The psychometric properties of the Turkish version of the Pre-sleep Arousal Scale

Abstract

Purpose

The aim of the present study was to examine the psychometric properties of the Turkish version of the Pre-sleep Arousal Scale (PSAS), which measures pre-sleep arousal, a significant predictor of insomnia symptoms.

Methods

651 participants were recruited via social media and the Internet. Confirmatory factor analysis was conducted in the total sample (65.28% females; M_age1 = 28.09 ± 14.00). Convergent, divergent, incremental, and known-groups validity and internal consistency coefficients were assessed in a subsample of 556 participants (62.77% females; M_age2 = 29.25 ± 14.81). A second separate sample of 88 participants (80.68% females; M_age3 = 22.19 ± 4.98) was used to evaluate three-week test–retest reliability.

Results

The results of factor analysis confirmed the two-factor structure of the Turkish PSAS with cognitive (PSAS-C) and somatic (PSAS-S), similar to the original scale. The correlations of the PSAS with convergent and divergent measures showed that the Turkish form had good convergent and acceptable divergent validity. PSAS-C and PSAS-S were able to explain an 18% additional variance in insomnia severity beyond depression and anxiety, an 18% additional variance in depression beyond insomnia severity, and a 35% additional variance in anxiety beyond insomnia severity. Moreover, insomnia patients had significantly higher PSAS-C and PSAS-S scores than good sleepers. Finally, the PSAS, PSAS-C, and PSAS-S had satisfactory internal consistency coefficients (α = 0.92, 0.91, and 0.86, respectively) and three-week test–retest correlations (ICC = 0.82, 0.82, and 0.71, respectively).

Conclusion

The Turkish form of the PSAS was a valid and reliable measure of pre-sleep arousal and can be utilized in sleep studies.

Introduction

Although insomnia can be defined as an inability to sleep or the presence of sleep problems, insomnia disorder is regarded as a distinct sleep problem characterized by three symptoms: inability to fall asleep, inability to maintain sleep, and early morning awakenings [1]. As shown in several studies, 6.4–19.2% of the adult population met the DSM-IV criteria for insomnia disorder [1, 2–3]. Despite being a risk factor for many psychiatric disorders [4], insomnia comes with several personal and societal costs, including decreased quality of life [5], decreased daytime functioning [6], and impaired work performance [7]. Insomnia models and prior research have revealed that cognitive and physiological hyperarousal contribute to the vulnerability to and maintenance of insomnia disorder [8]. It has been hypothesized that the hyper-aroused states of insomnia patients during the pre-sleep phase disrupt the onset and maintenance of sleep as if the sleep system is halted by the activation of the central nervous system [8]. Therefore, investigating the role of hyperarousal in insomnia disorder may lead to a better understanding of the disorder and contribute to the development of treatments for it. Indeed, this necessitates the use of valid and reliable measurement tools.

The Pre-sleep Arousal Scale (PSAS) [9] is one of the most frequently utilized sleep medicine tools for measuring pre-sleep arousal in cognitive and somatic dimensions. Cognitive arousal corresponds to excessively negative mental activity, such as worrying and ruminating, the presence of a racing mind, and the inability to stop thinking [9]. Somatic arousal refers to the physiological responses of the body to threats, such as an elevated heart rate, increased breathing rate, increased glucose release, and increased perspiration [9, 10]. So far, the evidence has shown that insomnia patients have higher cognitive and somatic pre-sleep arousal scores than good sleepers, and higher pre-sleep arousal is associated with longer sleep-onset latency, decreased total sleep time, more frequent night and early morning awakenings from sleep, reduced sleep quality, higher daytime impairment, and higher insomnia severity [9, 11, 12–13]. Since there have been no prior attempts to adapt the PSAS to Turkish, little is known about its psychometric properties in the Turkish-speaking population. The purpose of the present study was to investigate the validity and reliability of the PSAS in samples of Turkish speakers.

Materials and method

Participants

The study data was collected from 651 individuals recruited via social media and the Internet. Confirmatory factor analysis was conducted in the total sample. In a subsample of 556 individuals who completed all the questionnaires in the study, convergent, divergent, incremental, and known-groups validity were assessed. In addition, a second separate sample (N = 88) was utilized to examine the 3-week test–retest reliability. Figure 1 and Table 1 show the demographic details of the samples and the samples of the study, respectively.

[See PDF for image]

Fig. 1

Flowchart showing the samples of the study

Table 1. The demographic characteristics of the samples

	Total Sample		Subsample		Second Sample
	M	SD	M	SD	M	SD
Age	28.09	14.00	29.25	14.81	22.19	4.98
	N	%	N	%	N	%
Gender
Female	425	65.28	349	62.77	71	80.68
Male	226	34.72	207	37.23	17	19.32
Education
High School	15	2.30	15	2.70	0	0
University	577	88.63	482	86.69	100	100
Graduate	59	9.06	59	10.61	0	0
Employment
No	434	66.67	434	78.06	100	100
Yes	217	33.33	122	21.94	0	0
Marital Status
Single	516	79.26	421	75.72	83	94.31
Married	126	19.36	126	22.66	5	5.68
Divorced	9	1.38	9	1.62	0	0
Socioeconomic Status
Very low	35	5.38	31	5.58	1	1.14
Low	94	14.44	80	14.39	10	11.36
Middle	435	66.82	370	66.55	57	64.77
High	85	13.06	73	13.13	19	21.59
Very high	2	0.31	2	0.36	1	1.14

Instruments

Demographic information form

A demographic information form was used to examine participant characteristics. Included were questions about the participants' age, gender, education level, employment status, marital status, and socioeconomic status.

