European Psychiatry 20 (2005) 503–509 http://france.elsevier.com/direct/EURPSY/ Original article Reliability of the Spanish version of the Composite Scale of Morningness A. Adan a,*, H. Caci b, G. Prat a a Department of Psychiatry and Clinical Psychobiology, School of Psychology, University of Barcelona, Passeig Vall d’Hebrón 171, 08035 Barcelona, Spain b Service de pédiatrie, hôpital Archet-2, CHU de Nice, 151, route de Saint-Antoine-de-Ginestière, 06202 Nice cedex 3, France Received 19 September 2004; accepted 13 January 2005 Available online 25 March 2005 Abstract Aim. – The aim of this study was to examine the reliability of the Spanish version of Composite Scale of Morningness (CSM) and its ability to measure the circadian typology. Subjects and methods. – Voluntary and unpaid psychology students (N = 391; 132 men and 259 women), aged between 17 and 33, completed the questionnaire between the months of September and December. Results. – The total score was independent of age and gender, with a close to normal distribution and a non-significant negative skewness. The internal consistency was high (Cronbach’s ␣ = 0.87) and factor analysis extracted three factors labeled Time of Retiring (items 2 and 7), Activity Planning (items 8, 9, and 13) and Morning Affect (items 3–6, and 10–12). With the 10th and 90th percentiles as cut-off scores, scorers below 22 (N = 40; 10.2%) are classified as evening-types and scorers above 39 as morning-types (N = 28; 7.2%). Conclusion. – The Spanish questionnaire shares most of the good psychometric properties of other versions of the CSM, and thus can be used for Spanish-speaking student samples. Nevertheless, further studies of normative data in workers and aged subjects are needed in order to validate CSM. © 2005 Elsevier SAS. All rights reserved. Keywords: Chronobiology; Circadian rhythms; Chronotype; Morningness; Eveningness; Questionnaire; Gender differences 1. Introduction A pervasive characteristic of mammals is the existence of rhythmic changes that can be observed on several levels of organization from basic cellular phenomena to complex behaviors. The endogenous mechanisms, also called pacemakers, are adjusted (i.e. synchronized) to environmental cycles, chiefly the light–dark cycle [16,23,34]. The morningness–eveningness dimension is a continuum on which individuals can be arranged from the morning-type or “lark” to the evening-type or “owl”, with a larger intermediate group. It is associated with interindividual differences in rhythmic expression, such as academic, professional and sport performance [12,23,41], and is also related to personality traits and to psychopathologic risk factors at adolescence and adulthood [10–12,17,23]. Circadian typology is assessed by means of self-rating questionnaires; the first constructed and the most widely used * Corresponding author. E-mail address: aadan@ub.edu (A. Adan). 0924-9338/$ - see front matter © 2005 Elsevier SAS. All rights reserved. doi:10.1016/j.eurpsy.2005.01.003 is the Morningness–Eveningness Questionnaire (MEQ) [22]. Some of the criticisms addressed to the MEQ are that the total score may not be appropriate to measure a multidimensional construct [31], and that small subsets of items may convey most of the total variance of the measure [1,15]. Two other questionnaires have been proposed about which similar criticisms have been raised: the Circadian Type Questionnaire (CTQ) [21], the Diurnal Type Scale (DTS) [19,39]. With this in mind Smith et al. [36] developed a new questionnaire they called the Composite Scale of Morningness (CSM) which is composed by the ‘best’ items of the MEQ (nine items) and the CTQ (four items). Several studies have demonstrated the good CSM test–retest reliability [9,20], and an adequate external validity when considering different variables. The total score ranges from 13 (extreme evening-type) to 55 (extreme morning-type). The authors arbitrarily proposed to retain the 10th and 90th percentiles as cut-off scores to classify the subjects into one of the above three categories, which correspond to the scores of 22 and 44, respectively, in their sample. On the other hand, Alzani and Natale [4] proposed a fixed 504 A. Adan et al. / European Psychiatry 20 (2005) 