Introduction

Although sleep is a fundamental physiological necessity for humans, as it is for all other living organisms (^{Assefa et al., 2015}), our understanding of sleep's functions remains limited. Present literature focuses on the two most universally acknowledged functions of sleep: recovery of the body and brain, and conservation of energy. Additionally, sleep deprivation, particularly when prolonged, leads to significant physiological complications (^{Krueger et al., 2016}; ^{Yetkin, 2021}).

Various prevalence rates for insomnia and related sleep problems have been reported (^{Siegel, 2022}). In the United States, one-third of the population reports sleep problems with about 9% experiencing daily sleep deprivation (^{Ancoli-Israel & Roth, 1999}). Similarly, in Canada, 19% are dissatisfied with their sleep, and 13% meet all criteria for insomnia (^{Morin et al., 2011}). Prevalence rates in European countries vary, with 19% to 29% in France (^{Leger et al., 2000}) and 11.7% in Norway (^{Pallesen et al., 2001}). In South Korea, 17% reported experiencing insomnia symptoms at least three nights per week (^{Ohayon & Hong, 2002}). China's insomnia prevalence is 15% (^{Cao et al., 2017}), with a noticeable increase during the COVID-19 pandemic, reaching up to 25% in some areas (^{Huang et al., 2020}; ^{Zitting et al., 2021}). Post-pandemic, approximately one in five people suffer from insomnia, highlighting the urgency for standard treatment methods (^{Pudlo et al., 2022}). Previous studies have found insomnia prevalence rates ranging from 3.2% to 42% in various populations (^{Liljenberg et al., 1988}; ^{Kokou-Kpolou, et al., 2020}). Notably, among university students, an emerging adult group, the average prevalence is 18.5%, significantly higher than the general population (^{Jiang et al., 2015}). These remarkable prevalence rates, especially in emerging adults, indicate that insomnia is a common health problem among university students and that more attention should be paid to insomnia in university students (Jiang et al., 2015). The findings on insomnia rates in Türkiye are both remarkable and varied, mirroring trends observed in other countries (^{Şahin et al., 2020}; ^{Şahiner-Önal & Hisar, 2018}). It's crucial to highlight that globally, insomnia is a growing concern, affecting up to 51% of people, as indicated by ^{Benbir et al. (2015}). This increasing prevalence underscores the importance of sleep as a fundamental human need and elevates insomnia to a critical public health issue. Insomnia's impact on physical and mental health demands significant research and attention (^{Korkmaz-Arslan, et al., 2020}). Furthermore, while the clinical population faces serious challenges with insomnia and related sleep disorders, varying rates are also observed in the nonclinical population. This suggests that addressing insomnia should be a widespread health priority, encompassing a broad range of populations to effectively mitigate its impact.

The DSM-5 dedicates a significant section to sleep-related disorders, defining insomnia as the difficulty in initiating or maintaining sleep, even in a conducive sleep environment (^{American Psychiatric Association (APA), 2022}). Similarly, the ICSD criteria describe insomnia as problems in initiating or maintaining sleep despite adequate opportunities and conditions, emphasizing the daytime consequences of these sleep disturbances (^{Sateia, 2014}). Individuals with insomnia experience a range of physical and psychological challenges (^{Yetkin & Aydın, 2014}). Notably, insomnia impairs cognitive and physical functioning and is linked to diminished daytime functioning across emotional, social, and physical domains (^{Roth, 2007}). It's also associated with decreased well-being, independent of physical illness and psychiatric disorders (^{Hamilton et al., 2007}). Furthermore, insomnia correlates with various psychological issues such as emotion dysregulation (^{Palagini et al., 2017}), depression (^{Lustberg & Reynolds III, 2000}), addiction (^{Liu et al., 2022}), and even violent tendencies (^{Evli et al., 2023}). These associations suggest a significant link between insomnia and a range of psychological problems (^{Sahebi et al., 2021}). Overall, this comprehensive understanding of insomnia, as defined by DSM-5 and ICSD, along with its extensive impact on both physical and psychological health, underscores the necessity of addressing this pervasive condition.