Pre-sleep Arousal Scale (PSAS)

The Pre-sleep Arousal Scale (PSAS) was developed by Nicassio et al. [9] to assess levels of cognitive and somatic arousal in the pre-sleep phase. Sixteen items of the scale are rated on a 5-point Likert-type scale, ranging from 1 (not at all) to 5 (extremely). The PSAS consists of two factors, each measuring pre-sleep cognitive (PSAS-C) and somatic (PSAS-S) arousal with eight items. Higher scores from the scale imply higher arousal.

Beck Anxiety Inventory (BAI)

Beck, Epstein, Brown, and Steer [14] developed the Beck Anxiety Inventory (BAI) to assess the severity of anxiety symptoms. Two factors (somatic and subjective anxiety-panic) are measured by 21 items rated on a 4-point Likert-type scale, ranging from 0 (not at all) to 3 (seriously). Higher scores indicate increased anxiety levels. The BAI was adapted into Turkish by Ulusoy, Şahin, and Erkmen [15]. In the current study, the somatic subscale of BAI (BAI-Somatic) was used to investigate the convergent validity of PSAS-S.

Ruminative Thought Style Questionnaire (RTSQ)

Brinker and Dozois [16] developed the Ruminative Thought Style Questionnaire (RTSQ) to measure ruminative thought style independently of depression. Twenty items are used to assess positive, negative, and neutral thoughts, as well as past and future-oriented ruminative thoughts. The RTSQ is rated on a 7-point Likert-type scale, ranging from 1 (not at all) to 7 (very well). Higher scores on the scale reflect a higher propensity for ruminative thought. The RTSQ was adapted by Karatepe [17] into Turkish. In the current study, the scale was utilized to evaluate the convergent validity of the PSAS-C.

Insomnia Severity Index (ISI)

Bastien, Vallières, and Morin [18] developed the Insomnia Severity Index (ISI) to create a brief screening measure of insomnia severity. The ISI consists of seven items covering insomnia symptoms as well as subjective satisfaction, impairment, and distress regarding sleep problems. The scale is rated on a 5-point Likert-type scale. Higher scores indicate more severe cases of insomnia. The ISI was adapted into Turkish by Boysan, Güleç, Beşiroğlu, and Kalafat [19]. In the current study, the Turkish version of ISI was utilized to evaluate the convergent validity of the PSAS as well as the incremental validity of PSAS-C and PSAS-S beyond depression, anxiety, and insomnia severity.

Depression Anxiety Stress Scale (DASS-21)

Lovibond and Lovibond [20] developed a 21-item version of the Depression Anxiety Stress Scale (DASS-21). The DASS-21 subscales consist of seven items assessing depression, anxiety, and stress. The items are rated on a 4-point Likert-type scale, ranging from 0 (did not apply to me at all) to 3 (applied to me very much or most of the time). Higher ratings on each subscale imply higher symptom severity. Yıldırım, Boysan, and Kefeli [21] adapted the DASS-21 to Turkish. In the current study, the depression and anxiety subscales of the DASS-21 were used to assess the incremental validity of the PSAS. The internal consistency coefficients for the depression and anxiety subscales of the DASS-21 were 0.89 and 0.83.

Gender Role Attitudes Scale (GRAS)

García-Cueto et al. [22] developed the Gender Role Attitudes Scale (GRAS) to assess gender role attitudes from the viewpoint of gender equality. Twenty items are rated on a 5-point Likert-type scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Higher scores indicate a more substantial commitment to gender equality. Bakioğlu and Türküm [23] adapted the scale to Turkish. The GRAS was used to evaluate the divergent validity of the PSAS, as it measures a construct conceptually unrelated to pre-sleep arousal.

Procedure

After receiving the approval for the scale adaptation from the author of the original scale, the ethical approval for the study was obtained from the Middle East Technical University’s Human Subjects Ethics Committee. For scale translation, a double-translation and reconciliation procedure was utilized [24]. The scale was translated into Turkish by two senior psychology researchers, and the translations were then combined by a bilingual (English-Turkish) scholar. Twenty-five subjects participated in a pilot study that evaluated the translations' clarity. The final version was established based on the feedback from the pilot study. All scales were presented online via the Qualtrics Survey System in counterbalanced order, meaning that the measures of the study (excluding the demographics form) were presented to participants in random order. Participants were informed of the study's aims, confidentiality policies, and their right to decline or discontinue participation at any time during the study. The study took an average of 20 min to complete.

Statistical analysis

The data analyses were conducted in SPSS 25 and JASP 0.17.1. Multiple outliers were identified by the use of Mahalanobis distance. The analyses assessing the validity and reliability of the PSAS can be seen in Table 2.

