Implementation and evaluation of a teacher intervention program on classroom communication

Figures & data

Figure 1. Flow diagram based on the CONSORT 2010 Flow Diagram.

Table 1. Distribution of the 25 participating teachers’ years of teaching, number of certified teachers (teachers that completed their education) and age.

	Teachers teaching in grade 3	Teachers teaching in grade 4	Teachers teaching in grade 5	Teachers teaching in grade 6
N	11	10	2	2
Years of teaching
0–5	2	2
6–19	5	5	2	2
>20	4	3
Certified	10	9	2	2
Age
Mean	43.5	44.6	44.0	41.5
Range	34–63	27–57	41–47	36–47

Table 2. Overview of questionnaires with respective range and values from validation studies of which some are based on clinical populations.

Measure	Description	Direction of scores	Range	Values from previous studies
VHI11-VAS	Voice problems on 10 cm VA-scale	Lower = better	0–10	Patients: M = 4.38, SD = 3.12
				Controls: M = 1.43, SD = 1.98
				[2]
VHI11-sum	Sum of statements on VHI–11	Lower = better	0–44	Patients: M = 12.96, SD = 9.70
				Controls: M = 2.11, SD = 2.52
				[30]
PSQ index	Index from 0 = no stress to 1 = maximum stress	Lower = better	0–1	Low stress: ≤ 0.34
				Moderate stress: > 0.34 – ≤0.46
				High stress: > 0.46
				[45]
CBI total score	Mean of all 19 CBI-items	Lower = better	0–100	M = 33.3, SD = 17.3
				High degree of burnout > 50
				[33]
CBI subscale student-related	Mean of the 6 subscale items	Lower = better	0–100	M = 30.9, SD = 17.6
				[33]
CBI subscale work-related	Mean of the 7 subscale items	Lower = better	0–100	M = 33.0, SD =17.7
				[33]
CBI subscale personal-related	Mean of the 6 subscale items	Lower = better	0–100	M = 35.9, SD = 16.5
				[33]
TSES: subscale classroom management	Mean of all 8 TSES-items	Higher = better	1–9	Range of mean from five different countries:
				M = 6.59–7.65
				[46]
QPSNordic 34 + score	Mean of all 37 items	Lower = better	1–5	The long version QPS Nordic is validated, but not its short form: QPS Nordic 34+.

Table 3. Values from 0 weeks pre-training for all measures presented with mean, standard deviation, and number of responding participants.

Measure	M	SD	n
VHI11-VAS	1.04	1.30	21
VHI11-sum	1.17	1.72	23
PSQ index	0.28	0.13	23
CBI total score	28.11	12.60	22
CBI subscale student-related	26.52	12.70	22
CBI subscale work-related	28.11	14.38	23
CBI subscale personal-related	32.79	18.39	23
TSES: subscale classroom management	7.76	0.70	23
QPSNordic 34 + score	3.53	0.20	17

Figure 2. Aggregated data from all teachers participating in the study. Points correspond to mean values and confidence intervals are the same as in Table 4, for the respective scales (see titles of sub-panels). The dotted line through the first measure point (Pre-training) indicates the baseline so that changes at the other points are more clearly visible. Note that confidence intervals overlap the baseline for some effects that are nevertheless significant (cf. Table 4). This is due to small differences between predicted values and calculated means, as the model included covariates.

Table 4. Regression (LME) coefficients and confidence intervals for the analysis including all participants, all scales (subpanels a–f) and all measurements (except the delayed training-group’s five weeks pre-test) relative to training.

	B	CI	B	CI	B	CI
	(a) VHI11 – VAS (log)		(b) VHI11 – sum (log)		(c) PSQ – index
(Intercept)	2.45	1.78, 3.12	.78	.50, 1.06	.33	.27, .40
Post 0 weeks	.41	−.20, 1.02	−.04	−.33, .24	−.01	−.05, .03
Post 5 weeks	−.05	−.50, .40	−.32	−.70, .05^†	−.05	−.08, −.01*
Post 3 months	−.07	−.47, .33	−.32	−.54, −.10*	−.03	−.07, .00^†
WT = 2016 w43^1	−1.77	−2.66, −.88*	−.59	−.89, −.30*	−.14	−.23, −.04*
WT = 2017 w07^1	−.98	−2.17, .21	−.29	−.65, .07	−.09	−.22, .03
	(d) CBI – score		(e) TSES – mean		(f) QPS – score
(Intercept)	36.27	29.17, 43.36	7.56	7.21, 7.91	3.51	3.38, 3.63
Post 0 weeks	.84	−4.71, 6.40	.06	−.16, .28	−.06	−.16, .05
Post 5 weeks	−5.03	−8.84, −1.21*	.27	.03, .50*	−.01	−.11, .09
Post 3 months	−3.03	−6.35, .28^†	.16	−.07, .39	−.03	−.12, .07
WT = 2016 w43^1	−15.29	−27.01, −3.57*	.33	−.16, .83	.09	−.08, .26
WT = 2017 w07^1	−13.01	−27.34, 1.31^†	.56	−.07, 1.19^†	−.11	−.32, .10

Note: *p < .05, ^†p < .10. WT = week of training. ¹Reference = 2016 week 38.

The week of training (WT) is included as a covariate with 2016 week 38 (first sub-group) as baseline. Measure points Post 0 weeks, Post 5 weeks, and Post 3 months are thus compared against the pre-training measure point, and confidence intervals excluding 0 indicate significant training effects (also marked by *).

Lyberg-Åhlander V, Rydell R, Löfqvist A. Speaker's comfort in teaching environments: voice problems in Swedish teaching staff. J Voice. 2011;25:430–440.

 PubMed Web of Science ®Google Scholar

Nyman M, Åradsson S. Voice Handicap Index-Lund-11. Ett självskattningsformulär för patienter med röstbesvär. Översättning, bearbetning och validering. [Voice Handicap Index-Lund-11. A self-assessment form for patients with voice problems. Translation, adaptation and validation] [master’s thesis]. Lund (Sweden): Lund University; 2008.

 Google Scholar

Bergdahl J, Bergdahl M. Perceived stress in adults: prevalence and association of depression, anxiety and medication in a Swedish population. Stress Health. 2002;18:235–241.

 Web of Science ®Google Scholar

Kristensen TS, Borritz M, Villadsen E, et al. The Copenhagen burnout inventory: a new tool for the assessment of burnout. Work Stress. 2005;19:192–207.

 Web of Science ®Google Scholar

Klassen RM, Bong M, Usher EL, et al. Exploring the validity of a teachers’ self-efficacy scale in five countries. Contemp Educ Psychol. 2009;34:67–76.

 Web of Science ®Google Scholar