1.24)	.65	[.49, .78]	.58^V	[.57, .60]
Referring to the therapeutic relationship	1.60 (1.52)	.78	[.66, .87]	.65	[.64, .67]
Exploring	16.17 (5.84)	.90	[.84, .94]	.87	[.86, .87]
Adding new meaning	18.55 (4.78)	.43	[.14, .66]	.69	[.68, .70]
Creating causal links	2.22 (1.32)	.49^V	[.29, .67]	.51^V	[.50, .53]
Interpretation using metaphors	.60 (1.13)	.64	[.47, .77]	.53^V	[.51, .54]
Encouraging a view or impulse	1.07 (.98)	.30^V	[.10, .50]	.27^V	[.25, .29]
Validation	2.03 (1.54)	.68	[.53, .81]	.58^V	[.56, .59]
Suggestion	2.70 (2.68)	.72	[.57, .83]	.64	[.63, .65]
Self-disclosure	1.73 (2.03)	.68	[.52, .80]	.61	[.60, .63]
Association	1.89 (1.29)	.53^V	[.30, .71]	.53^V	[.51, .55]
Expression of emotional sympathy	.69 (.93)	.69	[.54, .81]	.51^V	[.50, .53]
Conveying professional knowledge	1.50 (5.45)	.97	[.94, .98]	.76	[.75, .77]
Other	10.65 (6.63)	.71	[.50, .84]	.60	[.58, .62]
Sentence fragments	2.85 (2.04)			.96	[.96, .96]
Single filler words	^*	.77	[.59, .87]	.72	[.71, .73]
*Mean form*		.72		.65
Superordinate category Drawing attention to parallels	.76 (.91)	.78	[.65, .87]	.71	[.70, .72]
Thematic content
Therapist	25.68 (24.95)	.87	[.80, .93]	.76	[.75, .77]
Current object	21.64 (23.30)	.90	[.84, .94]	.81	[.80, .81]
Mother	6.51 (9.69)	.98	[.97, .99]	.89	[.89, .90]
Father	4.34 (7.21)	.96	[.93, .98]	.84	[.84, .85]
Unspecified significant carer	.24 (.39)	.41^V	[.21, .60]	.36	[.34, .37]
Other significant carer	3.84 (12.82)	.95	[.92, .97]	.82	[.81, .82]
Abstract relationship behavior	13.92 (12.95)	.74	[.51, .86]	.66	[.63, .68]
Symptomatology	6.65 (9.91)	.89	[.82, .94]	.81	[.80, .81]
Other content without relationship or symptomology context	17.18 (15.21)	.81	[.70, .89]	.65	[.63, .66]
*Mean thematic content*		.90		.76
Temporal focus
Present	91.61 (10.37)	.88	[.80, .93]	.69	[.68, .71]
Childhood / adolescence	6.75 (9.49)	.91	[.83, .95]	.78	[.77, .79]
Symbol	1.02 (3.65)	.85	[.76, .91]	.72	[.71, .73]
Other temporal foci	.62 (1.21)	.13^V	[-.05, .35]	.10^V	[.08, .11]
*Mean temporal focus*		.79		.62

PIL = Psychodynamic Interventions List. ICC = Intraclass correlation coefficients (2,1) based on Shrout & Fleiss [30]. Classification = classification according to Fleiss [32]: >.75: excellent, .60-.75: good, .40-.59 moderate, < .40 poor. Percentage weighting = sum of the values of the category of interest divided by the sum of the values of all categories in the respective dimension across all sessions. SD = standard deviation. ^VReliability classified as less than ‘good’ which might be explained by small variance as the standard deviation was below-average when compared to the other categories of the same dimension.

*The percentage weighting of the category “Single filler words” was 43.34%. To better illustrate the distributions of interest this category was not included in the calculation of the percentage weightings of the other categories.