Table 2. Analyses to investigate the psychometric properties of the Turkish PSAS

Validity	Analysis	Indicators
Construct validity	Confirmatory Factor Analysis	Sample size ≥ 200 [25]
		Insignificant χ2 test [26]
		SRMR ≤ .08 [26]
		RMSEA ≤ .10 [27]
		TLI ≥ .90 [28]
		CFI ≥ .90 [28]
Convergent validity	Pearson Correlation	Small to high significant correlations [29]
Divergent validity	Pearson Correlation	No significant correlation or small significant correlations [29]
Incremental validity	Hierarchical Regression	Significantly explained variance beyond the control variable [29]
Known-groups validity	Student’s T-test, Welch's T-test	A significant difference [29]
Reliability	Analysis	Indicators
Internal consistency	Cronbach’s Alpha	α ≥ .70 [30]
Test–retest reliability	ICC	Sample size ≥ 66 [31]
		ICC ≥ .75 [32]

Note.SRMS Standardized Root-Mean-Square Residual, RMSEA Root-Mean-Square Error of Approximation, CFI Comparative Fit Index, TLI Tucker-Lewis Index, ICC Intra-Class Correlation

Results

Confirmatory factor analysis

To validate the Turkish PSAS's factor structure, a confirmatory factor analysis was conducted in the total sample. Eighteen outliers were eliminated from the subsequent analysis. The final sample included 633 individuals. A two-factor model with 16 items based on the original scale was hypothesized. Eight items were entered as PSAS-C indicators, while the remaining eight were added as the PSAS-S indicators. Due to the lack of multivariate normality (Mardia’s Z = 18.20, p < 0.001), confirmatory factor analysis was conducted with maximum likelihood parameter estimates, Satorra-Bentler correction, and robust error calculation. The goodness-of-fit indices indicated that the two-factor model barely fit the data (χ²(103)  = 702.04, p < 0.001, SRMR = 0.06, RMSEA = 0.10, 90% CI = [0.09, 0.10], TLI = 0.88, CFI = 0.90). Several error covariances were added to the initial model (items 6–7, items 3–4, and items 9–11) to improve it, as recommended by modification indices. Each modification significantly contributed to the model's fit. The final model fit the data adequately (χ²(100) = 480.80, p < 0.001, SRMR = 0.05, RMSEA = 0.08, 90% CI [0.07, 0.09], TLI = 0.92, CFI = 0.94) (see Fig. 2).

[See PDF for image]

Fig. 2

The confirmatory factor analysis results of the two-factor model of the Turkish PSAS. Note. *p < 0.001. All reported estimates are standardized

Convergent and divergent validity

Convergent and divergent validity were examined by assessing Pearson correlations between the PSAS, PSAS-C, PSAS-S, BAI-Somatic, the RTSQ, and the ISI in the subsample. PSAS-C and the RTSQ, PSAS-S, and BAI-Somatic, and the PSAS and the ISI had significant and moderate to high positive correlations. However, there were also small but significant correlations between the PSAS, PSAS-C, PSAS-S, and GRAS. The correlations between the study variables, descriptives, and internal consistency coefficients can be seen in Table 3.

Table 3. The correlations, descriptives, and internal consistency values of the scales

Scales	1	2	3	4	5	6		7	8	9	10
1. PSAS	-
2. PSAS-C	0.92**	–
3. PSAS-S	0.87**	0.61**	–
4. RTSQ	0.67**	0.71**	0.46**	–
5. BAI-Somatic	0.59**	0.43**	0.67**	0.38**	–
6. GRAS	0.25**	0.28**	0.15**	0.26**	0.11**	–
7. ISI-1	0.58**	0.60**	0.43**	0.38**	0.33**	0.12*		–
8. ISI-2	0.42**	0.37**	0.39**	0.20**	0.31**	− 0.03		0.53**	–
9. ISI-3	0.24**	0.18**	0.26**	0.08	0.21**	− 0.08		0.20**	0.48**	–
10. ISI	0.65**	0.62**	0.54**	0.42**	0.42**	0.08		0.79**	0.68**	0.46**	–
Mean	38.94	23.38	15.57	93.01	4.52	61.28		1.44	0.89	0.82	8.80
SD	12.65	7.86	6.22	31.35	4.19	10.40		1.22	1.04	1.12	5.20
Min	16.00	8.00	8.00	20.00	0.00	30.00		0.00	0.00	0.00	0.00
Max	76.00	40.00	37.00	160.00	22.00	75.00		4.00	4.00	4.00	26.00
Cronbach’s α	0.92	0.91	0.86	0.96	0.81	0.89		–	–	–	0.80

Note 1. *p < 0.01, **p < 0.001

Note 2.PSAS Pre-sleep Arousal Scale, PSAS-C Pre-sleep Arousal Scale—Cognitive Subscale, PSAS-S Pre-sleep Arousal Scale—Somatic Subscale, RTSQ Ruminative Thought Style Questionnaire, BAI-Somatic Beck Anxiety Inventory—Somatic Subscale, GRAS Gender Role Attitude Scale, ISI Insomnia Severity Index, ISI-1 Difficulty falling asleep, ISI-2 Difficulty staying asleep, ISI-3 Waking up too early

Incremental validity

Considering the bidirectional relationships between depression/anxiety and insomnia, three two-step hierarchical multiple regressions were conducted in the subsample to examine the incremental validity of cognitive and somatic pre-sleep arousal. In the first hierarchical multiple regression, it was tested whether cognitive and somatic pre-sleep arousal can predict insomnia severity beyond depression and anxiety. Two further hierarchical multiple regressions were also conducted to examine whether cognitive and somatic pre-sleep arousal predict depression and anxiety beyond insomnia severity. One outlier was removed from further analysis. The final sample consisted of 555 participants. Tolerance and VIF values indicated no multicollinearity for the regression analyses. The results showed that while predicting insomnia severity, adding PSAS-C and PSAS-S explained additional variance beyond depression and anxiety (ΔR² = 0.18, p < 0.001). Moreover, while predicting depression and anxiety, PSAS-C and PSAS-S accounted for additional variances beyond insomnia severity (ΔR² = 0.18, p < 0.001, ΔR² = 0.35, p < 0.001, respectively). The summaries of hierarchical multiple regressions can be seen in Tables 4, 5, and 6.