In recent years, there has been increasing interest in research on cognitive models of insomnia (^{Espie, 2007}). The cognitive model of insomnia focuses on confirming that people with insomnia have a strong and intense concern about the negative consequences of insomnia and the direct impact of insomnia on daily life. In particular, the effects of insomnia are thought to cause a serious deficit in the perception of performance loss and, according to the cognitive model of insomnia, the result is the manifestation of psychological distress (^{Harvey, 2002}). In addition, ways of thinking about sleep, in other words beliefs about sleep, are reported to exacerbate the problems associated with insomnia (^{Castillo et al., 2023}; ^{Morin et al., 2002}). When research on cognitive models of insomnia is examined, processes related to the day and night effects of sleep come to the fore. When individuals associate the effects of insomnia with daytime processes, dysfunctional beliefs come into play (Harvey, 2002). In addition, recent studies show that cognitive behavioural therapies are highly effective in addressing dysfunctional beliefs about insomnia, following the cognitive model of insomnia (^{Eidelman et al., 2016}; ^{Thakral et al., 2020}).

In addition to the growing body of literature on the cognitive model of insomnia, the theoretical framework of REBT provides a solid framework for insomnia. According to REBT, psychological disturbances are fed by 4 core irrational beliefs (see Figure 1). These are demandingness, frustration intolerance, awfulizing, and global evaluation of worth (^{Dryden, 2013}). The common characteristic of irrational beliefs is that they are usually absolute, rigid, and inflexible (^{DiGiuseppe et al., 2013}), and they generally do not contribute to the achievement of individual goals and cause unhealthy emotions and psychological disturbance (^{Turner, 2016}). Specifically, demandingness has a unique place for Albert ^{Ellis. Demandingness is the basis of other psychological disorders (Ellis, 1997}). Individuals' wishes and desires become absolute and obligatory over time (^{Ellis, 1995}), and demandingness transforms into irrational beliefs of "should, must, must have, have to, need to" (DiGiuseppe et al., 2013). Individuals tend to encounter negative or unpleasant events in their lives and avoid situations that require a reduction in their comfort zone. The tension that arises particularly in the face of unpleasant life events is defined as frustration intolerance (^{Ellis, 2003}). Individuals with irrational beliefs of frustration intolerance in their lives often experience ongoing psychological discomfort as they attempt to maintain their comfort zone or avoid unpleasant events (Dryden, 2011). Awfulizing is an individual's exaggerated evaluation of an issue and usually takes simple forms such as "terrible, bad, horrible". When individuals reflect on their awful beliefs, they usually act as if the worst-case scenario has occurred (DiGiuseppe et al., 2013). The value of the human being is suspended in the REBT framework, and it is recognized that individuals are changeable, capable of learning and making mistakes. Therefore, it is accepted that a holistic assessment of the human being is much more complex than it appears, and the assignment of human value is avoided (^{Dryden, 1994}). According to this approach, it is only a person's behaviour that should be evaluated and therefore devaluation should be avoided (^{Dryden, 2019}).