503–509 cut-off approach dividing the theoretical range of the total score (i.e. 55 – 13 = 42) in three equal parts: 13 + (55– 13)/3 = 27 and 27 + (55 – 13)/3 = 41. Unfortunately, international research has established that the centiles do vary across samples [35], e.g. 27 and 44, respectively, in Australia [20], or 27 and 41, respectively, in Italy [30], or 30 and 45, respectively, in France [9]. Consequently the blind conversion of the total score into a typology using whatever cut-off scores implies that one ignores this critical point. One solution may be to establish normalized cut-off scores with regard to nationality, age and gender before using the questionnaire in smaller groups or individually. Many studies have pointed out that age and gender both influence morningness. In adults the trend toward morningness increases with age [13,18,38,43], and women show a greater trend toward morningness than men in their rhythm expression [3,32]. But the relation between gender and morningness scores remains controversial: some researchers did not obtain significant differences between men and women [2,8,15,20] while others did using larger samples [3]. Moreover, like what is reported for some other personality traits, the relationship between age and morningness may not be linear across the age span [3,14,29]. It could be of interest to [25,31,42] analyze more accurately the potential differences between men and women in morningness–eveningness scores on the CSM. In the last decade, the use of CSM has progressively increased as the questionnaire was translated into several languages [4,8,30,33]. Regarding the structure of CSM, a threefactor solution was found in three independent Englishspeaking samples. The factors were called Morningness/ Effort, Evening, and Morning Affect, this latter being the more robust factor [6,36]. The Thaï version also obtained three components with some differences from the English version [33]. Although the authors of the Italian version retained a onefactor solution in both a student and a shiftworker samples, they also provided a plot of the eigenvalues suggesting the appropriateness of a three-factor solution [4]. In France, a one-factor solution was provisionally accepted [8] but a threefactor solution emerged later [9,10]. A transcultural work with large samples (Australia, France, Italy, Spain, and Thailand) also obtained a three-factor solution in all five countries [7]. The aim of this study is to examine the reliability of the Spanish version of the CSM in student subjects, including the analysis of the factor structure, and to take into account the potential gender differences in the structure and the raw scores. The typology will be determined following previous works [30,36] to establish the optimal categorization system. 2. Materials and methods 2.1. Subjects Participants were 391 undergraduate students (132 men and 259 women), aged between 17 and 33 (19.94 ± 2.22 years). All were voluntary, anonymous and unpaid. As expected in student samples, the Shapiro–Wilk tests indicate that age is positively albeit slightly skewed in total sample (W = 0.883, P < 0.001) and for both genders (men: W = 0.870, P < 0.001; women: W = 0.892, P < 0.001). In fact, 65.7% of the subjects were younger than 21. Men were about 9 months older than women (Z = 2.71, P < 0.007). 2.2. Measures and procedure The CSM was translated into Spanish and then back translated into English by two bilingual English speakers to ensure translation quality (see Appendix 1 for the Spanish version). Following we briefly described each one of the 13 items included in the CSM: 1 (At what time would you get up?), 2 (At what time would you go to bed?), 3 (How easy do you find getting up in the morning?), 4 (How alert do you feel during the first half hour after awakening?), 5 (How tired do you feel during the first half hour after awakening?), 6 (Physical exercise at early morning), 7 (At what time in the evening do you feel tired and in need of sleep?), 8 (Hour of peak for mental performance), 9 (Self-assessment of circadian typology), 10 (When would you prefer to rise?), 11 (If you were always to rise at 6:00 a.m. what were would you be like?), 12 (How long a time does it take before you “recover your senses” after rising?), and 13 (To what extent is an individual oriented toward morning or evening). Subjects completed the CSM in a morning session of a Psychology course at the University of Barcelona between the months of September and December. The CSM was administered to 400 students (a sample-size adequate to the aim of the study) ranged into five class-group. Before the administration of the CSM, students received a brief introduction about the nature of research, ethical requirements for confidentiality and voluntary participation. No informed consent was required. The response rate was close to 100% of the subjects present in each class and we have information about all the items since we control missing values. 2.3. Statistical analysis Scores are expressed as mean, standard deviation (S.D.) and range, and their distribution shapes were assessed for normality with the Shapiro–Wilk test. The reliability of the scores was estimated by the unstandardized unit-weighted Cronbach’s a coefficient. The CSM total score was submitted to an ANCOVA with gender as factor and age as covariate. Gender differences on item scores were tested by the two-sample Wilcoxon rank-sum test since the item responses are on an ordinal scale. We explored the structure underlying the 13 items of the CSM by exploratory factor analysis. Because the item scores are not normal by nature, we used the Weighted Least Squares estimator with Mean- and Variance-adjustment (WLSMV) [26]. The number of factors to extract was determined by Cattell’s scree test and an oblique PROMAX rotation was applied. A. Adan et al. / European Psychiatry 20 (2005) 503–509 The solution was considered as satisfactory if the Standardized Root Mean Residual (SRMR) index was lower than 0.050. The similarity of each factor between genders was assessed by the Tucker’s congruence coefficient, which can be used with oblique solutions. Previous work on factor similarity with large real data samples has showed that coefficients greater than 0.800 indicate good congruence between factors [5]. Analyses were run on Stata 7 [37] and Mplus 2.02 [27], and statistical tests were two-way with a type-I error set at 5%. 3. Results 3.1. Descriptive statistics and reliability The mean score for CSM was 30.74 (S.D. = 6.54; range: 14–48) in total sample, 30.23 (S.D. = 6.74; range: 16–48) in men and 31.00 (S.D. = 6.43; range: 14–46) in women. The CSM score had a marginally significant Shapiro–Wilk test in the total sample (W = 0.993; P = 0.047), while it is normally distributed in both genders as indicated by non-significant (ns) Shapiro–Wilk tests (men: W = 0.994, P = 0.256; women: W = 0.985, P = 0.165). The distributions shapes are similar between genders (Kolmogorov–Smirnov D = 0.084; ns), and not significantly skewed to morningness were never obtained (total sample: skewness = –0.202, P = 0.100; men: skewness = –0.096, P = 0.636; women: skewness = –0.253; P > 0.09). There was no effect on the CSM total score of gender (F(1,387) = 1.26; ns) and age (F(1,387) = 1.63; ns) nor gender x age interaction (F(1,387) = 0.86; ns). Spearman rankorder coefficient between age and CSM score did not show any significant correlation when considering all the subjects (q = 0.049), only men (q = –0.008) or women (q = 0.088). The reliability of the entire scale is excellent in total sample (␣ = 0.877) and similar between genders (men: ␣ = 0.889; women: ␣ = 0.985). The values of the 10th and 90th percentiles in our sample were, respectively, 21 and 39 in the total sample, 20 and 39 in men, and 22 and 39 in women. These values are close to those reported in the original study [36] but the resulting categorization would be slightly biased toward eveningness: 40 evening-types (10.2%; 18 men and 22 women), 323 neither-types (82.6%; 150 men and 218 women), and 28 morning-types (7.2%; nine men and 19 women). There is no difference between genders in the assignment to these three groups (v2(2) = 2.52; ns). Now considering the fixed cut-offs of 27 and 41, the resulting categorization shows a greater bias toward eveningness: 113 evening-types (28.9%; 39 men and 74 women), 261 neither-types (66.8%; 85 men and 176 women), and 17 morningtypes (4.35%; eight men and nine