Table 4. Hierarchical regression analysis summary for cognitive and somatic pre-sleep arousal predicting insomnia severity beyond depression and anxiety

Model	F (df1, df2)	Adjusted R2	B	SE	β	t	p	Tolerance	VIF
Step 1	89.31 (2, 552)	.24					< 0.001
Intercept			5.06	0.35		14.38	< 0.001
Depression			0.22	0.05	0.22	4.47	< 0.001	0.57	1.75
Anxiety			0.41	0.06	0.32	6.61	< 0.001	0.57	1.75
Step 2	101.57 (4, 550)	.42					< 0.001
Intercept			− 1.39	0.59		− 2.36	0.02
Cognitive Arousal			0.29	0.03	0.43	9.97	< 0.001	0.55	1.81
Somatic Arousal			0.19	0.04	0.23	4.31	< 0.001	0.38	2.61
Depression			0.05	0.05	0.05	1.06	0.29	0.51	1.95
Anxiety			0.05	0.07	0.04	0.65	0.52	0.35	2.85

Table 5. Hierarchical regression analysis summary for cognitive and somatic pre-sleep arousal predicting depression beyond insomnia severity

Model	F (df1, df2)	Adjusted R2	B	SE	β	t	p	Tolerance	VIF
Step 1	125.30 (1, 553)	.18					< .001
Intercept			4.11	0.39		10.53	< .001
Insomnia Severity			0.43	0.04	0.43	11.19	< .001	1.00	1.00
Step 2	106.68 (3, 551)	.36					< .001
Intercept			− 1.61	0.54		− 2.81	0.005
Insomnia Severity			0.07	0.05	0.07	1.49	0.14	0.58	1.73
Cognitive Arousal			0.23	0.03	0.35	7.29	< .001	0.51	1.94
Somatic Arousal			0.23	0.04	0.28	6.34	< .001	0.59	1.70

Table 6. Hierarchical regression analysis summary for cognitive and somatic pre-sleep arousal predicting anxiety beyond insomnia severity

Model	F(df1, df2)	Adjusted R2	B	SE	β	t	p	Tolerance	VIF
Step 1	153.34 (1, 553)	0.22					< 0.001
Intercept			1.63	0.30		5.38	< 0.001
Insomnia Severity			0.37	0.03	.47	12.38	< 0.001	1.00	1.00
Step 2	239.95 (3, 551)	0.56					< 0.001
Intercept			− 3.48	0.38		− 9.22	< 0.001
Insomnia Severity			0.04	0.03	.05	1.24	0.22	0.58	1.73
Cognitive Arousal			0.06	0.02	.11	2.83	0.005	0.51	1.94
Somatic Arousal			0.43	.02	.65	17.82	< 0.001	0.59	1.70

Known-groups validity

It was tested whether there were significant differences between the scores of good sleepers and insomnia patients on the PSAS subscales in the subsample. The ISI scores (≥ 8) were used to differentiate good sleepers from insomnia patients [18]. Two independent t-tests in the subsample were performed. PSAS-C scores of good sleepers (M = 19.03, SD = 6.91, N = 239) and insomnia patients (M = 26.65, SD = 6.89, N = 317) differed significantly; t(554) = − 12.84, p < 0.001, d = − 1.11. Due to the violation of the assumption of equality of variances, a Welch t-test was performed for PSAS-S. There was a significant difference between the PSAS-S scores of good sleepers (M = 12.62, SD = 4.69, N = 239) and insomnia patients (M = 17.79, SD = 6.32, N = 317); t(554) = -10.63, p < 0.001, d = -0.91. The result indicated that the Turkish version of PSAS showed satisfactory known-groups validity.

Internal consistency and test–retest reliability

The reliability analysis conducted in the subsample (N = 556) yielded that Cronbach’s Alpha coefficients were 0.92 (the PSAS), 0.91 (PSAS-C), and 0.86 (PSAS-S). In a second sample of 88 individuals, the three-week test–retest reliability of the PSAS was assessed. The intra-class correlation coefficients between the baseline and 21-day follow-up scores were found as 0.82 (p < 0.001) for the PSAS, 0.82 (p < 0.001) for PSAS-C, and 0.71 (p < 0.001) for PSAS-S. The findings revealed that the Turkish PSAS had good reliability.

Discussion

The present study aimed to adapt the PSAS, a well-known and frequently used self-report scale of cognitive and somatic pre-sleep arousal, to Turkish. A confirmatory factor analysis was conducted to validate the two-factor structure with 16 items of the original scale [9]. After adding correlations between the errors of several items, the confirmatory factor analysis results indicated that the two-factor model with 16 items had an adequate fit for the data. Despite having different total numbers of items, Swedish, German, Pakistani, Portuguese, and Japanese adaptation studies found a two-factor structure for the PSAS [11, 12–13, 33, 34]. Among the adaptations of the PSAS to different languages, only the Japanese adaptation study [33] included a confirmatory factor analysis and similarly found adequate fit values for the original two-factor structure with 16 items.