Previous research confirms the existence of measurement instruments that can reveal the cognitive model of insomnia and satisfy the severe and increasing need (^{Adan et al., 2006}; ^{Morin et al., 2011}; ^{Espie et al., 2000}). The Dysfunctional Beliefs and Attitudes about Sleep (DBAS), a widely used instrument to measure the cognitive model of insomnia, includes 5 sub-dimensions. However, this instrument had some shortcomings (Espie et al., 2000). On the other hand, the Dysfunctional Beliefs and Attitudes about Sleep Brief Version (DBAS-16) reflects 4 dimensions including perceived consequences of insomnia, worry/helplessness about insomnia, sleep expectations and medication. In addition, the Sleep Beliefs Scale (SBS) is divided into 3 factors: sleep-incompatible behaviors, sleep-wake cycle behaviors, and thoughts and attitudes about sleep (Adan et al., 2006). The current literature provides some measures of the cognitive model of insomnia. The cognitive beliefs addressed in these instruments are similar to, but independent of, the REBT theoretical framework. Furthermore, when the structural features of the scales are considered, it is apparent that the scales have a conceptual structure that deals with the behavioural consequences of thoughts rather than thinking styles. This situation provides a framework that is disconnected from the theory in terms of identifying the effectiveness of therapy or the cognitive thinking styles that need to be intervened on, especially in treatment-oriented approaches. From this point of view, we believe that a scale that directly addresses thoughts about insomnia according to the theoretical background would be effective. However, none of these provide a comprehensive measure of beliefs about insomnia based on cognitive behavioural theories such as Ellis' theory.

Recent literature in cognitive-behavioral theories offers various conceptualizations and measurement instruments related to catastrophizing (^{Clemente et al., 2023}; ^{Jansson-Fröjmark et al., 2020}; ^{Morin et al., 2007}; ^{Tan et al., 2017}). In this domain, measures of insomnia catastrophizing are continually evolving, with new instruments being developed (Jansson-Fröjmark et al., 2012) and existing ones adapted into different languages (^{Ballesio et al., 2018}; ^{Kabadayi et al., 2021}). While catastrophizing is a significant concept in Rational Emotive Behavior Therapy (REBT), it does not encompass all irrational beliefs. Our study distinguishes itself by exploring the psychometric properties of an insomnia measure grounded in the well-established principles of REBT. We use the theoretical framework of REBT, which has been shown to be effective in treating insomnia (^{Qin et al, 2023}; ^{Sakakini et al., 2020}). Specifically, we utilize Ellis' theory to develop a comprehensive assessment tool for insomnia-related problems, following the cognitive model of insomnia. This instrument, the Insomnia Belief Scale (IBSI), aims to thoroughly measure various aspects of insomnia-related sleep patterns, including difficulties with falling asleep, maintaining sleep, and the adverse daytime consequences of poor sleep. It does so through the dimensions of demandingness, frustration intolerance, awfulizing, and global development of value. This instrument can be used to more effectively measure the diagnostic and treatment outcomes of interventions in mental health studies based on the theoretical principles of REBT. The primary objective of our study is to establish the validity and reliability of the IBSI, thereby contributing a novel and comprehensive instrument for assessing insomnia within the framework of REBT.

Figure 1. Theoretical Framework of Psychological Disturbance. 

Methods

Results

Exploratory Factor Analysis

To establish the IBSI factor structure, 420 participants completed the questionnaire. After preliminary analysis of the data set, 29 data were excluded due to outliers. The remaining 391 data were analyzed using EFA. EFA was performed using a 4-factor maximum likelihood analysis based on the correlation matrix (model fit indices; (Χ²/df) = 2.71, CFI = .94, SRMR = .03 and RMSEA = .066). The Oblimin ratio was preferred because the factors were assumed to be related. The KMO value was determined to be .91 and the Bartlett test value to be 5636.458/276 (p = .000). Based on the EFA results, evaluations for each item in the scale were based on factor loadings, and the scale structure was thoroughly examined. Items were removed if they had high loadings on multiple factors, a factor loading below .40, or if the difference between loadings on two factors was less than .15. The main criteria for item removal were significant loading magnitude and cross-loadings, with theoretical alignment also being critical. Items with the highest inter-item correlations were examined to ensure that the scale includes distinct items, preventing artificially inflated reliability. Additionally, each factor was required to have at least three items throughout the deletion process (^{Carpenter, 2018}; ^{Worthington & Whittaker, 2006}). The EFA results indicated that the best 6 items for each dimension should be included in the scale with satisfactory factor loading (see Table 1). The uniqueness values of the items should be further away from 1 and close to zero or lower. The instrument was found to have all items with acceptable uniqueness scores.