women). There is no difference between genders in the assignment to these three groups (v2(2) = 1.54; ns). Therefore, it would seem more adequate for Spanish student samples to maintain the percentile-based approach. However, in studies aimed at selecting subjects from the different categories less restrictive cut-off scores should be used, such as the percentiles 20th for the evening-type (cut-off = 25; total sample: 91 (23.3%); 35 men and 56 women) and 80th for the morning-type (cutoff = 37; total sample: 61 (15.6%); 19 men and 42 women). Then again there is no difference between genders in the assignment to these three groups (v(2) = 1.22; ns). Table 1 shows the mean scores of all the 13 items of the CSM with regard to gender (high scores indicating a trend toward morningness), along with the rank-order correlation between an item and the sum of the other twelve items. As in many previous studies in other languages, the lowest correlation appears with item 7 and the highest with item 9, which is a “validity check item” for the morningness–eveningness dimension. Women have higher scores than men on item 2 (Z = –3.02; P < 0.003), 7 (Z = –1.97; P = 0.049) and 9 (Z = – 2.83; P < 0.005), whereas men have a higher score on item 1 only (Z = 2.88; P = 0.004). 3.2. Factor analysis With respect to Cattell’s scree test, we extracted three factors from the 13 items of CSM in the total sample and in both Table 1 Mean score ± S.D. and item-corrected total rank-order correlation coefficients for all 13 items of the CSM in the total sample and according to gender Item 1 2 3 4 5 6 7 8 9 10 11 12 13 Total sample (N = 391) 2.39 ± 0.78 2.42 ± 0.75 2.36 ± 0.76 2.26 ± 0.73 2.49 ± 0.66 2.06 ± 0.72 2.37 ± 0.75 2.94 ± 0.95 2.25 ± 0.78 1.79 ± 0.67 2.15 ± 0.84 2.90 ± 0.85 2.37 ± 0.97 Items scores Men (N = 132) 2.55 ± 0.80 2.27 ± 0.77 2.34 ± 0.78 2.20 ± 0.73 2.55 ± 0.67 2.05 ± 0.75 2.26 ± 0.78 2.92 ± 0.88 2.08 ± 0.79 1.81 ± 0.71 2.08 ± 0.83 2.81 ± 0.82 2.32 ± 0.96 Women (N = 259) 2.31 ± 0.76 2.50 ± 0.73 2.38 ± 0.75 2.30 ± 0.74 2.45 ± 0.65 2.06 ± 0.71 2.42 ± 0.73 2.95 ± 0.98 2.34 ± 0.77 1.78 ± 0.65 2.18 ± 0.84 2.94 ± 0.87 2.40 ± 0.98 505 Total sample (N = 391) 0.571 0.482 0.630 0.534 0.534 0.491 0.415 0.499 0.750 0.531 0.537 0.546 0.649 Item-corrected total rank order correlation Women Men (N = 259) (N = 132) 0.540 0.622 0.562 0.446 0.602 0.647 0.555 0.518 0.554 0.533 0.552 0.460 0.465 0.385 0.571 0.460 0.780 0.738 0.501 0.552 0.498 0.554 0.538 0.550 0.693 0.623 506 A. Adan et al. / European Psychiatry 20 (2005) 503–509 Table 2 Eigenvalues of the correlation matrices Total sample (N = 391) 6.154 1.600 0.943 0.921 0.610 0.539 0.494 0.403 0.344 0.315 Men (N = 132) 6.479 1.611 0.969 0.872 0.710 0.611 0.544 0.319 0.258 0.246 The oblique solutions were assessed for similarity between genders using Tucker’s coefficient of congruence. All the coefficients for heterogeneous pairs are very low, ranging from –0.111 (F1–F3 pair) to + 0.473 (F1–F2 pair). The coefficients for homologous pairs were highest for the Morning Affect factor (F3–F3 = 0.937), and close to the threshold of 0.800 for the Time of Rising (F1–F1 = 0.795) and Activity Planning (F2–F2 = 0.778). Women (N = 259) 6.083 1.675 0.966 0.888 0.573 0.517 0.490 0.457 0.350 0.313 4. Discussion gender groups (Table 2). With regard to the SRMR, the oblique three-factor solutions are good in the total sample and in the women group (SRMR < 0.05) but less satisfactory in the men group (Table 3). We labeled the first factor Time of Retiring, composed by items 2 and 7. The second factor, named Activity Planning, is composed by items 8, 9, and 13. The third factor, labeled Morning Affect, counts seven items and previously appeared in several studies independently of the language. This latter factor is composed by items 3, 4, 5, 6, 10, 11, and 12. In women, item 11 was allocated on the Morning Affect factor, according to its primary loading of 0.493, the low secondary loadings and its salient loading in men. Although item 1 loads on the Time of Retiring factor in the men group only, we think it should not be discarded from the CSM as it reports relevant informations on sleep preference. Items 1 and 10 both refer to the time of rising in the morning. Thus if we accept that