The correlations between PSAS-C, PSAS-S, the RTSQ, BAI-Somatic, and the ISI were used to test the convergent validity of the PSAS. Although the items of the RTSQ are not specific to the pre-sleep phase, they pertain to ruminative thinking style or intrusive thoughts and therefore correspond to the cognitive mental activity items of PSAS-C. Similarly, BAI-Somatic consists of items that can be regarded as somatic manifestations of anxiety. In this sense, the items of PSAS-S overlap with the items of BAI-Somatic. The association between insomnia and pre-sleep arousal has long been argued and studied, therefore, the correlation between the ISI and the PSAS was also investigated for convergent validity [8]. The high correlations between the RTSQ and PSAS-C (r = 0.71), BAI-Somatic and PSAS-S (r = 0.67), and the ISI and PSAS (r = 0.65) implied that the PSAS had good convergent validity. In line with the previous studies [12, 13, 33], pre-sleep arousal was more strongly associated with difficulty falling asleep than difficulty staying asleep or early morning awakenings. In terms of divergent validity, it was anticipated that the GRAS would not be significantly or highly correlated with the PSAS and its subscales. Nevertheless, the findings revealed small positive correlations between the GRAS and the PSAS (r = 0.25), PSAS-C (r = 0.28), and PSAS-C (r = 0.15). Rönkkö and Cho [29] asserted that the divergent validity must be evaluated by not only statistical testing but also considering the context and relevant theories of the measures. The PSAS assesses cognitive and somatic arousal in the pre-sleep phase, while the GRAS measures attitudes toward gender roles. Both the context and theories of these constructs are clearly distinct. In light of the context and theories underlying the two scales, the small correlations between the GRAS and the subscales of the PSAS may therefore support the divergent validity of the PSAS.

Hierarchical multiple regression analyses were conducted to evaluate the incremental validity of the PSAS. The results yielded that PSAS-C and PSAS-S could explain additional variance in insomnia severity beyond depression and anxiety and additional variances in depression and anxiety beyond insomnia severity. Previous studies found that insomnia severity and pre-sleep arousal were significantly associated with depression and anxiety [12, 36]. In the current study, depression and anxiety were not significant predictors of insomnia severity after cognitive and somatic pre-sleep arousal were added to the regression model. Similarly, insomnia severity was not a significant predictor of depression and anxiety when cognitive and somatic pre-sleep arousal were added to the models. Considering the bidirectional relationship between depression, anxiety, and insomnia [37, 38], these findings suggest that cognitive and somatic pre-sleep arousal may be mediators of the relationship between depression/anxiety and insomnia severity. In this context, there are some clinical implications to consider. Regarding pre-sleep arousal stemming from depression and anxiety, targeting cognitive pre-sleep arousal characterized by rumination and worry with cognitive and mindfulness-based interventions (e.g., cognitive restructuring and mindfulness meditation) and somatic pre-sleep arousal with relaxation techniques (e.g., deep breathing and progressive muscle relaxation) may effectively lessen insomnia severity [36, 39, 40]. On the other hand, behavioral techniques and pharmacotherapy may be preferable to target pre-sleep arousal caused by insomnia and relieve symptoms of depression and anxiety. Insomnia, especially sleep-onset insomnia, may extend the pre-sleep time in bed, and this period eventually can contribute to one’s engagement in rumination and worry [41]. Therefore, behavioral techniques (e.g., stimulus control and sleep restriction) that limit pre-sleep time in bed may counteract this negative effect of insomnia [42]. Regarding somatic pre-sleep arousal evoked by insomnia, pharmacotherapeutic agents affecting the central nervous system and HPA axis (e.g., benzodiazepine receptor agonists or selective melatonin receptor agonists) can be used to target somatic arousal since physiological hyperarousal in insomnia is related to increased central nervous system activity and HPA activation [43, 44–45].

In terms of known-groups validity, insomnia patients (ISI ≥ 8) were shown to have significantly higher PSAS-C and PSAS-S scores than good sleepers, similar to the original study [9], German [11], and Portuguese [13] adaptation studies. Internal consistency and test–retest analyses were used to evaluate the PSAS's reliability. The Cronbach's Alpha values of the PSAS (α = 0.92), PSAS-C (α = 0.91), and PSAS-S (α = 0.86) were greater than the suggested values of 0.70-0.80 [30]. Three-week test–retest intra-class correlations of the PSAS (ICC = 0.82) and PSAS-C (ICC = 0.82) were higher than the recommended value of 0.75, indicating good reliability, while the correlation of PSAS-S (ICC = 71) was higher than 0.50 indicating moderate reliability [46]. The overall findings implied that the Turkish PSAS had good reliability.