Table 1. Item Factor Loadings and Uniqueness Scores. 

^Note.The table only shows item factor loadings of .32 and above.

As shown in Table 1, the factor loads for each dimension are .55 and above. The results show that the item factor loads for each dimension are at the least good level. In addition, the explained variance rate has a significant ratio of 65.10% for the total IBSI. Furthermore, the uniqueness values are close to 0.00 and far from 1.00. The items of the expressed uniqueness scores meet the criteria. Scree plot evidence is also presented to examine 4 dimensions in EFA (see Figure 2). The eigen scores of each dimension also show consistent results. The eigen scores for the 4 dimensions of the instrument are 8.58, 3.24, 2.24 and 1.56. Even the eigen score for the 5th possible dimension is below 1.00 (Nex possible dimension 5 for = .88).

Figure 2. Scree plot of IBSI. 

Network analysis was used to provide additional evidence supporting the factor structure of the IBSI (see Figure 3). In the network analysis, yellow colors symbolize the dimension of demandingness, red colors symbolize awfulizing, green colors symbolize frustration intolerance, and blue colors symbolize the global evolution of worth. First, it is noteworthy that the items related to each dimension are located on the network plane by a factor. Secondly, there are significant positive relationships between the items within the dimensions. Moreover, as expected, the theoretical relevance of irrational beliefs is the apparent relationship between items within dimensions.

Figure 3. Network Analysis of IBSI. 

Confirmatory Factor Analysis

CFA with maximum likelihood estimation was used for the four-dimensional IBSI. (See Appendix 4). As shown, the factor model (Χ²/df) = 1.83, CFI = .95, TLI = .95, GFI = .97, SRMR = .06, and RMSEA = .05. In summary, the fit indices obtained because of CFA were found to be at least acceptable.

Reliability

Cronbach’s α and McDonald’s ω were calculated in EFA and CFA. Reliability scores obtained from IBSI total and sub-dimensions in the EFA and CFA processes were satisfactory (see Table 2).

Table 2. Reliability results of the IBSI. 

Test-Criterion Validity

For test-criterion validity, we analyzed the correlation between IBSI and ISI, depression, anxiety, and stress (see Table 3). The correlation between IBSI, ISI, depression, anxiety, and stress was analyzed, and the results showed positive correlations between many variables as expected.

Table 3. Pearson correlation coefficients among variables. 

^{Note. *}p <.01,

^**p <.001

We also tested network analysis with variables (see Figure 4). We analyzed data with the correlation estimator using JASP. Each node is labelled with variable and only significant relationships are shown as edges. Blue colors indicate, as expected, significant positive correlations between variables. The red color confirms the significant negative relationship between insomnia severity and sleep time. The thicker edges between anxiety, depression and stress reflect the strength of the relationship. Furthermore, the IBSI confirms significant positive correlations with depression, stress, anxiety, and insomnia severity. In summary, the findings of the network analysis strongly support the correlation coefficients.

Figure 4. Network Analysis of Variables. 

Intergroup Differences

For differences in IBSI total scores and between groups, the differences between the clinical group who had insomnia symptoms, and the nonclinical group are shown in detail by independent samples t-test (see Table 4).

Table 4. Differences Between Clinic and Nonclinical Groups. 

^Note.Student’s t-test,

^*p <.05

Intergroup differences showed that there was a significant difference between the clinic group with insomnia symptoms and the non-clinic group without insomnia symptoms on IBSI (t = 5.62, p < .05, Cohen’s d = .750), awfulizing (t = 5.24, p < .05, Cohen’s d =.698), frustration intolerance (t = 5.21, p < .05, Cohen’s d = .695) and global evolution of worth (t = 4.74, p < .05, Cohen’s d = .632). However, there was no significant difference in the demandingness dimension between the clinic group with insomnia symptoms and the non-clinic group considered healthy without insomnia symptoms (t =.13, p > .05, Cohen’s d = .017).