morningness is independent from the duration of sleep then it is not surprising that both these items load positively on the Time of Retiring factor (which may be renamed Sleep Planning). Anyway, we do not have any explanation for the low loadings of items 1 and 10 in women. Psychometric properties reported in the present work are similar to those reported in other countries and languages [4,8,33,36], suggesting the Spanish version of the questionnaire is effective and reliable. Our sample was larger than those used in standardizing studies realized in these other countries but this does not mean that it was representative of the Spanish student population. Nevertheless, our work is the first step to use this scale in circadian typology research in Spain and in Spanish-speaking countries. There is now a need to obtain normative data in a larger sample of subjects and to validate the questionnaire with biological and behavioral external parameters. The CSM total score was normally distributed in both gender groups with a marginally significant Shapiro–Wilk test in the entire sample, and a non-significant skewness in all cases. This result allows us to use the percentile-based method to categorize the subjects. With regard of the MEQ, it is admitted that 15–20% of the subjects are classified in each of the extreme categories (morning- and evening-type) [2,14,28, 29,38]. The fixed cut-off scores approach is biased toward eveningness and may imply a difficulty in research studies if CSM was used to categorize subjects into extreme typologies [17,41]. Spanish subjects are classified as evening-type (scores lower or equal to 21), as neither-type (scores between Table 3 Exploratory factor analysis (WLSMV estimator) of the CSM for the total sample and each gender group, with factor intercorrelations and Square-Root Mean Residual (SRMR). Loadings equal or greater than 0.500 in absolute value are in bold Item 1 Item 2 Item 3 Item 4 Item 5 Item 6 Item 7 Item 8 Item 9 Item 10 Item 11 Item 12 Item 13 F2 F3 SRMR 0.322 0.875 –0.146 –0.056 –0.152 0.113 0.939 –0.075 0.203 0.218 0.009 0.039 0.040 0.537 0.447 0.043 All sample (N = 391) 0.105 0.012 0.161 –0.100 0.037 –0.055 0.052 0.704 0.664 0.018 0.094 0.108 0.777 1.000 0.622 0.405 0.006 0.778 0.845 0.796 0.632 –0.154 0.055 0.164 0.513 0.606 0.580 0.070 — 1.000 0.655 0.750 –0.110 –0.125 –0.085 0.264 0.893 0.065 0.158 0.628 0.040 0.041 0.091 0.323 0.474 0.063 Men (N = 132) –0.013 0.400 0.170 0.056 0.069 0.018 0.375 0.532 0.540 –0.213 0.050 0.166 0.644 1.000 0.374 0.218 –0.104 0.763 0.855 0.767 0.502 –0.293 0.300 0.442 0.363 0.637 0.555 0.343 — 1.000 0.177 0.964 –0.124 0.074 –0.019 0.004 0.735 –0.122 0.172 –0.001 –0.066 0.088 0.038 0.488 0.294 0.039 Women (N = 259) 0.343 –0.043 0.213 –0.200 –0.033 0.075 0.209 0.712 0.746 0.276 0.284 0.073 0.773 1.000 0.619 0.371 0.071 0.736 0.853 0.796 0.580 –0.132 –0.005 0.101 0.453 0.493 0.607 0.018 — 1.000 A. Adan et al. / European Psychiatry 20 (2005) 503–509 22 and 39) and as morning-type (scores greater or equal to 40). Finally, research studies should apply the looser criteria (such as the percentiles 20 and 80) to select moderate eveningtype and morning-type, respectively. That is, scores lower or equal to 25 correspond to the evening-type and, scores equal or greater to 37 correspond to the morning-type subjects. It could be interesting to study other population groups, such as workers and especially those doing shift work. A previous work has obtained similar descriptive data and cut-off scores for CSM in shift workers [4]. In this study we only selected students because this group of subjects had fewer constraints to social synchronizers [24] (i.e. work schedule) and can express with greater liberty their rhythmic preference. This is the reason why students are more frequently included in circadian typology research. As a corollary, we were not able to explore the well known relationship between scores of morningness–eveningness and age since our age interval was smaller [13,18,23,38,40]. However, a clear morning orientation seems to appear by age 50 with associated sleep problems, sensitivity to jet-lag, and less adaptation to changes in hourly schedule and work shifts [32]. The analysis of factor structure of