The present study was not without limitations. First, the sample's male-to-female ratio was biased toward females, and most of the participants were university students. Therefore, future studies should be conducted with larger and more representative samples. Second, the modification indices were utilized to determine which residual variances in the same latent factor are correlated. Correlated errors lead to neglecting variables not included in the model but which themselves cause errors [40]. However, Bollen and Lennox [48] argued that correlated errors could result from similar wording and adjacent items. Analysis of item dyads indicated that correlated errors may be a result of wording and being adjacent for Items 6–7, while a result of being adjacent for Items 3–4 and 9–11. Third, the selection of gender role attitudes to assess divergent validity was not strictly theory based. It was used since it is regarded as a construct conceptually unrelated to pre-sleep arousal. Finally, the participants in the known-groups validity analysis were categorized based on their ISI scores. A more robust way to diagnose people with insomnia disorder may be to use clinical interviews with psychiatrists.

To conclude, the findings indicated that the 16-item Turkish version of the PSAS was a valid and reliable measure of pre-sleep arousal. There may be several ways of utilizing the PSAS in sleep studies. It can be used in experimental and observational sleep and behavioral medicine studies that examine pre-sleep arousal. In addition, future studies can be conducted to determine threshold values for the PSAS scores, which can be used to predict the diagnosis of chronic insomnia and to distinguish chronic insomnia patients from patients with other disorders in the Turkish population, as done in a previous study for the USA population [36]. Finally, the PSAS scores can serve as outcome measures for intervention studies with Turkish-speaking participants that target pre-sleep arousal to alleviate insomnia, sleep difficulties, or psychiatric symptoms in insomnia patients [49, 50].

Author contributions

KKT and DCÇ contributed to the study conception and design. Material preparation, data collection, and analyses were performed by KKT. The first draft of the manuscript was written by KKT, and all authors commented on previous versions. All authors read and approved the final manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for- profit sectors.

Declarations

Conflicts of interest

The authors declare that they have no conflicts of interest.

Ethics approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Ethical committee permission

Permission Number: The study was reviewed and approved by Middle East Technical University’s Human Subjects Ethics Committee with approval number 259-ODTU-2020.

Authors’ note

The article was generated from the Ph.D. thesis that the first author produced under the supervision of the second author.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Ohayon, MM; Sagales, T. Prevalence of insomnia and sleep characteristics in the general population of Spain. Sleep Med; 2010; 11, 10 pp. 1010-1018. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21093362][DOI: https://dx.doi.org/10.1016/j.sleep.2010.02.018]

2. Cao, XL; Wang, SB; Zhong, BL; Zhang, L; Ungvari, GS; Ng, CH et al. The prevalence of insomnia in the general population in China: a meta-analysis. PLoS ONE; 2017; 12, 2 [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28234940][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5325204][DOI: https://dx.doi.org/10.1371/journal.pone.0170772] e0170772.

3. Ford, ES; Cunningham, TJ; Giles, WH; Croft, JB. Trends in insomnia and excessive daytime sleepiness among US adults from 2002 to 2012. Sleep Med; 2015; 16, 3 pp. 372-378. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25747141][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4763609][DOI: https://dx.doi.org/10.1016/j.sleep.2014.12.008]

4. Hertenstein, E; Feige, B; Gmeiner, T; Kienzler, C; Spiegelhalder, K; Johann, A et al. Insomnia as a predictor of mental disorders: a systematic review and meta-analysis. Sleep Med Rev; 2019; 1, 43 pp. 96-105. [DOI: https://dx.doi.org/10.1016/j.smrv.2018.10.006]

5. LeBlanc, M; Beaulieu-Bonneau, S; Mérette, C; Savard, J; Ivers, H; Morin, CM. Psychological and health-related quality of life factors associated with insomnia in a population-based sample. J Psychosom Res; 2007; 63, 2 pp. 157-166. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17662752][DOI: https://dx.doi.org/10.1016/j.jpsychores.2007.03.004]

6. Léger, D; Morin, CM; Uchiyama, M; Hakimi, Z; Cure, S; Walsh, JK. Chronic insomnia, quality-of-life, and utility scores: comparison with good sleepers in a cross-sectional international survey. Sleep Med; 2012; 13, 1 pp. 43-51. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22093806][DOI: https://dx.doi.org/10.1016/j.sleep.2011.03.020]

7. Léger, D; Massuel, MA; Metlaine, A. SISYPHE study group. Professional Correlates of Insomnia Sleep; 2006; 29, 2 pp. 171-178. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/16494084]

8. Perlis ML, Ellis JG, Kloss JD, Dieter WR. Etiology and pathophysiology of insomnia. In: Principles and Practice of Sleep Medicine. 6th ed. Elsevier; 2016. p. 769–84.

9. Nicassio, PM; Mendlowitz, DR; Fussell, JJ; Petras, L. The phenomenology of the pre-sleep state: the development of the pre-sleep arousal scale. Behav Res Ther; 1985; 23, 3 pp. 263-271.1:STN:280:DyaL2M3isFalsw%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/4004706][DOI: https://dx.doi.org/10.1016/0005-7967(85)90004-X]

10. Vargas, I; Nguyen, AM; Muench, A; Bastien, CH; Ellis, JG; Perlis, ML. Acute and chronic insomnia: what has time and/or hyperarousal got to do with it?. Brain Sci; 2020; 10, 2 71. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32013124][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7071368][DOI: https://dx.doi.org/10.3390/brainsci10020071]

11. Gieselmann, A; de Jong-Meyer, R; Pietrowsky, R. Kognitive und körperliche Erregung in der Phase vor dem Einschlafen. Z Für Klin Psychol Psychother; 2012; 41, 2 pp. 73-80. [DOI: https://dx.doi.org/10.1026/1616-3443/a000134]