Discussion

The present study focused on the validity and reliability of the IBSI in Turkish emerging adults. The IBSI mainly measures irrational beliefs about insomnia, focusing on irrational beliefs, the key role in the theory of rational emotive behavior. The scale development steps were as follows. Comprehensive literature review, item pooling, focus group interview, expert opinion on items, pilot study, main study for EFA, and then main study for CFA.

First, it was noteworthy that some of the items in the item pool were revised through the focus group interview. Items were of higher quality in a focus group interview with identified emerging adults with different characteristics of insomnia. The authors reached a common consensus for the draft IBSI. Second, item statistics and correlations between items from the pilot study yielded significant positive correlations as expected. Third, Bartlett's test ensured that there was a sufficient sample size in the main study for the EFA. The findings revealed a 4-dimensional (demandingness, awfulizing, frustration intolerance, global evolution of worth) and 24-item structure for the IBSI in the EFA. Specifically, item factor loadings ranged from .59-.75 for demandingness, .68-.90 for awfulizing, .59-.83 for frustration intolerance, and .65-.90 for global evolution of worth. The factor loadings of these items have the lowest "good" level score according to the criteria proposed by ^{Comrey and Lee (1992}). The uniqueness scores of these items were far from 1.00 and close to 0.00. In addition, the Scree Plot and network analysis of the factor structure also provided additional evidence of whether the IBSI theoretically satisfies the 4-dimensions. Scree Plot showed that the breakdown occurs after 4 dimensions. Furthermore, the low-eigenvalue produced for the fifth dimension of IBSI is also evidence for the 4-dimensional structure of the instrument. According to the network analysis, it confirmed that the items are located together in the network plane as expected and that the items show positive correlations. Fourth, CFA was performed to confirm the factor structure of the IBSI. The results showed that the fit indices satisfy at least the acceptable level or higher.

Reliability is a point that cannot be ignored for measurement instruments. In the present study, Cronbach's α and McDonald's ω scores were calculated for reliability. The findings of both EFA and CFA studies showed that the Cronbach's α and McDonald's ω scores of IBSI and its sub-dimensions were satisfactory.

Test-criteria validity has shown many significant correlations between the IBSI and insomnia severity, depression, anxiety, and stress as demandingness, awfulizing, frustration intolerance, and global evolution of worth. These correlational results should be considered as evidence of the theoretically expected significant relationships. However, it was noteworthy that while there were positive significant relationships between the demandingness dimension and awfulizing, frustration intolerance, global evolution of worth, and IBSI, there was no significant relationship between insomnia severity, depression, anxiety, and stress. While interpreting these findings, the theoretical focus should be on the difference between demandingness and desires.

From the perspective of intergroup differences, there is an important aspect of the demandingness dimension that should be considered theoretically. According to the theory, it is not as straightforward as it seems to distinguish between demandingness and desire. Ellis distinguishes between demands and preferences. In terms of preferences, wants and desires cannot be evaluated as rational or irrational. Consequently, individuals experience psychological disorders only when their wants and desires turn into excessive demands (^{DiGiuseppe et al., 2013}). While these discourses are more pronounced in English in terms of grammatical structure, Turkish is different in terms of language structure. Instead of obligations, wishes and preferences increase the likelihood of similar meaning. As an example, instead of "I would like ..." can mean "I should ..." and "I must ..." in the cognitive structure of the Turkish language. Therefore, it should not be considered a surprising result that there was no significant difference in the demandingness dimension between the clinic group with insomnia symptoms and the non-clinic group considered without insomnia symptoms. For the current situation, more in-depth qualitative studies are needed in Turkish emerging adulthood groups.