the CSM suggested to obliquely rotate a three-factor solution, namely: Time of Rising (items 2 and 7), Activity Planning (items 8, 9 and 13), and Morning Affect (items 3–6 and 10–12). This factor solution is similar to that obtained in previous studies [6,7,10] with the Morning Affect factor being to the most robust. The factor structure was also similar between genders, with a greater congruency for the Morning Affect factor and a close to satisfactory congruency for the other two factors. This factor structure has profound theoretical and practical implications that future studies should take into account in students and in other population groups. The factor Time of Rising is more influenced by environmental constraints and could be less endogenously determined. The Activity Planning factor involves mainly the self-perception of circadian typology. And, the Morning Affect factor could be a short CSM scale useful when a rapid and accurate measure of circadian endogenous organization is needed. In all analyses men and women were considered separately. The differences between genders appear mainly when we explore the factor structure of questionnaire, and descriptive data on the entire scale did not show any significant difference between genders. This latter result is not in line with previous studies that reported a gender effect on the morningness–eveningness scores using larger samples [3,14,28,29]. This difference is not very large, however, casting doubt on the possibility to reveal such an effect using questionnaires only whereas its genetic basis [40] and its relationship with the different functionality of the women endocrine system have been reported [28]. Obviously, our sample was large enough to establish the psychometric properties of the Spanish version of the CSM but lacked in power to detect a small gender effect. 507 5. Conclusion The CSM is a psychometrically sound instrument in its Spanish version also. The factor structure was constituted by three factors: Time of Retiring (items 2 and 7), Activity Planning (items 8, 9 and 13), and Morning Affect (items 3–6, and 10–12). Considering the more restrictive criteria of the cutoff (percentiles 10 and 90), evening-type subjects have a score of 21 or less, and morning-type subjects have a score of at least 40. We did not find any gender effect on item scores, total score and distribution shape of the total score. There were some discrepancies in the three-factor solution between genders, especially regarding the Time of Retiring factor. Future works should be carried out in order to obtain normative data and involving other population groups such as shift workers and subjects with a wider age range, as well as using external criteria, in order to validate the CSM in Spanishspeaking subjects. Appendix 1 Spanish version of the Composite Scale of Morningness (CSM) Escala compuesta de matutinidad Instrucciones. Por favor, lea atentamente cada pregunta antes de responder. Conteste a todas las preguntas en el orden presentado y sin revisar las respuestas precedentes. Su información permanecerá en secreto, intente por lo tanto responder lo más sinceramente posible. Las preguntas prevén una serie de respuestas, escoja sólo la que mejor le describe. Gracias por su colaboración. 1. Considerando solamente su ritmo personal de “sentirse en forma”, ¿a qué hora se levantaría si fuese completamente libre para planificar su jornada ? 5:00–6:30 a.m. 6:30–7:45 a.m. 7:45–9:45 a.m. 9:45–11:00 a.m. 11:00–12:00 (mediodía) 2. Considerando solamente su ritmo personal de “sentirse en forma”, ¿a qué hora se iría a la cama si fuese completamente libre para planificar su jornada ? 20:00–21:00 p.m. 21:00–22:15 p.m. 22:15–00:30 a.m. 00:30–1:45 a.m. 1:45–3:00 a.m. 3. En circunstancias normales, ¿en qué medida le resulta fácil levantarse por la mañana ? nada fácil no demasiado fácil bastante fácil 508 A. Adan et al. / European Psychiatry 20 (2005) 503–509 muy fácil 4. ¿Cuán despabilado se encuentra durante la primera media hora después de haberse despertado ? nada despabilado un poco despabilado bastante despabilado muy despabilado 5. Durante la primera media hora de haberse despertado se encuentra usted... muy cansado bastante cansado bastante descansado muy descansado 6. Ha decidido practicar algún ejercicio físico. Un amigo le sugiere que debería hacerlo durante une hora dos veces a la semana y que la mejor hora es entre las 7:00 y las 8:00 de la mañana. Considerando exclusivamente su ritmo de “sentirse en forma”, ¿cómo se encontraría a esa hora ? estaría en optima forma estaría en buena forma me resultaría difícil me resultaría muy difícil 7. ¿A que hora se encuentra cansado/a por la noche y siente la necesidad de dormir ? 