12. Jansson-Fröjmark, M; Norell-Clarke, A. Psychometric properties of the Pre-Sleep Arousal Scale in a large community sample. J Psychosom Res; 2012; 72, 2 pp. 103-110. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22281450][DOI: https://dx.doi.org/10.1016/j.jpsychores.2011.10.005]

13. Ruivo Marques, D; Allen Gomes, A; Nicassio, PM; de Azevedo, MHP. Pre-Sleep Arousal Scale (PSAS): psychometric study of a European Portuguese version. Sleep Med; 2018; 1, 43 pp. 60-65. [DOI: https://dx.doi.org/10.1016/j.sleep.2017.10.014]

14. Beck, AT; Epstein, N; Brown, G; Steer, RA. An inventory for measuring clinical anxiety: Psychometric properties. J Consult Clin Psychol; 1988; 56, 6 pp. 893-897.1:STN:280:DyaL1M%2FnvF2mtg%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/3204199][DOI: https://dx.doi.org/10.1037/0022-006X.56.6.893]

15. Ulusoy, M; Şahin, NH; Erkmen, H. Turkish version of the beck anxiety inventory: psychometric properties. J Cogn Psychother Int Q.; 1998; 12, 2 pp. 163-172.

16. Brinker, JK; Dozois, DJA. Ruminative thought style and depressed mood. J Clin Psychol; 2009; 65, 1 pp. 1-19. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19048597][DOI: https://dx.doi.org/10.1002/jclp.20542]

17. Karatepe HT. Ruminatif Düşünce Biçimi Ölçeğinin Türkçe uyarlaması, geçerlik ve güvenirlik çalışması. [İstanbul]: T.C. Sağlık Bakanlığı İstanbul Bakırköy Prof. Dr. Mazhar Osman Ruh Sağlığı ve Sinir Hastalıkları Eğitim ve Araştırma Hastanesi; 2010.

18. Bastien, CH; Vallières, A; Morin, CM. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med; 2001; 2, 4 pp. 297-307. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11438246][DOI: https://dx.doi.org/10.1016/S1389-9457(00)00065-4]

19. Boysan, M; Güleç, M; Beşiroğlu, L; Kalafat, T. Psychometric properties of the Insomnia Severity Index in Turkish sample. Anadolu Psikiyatri Derg; 2010; 11, pp. 248-252.

20. Lovibond, PF; Lovibond, SH. The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther; 1995; 33, 3 pp. 335-343.1:STN:280:DyaK2M3ktl2jsA%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/7726811][DOI: https://dx.doi.org/10.1016/0005-7967(94)00075-U]

21. Yıldırım, A; Boysan, M; Kefeli, MC. Psychometric properties of the Turkish version of the depression anxiety stress scale-21 (DASS-21). Br J Guid Couns; 2018; 46, 5 pp. 582-595. [DOI: https://dx.doi.org/10.1080/03069885.2018.1442558]

22. García-Cueto, E; Rodríguez-Díaz, FJ; Bringas-Molleda, C; López-Cepero, J; Paíno-Quesada, S; Rodríguez-Franco, L. Development of the Gender Role Attitudes Scale (GRAS) amongst young Spanish people. Int J Clin Health Psychol; 2015; 15, 1 pp. 61-68. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30487822][DOI: https://dx.doi.org/10.1016/j.ijchp.2014.10.004]

23. Bakioğlu, F; Türküm, AS. Toplumsal Cinsiyet Rolleri Ölçeği’nin Türkçe’ye Uyarlaması: Geçerlik ve Güvenirlik Çalışması. Kastamonu Eğitim Derg; 2019; 15, pp. 701-716.

24. International Test Commission. ITC Guidelines for Translating and Adapting Tests (Second Edition) [Internet]. International Test Commission; 2017. https://www.intestcom.org/files/guideline_test_adaptation_2ed.pdf

25. Kyriazos, TA. Applied Psychometrics: Sample Size and Sample Power Considerations in Factor Analysis (EFA, CFA) and SEM in General. Psychology; 2018; 09, 08 pp. 2207-2230. [DOI: https://dx.doi.org/10.4236/psych.2018.98126]

26. Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis. CengageLearning; 2019.

27. Hu, L; Bentler, PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Model Multidiscip J; 1999; 6, 1 pp. 1-55. [DOI: https://dx.doi.org/10.1080/10705519909540118]

28. Schumacker, RE; Lomax, RG. A beginner’s guide to structural equation modeling; 2016; 4 New York, NY, Routledge: 351.

29. Hughes DJ. Psychometric Validity. In: The Wiley Handbook of Psychometric Testing [Internet]. John Wiley & Sons, Ltd; 2018 [cited 2023 Apr 24]. p. 751–79. Available from: https://onlinelibrary.wiley.com/doi/abs/https://doi.org/10.1002/9781118489772.ch24

30. Hoekstra, R; Vugteveen, J; Warrens, MJ; Kruyen, PM. An empirical analysis of alleged misunderstandings of coefficient alpha. Int J Soc Res Methodol; 2019; 22, 4 pp. 351-364. [DOI: https://dx.doi.org/10.1080/13645579.2018.1547523]

31. Bujang, MA; Baharum, N. A simplified guide to determination of sample size requirements for estimating the value of intraclass correlation coefficient: a review. Arch Orofac Sci; 2017; 12, 1 12.