Conclusion

The present scale development study for the IBSI has been carried out for validity and reliability. We share the instrument to inspire researchers (see Appendix 5). The demandingness dimension should be considered carefully in the present study. The correlation results showed that there was a positive medium correlation with the IBSI for demandingness. However, demandingness did not show a significant relationship with depression, anxiety, and stress. ^{Ellis highlights the risk that wishes and desires become demands over time (Ellis et al., 2010}). Moreover, some people become more demanding in their living conditions but do not develop serious mental disorders. However, there are still various negative consequences of demandingness in different forms (self-deception, withdrawal, alcohol, and drug abuse, etc.) (Ellis, 1997). Consequently, the responses of the nonclinical population may essentially involve more idealized wishes and desires for insomnia. At the same time, the nonclinical population may have resorted to alternative unhealthy coping strategies.

The present study provides an opportunity for following studies on the demandingness dimension. Future studies would benefit from exploring rational and flexible beliefs (the nature of wants and desires) about insomnia in clinical and nonclinical populations. An important limitation of this study was the use of focus group interviews with clinical cases. Future studies should further explore and differentiate demandingness in clinical and nonclinical populations, particularly in line with the cognitive model of insomnia. At the same time, understanding the mechanism by which high demandingness leads to dysfunctional outcomes in nonclinical populations would make a particular contribution.

Frustration intolerance is characterized by a response to threats and tensions that eliminate some form of comfort (^{Ellis, 2003}). Avoidance of unpleasant life events is associated with frustration intolerance and consequently causes psychological distress (^{Dryden, 2011}). Everyone spends sleepless nights for a variety of reasons, but the discomfort that can result from an irrational desire for comfort and sleep is supported by the belief that insomnia is only frustration intolerance. In this perspective, it is important to identify frustration intolerance beliefs about insomnia. Mental health professionals can help the clients by identifying the frustration intolerance beliefs of their clients during unpleasant and unwanted nights.

Global evolution of worth is a specific dimension of the cognitive model of sleep. Global evolution of worth focuses on self-acceptance and accepts the human capacity to make mistakes and errors (^{DiGiuseppe et al., 2013}). Devaluation is particularly closely related to low self-esteem (^{Chamberlain & Haaga, 2001}; ^{Thompson & Waltz, 2008}). Therefore, the global evolution of the worth dimension may be a predictor of potentially serious emotional and behavioral disorders such as depression, self-harm, suicide, and low self-esteem in clinical and nonclinical populations. Thus, the devaluation dimension of insomnia may have an important function in both clinical cases and research.

The awfulizing dimension is shorter than the instruments in the field and covers general attitudes. There are important conceptualizations of the cognitive model of insomnia in the literature (^{Jansson-Fröjmark et al., 2012}; ^{Tan et al., 2017}).

None of them include Ellis' model. The awfulizing dimension may have an important application in literature, as it is 6 items, short and practical.

The findings of the study provided significant evidence that the IBSI was valid and reliable. The IBSI, which consists of 6 items and 4 dimensions, has several potential benefits for the Turkish literature. It is a theoretically and potentially sound instrument for measuring increased insomnia and insomnia-related cognitive problems. As far as we know, it is the first REBT-based measurement instrument that is appropriate for the cognitive model of insomnia in Turkish literature. The measurement instrument based on cognitive behavioral and rational emotive behavior theory will be an important tool for those researching insomnia in Türkiye. As in many countries, cognitive behavioral therapy is the most common psychotherapy approach in Türkiye. The instrument can be used in both clinical and nonclinical groups of emerging adults aged 18-29 years. Moreover, mental health practitioners can identify irrational beliefs for insomnia in therapy in line with the cognitive model of insomnia. This could contribute to a more effective helping model. It should not be overlooked that the instrument potentially needs validity and reliability studies in different cultures, especially in the English language. Moreover, IBSI should be encouraged for adaptation studies in different languages and cultures.