20:00–21:00 p.m. 21:00–22:15 p.m. 22:15–00:30 a.m. 00:30–1:45 a.m. 1:45–3:00 a.m. 8. Usted debe hallarse en el momento de máximo rendimiento para realizar una prueba que le cansará mentalmente y que dura dos horas. Si es enteramente libre para planificar su jornada, ¿cuál de los siguientes periodos horarios elegiría ? 8:00–10:00 a.m. 11:00 a.m.–13:00 p.m. 15:00–17:00 p.m. 19:00–21:00 p.m. 9. Suele hablarse de personas de tipo “matutino” y de tipo “vespertino”. ¿A cuál de estos dos tipos cree pertenecer ? claramente “matutino” más “matutino” que “vespertino” más “vespertino” que “matutino” claramente “vespertino” 10. Suponiendo que realiza un trabajo de 8 horas al día y que es totalmente libre de gestionar su tiempo, ¿cuándo preferiría levantarse ? antes de las 6:00 a.m. 6:00–7:30 a.m. 7:30–8:30 a.m. después de las 8:30 a.m. 11. Si tuviese que levantarse siempre a las 6:00 de la mañana, ¿como lo valoraría ? muy difícil y molesto bastante difícil y molesto un poco molesto pero no sería un problema fácil y no molesto 12. ¿Cuánto tiempo transcurre normalmente por la mañana para que se “despierte del todo” después de una noche de sueño ? 0–10 minutos 11–20 minutos 21–40 minutos más de 40 minutos 13. Indique cuál de las siguientes afirmaciones define mejor su patrón de actividad. marcadamente activo por la mañana (despierto por la mañana y cansado por la tarde) moderadamente activo por la mañana moderadamente activo por la tarde marcadamente activo por la tarde (cansado por la mañana y despierto por la tarde) References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] Adan A, Almirall H. Standardization of a reduced scale of morningness in Spanish population: individual differences. Psicothema 1990; 2:137–49. Adan A, Almirall H. Horne and Östberg Morningness–Eveningness Questionnaire: a reduced scale. Person Individ Diff 1991;12:241–53. Adan A, Natale V. Gender differences in morningness–eveningness preference. Chronobiol Int 2002;19:709–20. Alzani A, Natale V. Uno strumento per la valutazione delle differenze individuali nei ritmi circadiani: une versione italiana della Composite Scale of Morningness. Testing Psicom Metodol 1998;5:19–31. Barrett P. Factor comparison: an examination of three methods. Person Individ Diff 1986;7:327–40. Brown FM. Psychometric equivalence of an improved Basic Language Morningness (BALM) Scale using industrial population within comparisons. Ergonom 1993;36:191–7. Caci H, Adan A, Bohlé P, Natale V, Pornpitakpan C, Tilley A. Transcultural properties of the Composite Scale of Morningness: the relevance of the Morning Affect factor. Chronobiol Int 2005;22. Caci H, Nadalet L, Staccini P, Myquel M, Boyer P. Psychometric properties of the French version of the Composite Scale of Morningness in adults. Eur Psychiatry 1999;14:284–90. Caci H, Nadalet L, Staccini P, Myquel M, Boyer P. The Composite Scale of Morningness: further psychometric properties and temporal stability. Eur Psychiatry 2000;15:278–81. Caci H, Robert P, Boyer P. Novelty seekers and impulsives subjects are low in morningness. Eur Psychiatry 2004;19:79–84. Caci H, Robert PH, Dossios C, Boyer P. Morningness–Eveningness for Children Scale: psychometric properties and month of birth effect. L’échelle de Matinalité pour enfants et adolescents: propriétés psychométriques et effet du mois de naissance. Encephale 2005; in press]. Carrier J, Monk TH. Circadian rhythms of performance: new trends. Chronobiol Int 2000;17:719–32. Carrier J, Monk TH, Buysse DJ, Kupfer DJ. Sleep and morningness– eveningness in the middle years of life (20–59 years). J Sleep Res 1997;6:230–7. Chelminski I, Ferraro FR, Petros T, Plaud JJ. Horne and Östberg questionnaire: a score distribution in a large sample of young adults. Person Individ Diff 