32. Matheson GJ. We need to talk about reliability: making better use of test-retest studies for study design and interpretation. PeerJ [Internet]. 2019 May 24 [cited 2021 Mar 27];7. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6536112/

33. Okajima, I; Ishii, M; Ochi, M; Nicassio, PM. Development and validity of the japanese version of the pre-sleep arousal scale. Tohoku J Exp Med; 2020; 252, 2 pp. 169-176. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33055430][DOI: https://dx.doi.org/10.1620/tjem.252.169]

34. Namra, S; Tazvin, I. Reliability and validity of pre-sleep arousal scale for Pakistani University Students—ProQuest. FWU J Soc Sci; 2014; 8, 1 pp. 78-82.

35. Rönkkö, M; Cho, E. An updated guideline for assessing discriminant validity. Organ Res Methods; 2022; 25, 1 pp. 6-14. [DOI: https://dx.doi.org/10.1177/1094428120968614]

36. Puzino, K; Amatrudo, G; Sullivan, A; Vgontzas, AN; Fernandez-Mendoza, J. Clinical significance and cut-off scores for the pre-sleep arousal scale in chronic insomnia disorder: a replication in a clinical sample. Behav Sleep Med; 2019; 18, 6 pp. 705-718. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31545084][DOI: https://dx.doi.org/10.1080/15402002.2019.1669604]

37. Jansson, M; Linton, SJ. The role of anxiety and depression in the development of insomnia: cross-sectional and prospective analyses. Psychol Health; 2006; 21, 3 pp. 383-397. [DOI: https://dx.doi.org/10.1080/14768320500129015]

38. Sivertsen, B; Salo, P; Mykletun, A; Hysing, M; Pallesen, S; Krokstad, S et al. The bidirectional association between depression and insomnia: the HUNT study. Psychosom Med; 2012; 74, 7 pp. 758-765. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22879427][DOI: https://dx.doi.org/10.1097/PSY.0b013e3182648619]

39. Hantsoo, L; Khou, CS; White, CN; Ong, JC. Gender and cognitive–emotional factors as predictors of pre-sleep arousal and trait hyperarousal in insomnia. J Psychosom Res; 2013; 74, 4 pp. 283-289. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23497828][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3655522][DOI: https://dx.doi.org/10.1016/j.jpsychores.2013.01.014]

40. Dietch JR, Manber R. Insomnia and cognitive arousal are important potential targets to reduce perinatal depression risk. Sleep. 2021 Jun 1;44(6):zsab091.

41. Kalmbach DA, Cheng P, Drake CL. A pathogenic cycle between insomnia and cognitive arousal fuels perinatal depression: exploring the roles of nocturnal cognitive arousal and perinatal-focused rumination. Sleep. 2021 Apr 8;44(6):zsab028.

42. Bootzin, RR; Epstein, DR. Understanding and Treating Insomnia. Annu Rev Clin Psychol; 2011; 7, 1 pp. 435-458. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/17716026][DOI: https://dx.doi.org/10.1146/annurev.clinpsy.3.022806.091516]

43. Rosen, RC; Lewin, DS; Goldberg, L; Woolfolk, RL. Psychophysiological insomnia: combined effects of pharmacotherapy and relaxation-based treatments. Sleep Med; 2000; 1, 4 pp. 279-288.1:STN:280:DC%2BC2sbjtlGktA%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/11040460][DOI: https://dx.doi.org/10.1016/S1389-9457(00)00042-3]

44. Neubauer, DN; Pandi-Perumal, SR; Spence, DW; Buttoo, K; Monti, JM. Pharmacotherapy of Insomnia. J Cent Nerv Syst Dis; 2018; 19, 10 1179573518770672.

45. Bonnet, MH; Arand, DL. Hyperarousal and insomnia: State of the science. Sleep Med Rev; 2010; 14, 1 pp. 9-15. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/19640748][DOI: https://dx.doi.org/10.1016/j.smrv.2009.05.002]

46. Koo, TK; Li, MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med; 2016; 15, 2 pp. 155-163. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/27330520][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4913118][DOI: https://dx.doi.org/10.1016/j.jcm.2016.02.012]

47. Hermida, R. The problem of allowing correlated errors in structural equation modeling: concerns and considerations. Comput Methods Soc Sci.; 2015; 3, 1 pp. 5-17.

48. Bollen, K; Lennox, R. Conventional wisdom on measurement: a structural equation perspective. Psychol Bull; 1991; 110, 2 305. [DOI: https://dx.doi.org/10.1037/0033-2909.110.2.305]

49. Tinajero, R; Williams, PG; Cribbet, MR; Rau, HK; Silver, MA; Bride, DL et al. Reported history of childhood trauma and stress-related vulnerability: Associations with emotion regulation, executive functioning, daily hassles and pre-sleep arousal. Stress Health; 2020; 36, 4 pp. 405-418. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32073201][DOI: https://dx.doi.org/10.1002/smi.2938]

50. Huberty, J; Puzia, ME; Green, J; Vlisides-Henry, RD; Larkey, L; Irwin, MR et al. A mindfulness meditation mobile app improves depression and anxiety in adults with sleep disturbance: analysis from a randomized controlled trial. Gen Hosp Psychiatry; 2021; 73, pp. 30-37. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34537477][DOI: https://dx.doi.org/10.1016/j.genhosppsych.2021.09.004]