1997;23:647–52. Chelminski I, Petros TV, Plaud JJ, Ferraro FR. Psychometric properties of the reduced Horne and Östberg questionnaire. Person Individ Diff 2000;29:469–78. De La Iglesia HO, Cambras T, Schwartz WJ, Diez-Noguera A. A forced desynchronization of dual circadian oscillators within the rat suprachiasmatic nucleus. Curr Biol 2004;14:796–800. A. Adan et al. / European Psychiatry 20 (2005) 503–509 [17] Diaz JF, Aparicio M. Relationships between morningness–eveningness and personality styles. Ann Psychol 2003;19:247–56. [18] Duffy JF, Czeisler CA. Age-related change in the relationship between circadian period, circadian phase, and diurnal preference in humans. Neurosci Lett 2002;318:117–20. [19] Greenwood KM. Psychometric properties of the Diurnal Type Scale of Torsvall and Åkerstedt (1980). Ergonom 1991;34:435–43. [20] Greenwood KM. Long-term stability and psychometric properties of the Composite Scale of Morningness. Ergonom 1994;37:377–83. [21] Greenwood KM. An evaluation of the Circadian Type Questionnaire. Ergonom 1995;38:347–60. [22] Horne JA, Östberg O. A self-assessment questionnaire to determine morningness–eveningness in human circadian rhythms. Int J Chronobiol 1976;4:97–110. [23] Kerkhof GA. Inter-individual differences in the human circadian system: a review. Biol Psychol 1985;20:83–112. [24] Kim S, Dueker GL, Hasher L, Goldstein D. Children’s time of day preference: age, gender and ethnic differences. Person Individ Diff 2002;33:1083–90. [25] Mecacci L, Zani A, Rocchetti G, Lucioli R. The relationships between morningness–eveningness, ageing and personality. Person Individ Diff 1986;7:911–3. [26] Muthén BO. Beyond SEM: general latent variable modeling. Behaviormetrika 2002;29:81–117. [27] Muthén LK, Muthén BO. Mplus 2.0 User’s Guide. Los Angeles, CA: Muthén and Muthén; 1998. [28] Natale V, Adan A. Season of birth modulates morningness–eveningness preference in humans. Neurosci Lett 1999;274:139–41. [29] Natale V, Adan A, Chotai J. Further results on the association between morningness–eveningness preference and the season of birth in human adults. Neuropsychopharmacology 2002;46:209–14. [30] Natale V, Alzani A. Additional validity evidence for the Composite Scale of Morningness. Person Individ Diff 2001;30:293–301. [31] Neubauer AC. Psychometric comparison of two circadian rhythm questionaires and their relationship with personality. Person Individ Diff 1992;13:125–31. 509 [32] Park YM, Matzumoto K, Seo YJ, Kang MG, Nagashima H. Effects of age and gender on sleep habits and sleep trouble for aged people. Biol Rhythms Res 2002;33:39–51. [33] Pornpitakpan C. Psychometric properties of the Composite Scale of Morningness: a shortened version. Person Individ Diff 1998;25:699– 709. [34] Rensing L, Meyer-Grahle U, Ruoff P. Biological timing and the clock metaphor: oscillatory and hourglass mechanisms. Chronobiol Int 2001;18:329–69. [35] Smith CS, Folkard S, Schmieder RA, Parra LF, Spelten E, Almirall H, et al. Multinational investigation of morning-evening orientation using a new measure of morningness. Person Individ Dif 2002;32: 949–68. [36] Smith CS, Reilly C, Midkiff K. Evaluation of three circadian rhythm questionnaires with suggestions for an improved measure of morningness. J Appl Psychol 1989;74(5):728–38. [37] Statacorp. Stata Statistical Software: Release 7.0. College Station, TX: Stata Press; 2000. [38] Tankova I, Adan A, Buela-Casal G. Circadian typology and individual differences. A review. Person Individ Diff 1994;16:671–84. [39] Torsvall L, Åkerstedt T. A Diurnal Type Scale. Construction, consistency and validation in shift work. Scand J Work Environ Health 1980;6:283–90. [40] Vink JM, Groot AS, Kerkhof GA, Boomsma DI. Genetic analysis of morningness and eveningness. Chronobiol Int 2001;18:809–22. [41] Waterhouse J, Folkard S, Van Dongen H, Minors D, Osens D, Kekhof G, et al. Temperature profiles, and the effect of sleep on them, in relation to morningness–eveningness in healthy female subjects. Chronobiol Int 2001;18:227–47. [42] Wilson GD. Personality, time of day and arousal. Person Individ Diff 1990;11:153–68. [43] Zickar MJ, Russell SS, Smith CS, Bohlé P, Tilley AJ. Evaluating two morningness scales with item response theory. Person Individ Diff 2002;33:11–24.