Time in bed, quality of sleep and school functioning of children


Meijer Dr University of Amsterdam, Faculty of Social and Behavioural Sciences, Department of Education, PO Box 94208, 1090 GE Amsterdam, the Netherlands. Tel.: 31 20 525 1572; fax: 31 20 525 1500; e-mail: anne@educ.uva.nl


This study describes the relationship of time in bed and quality of sleep with concentration and functioning at school. Neurotic and psychosomatic symptoms have been used as control variables. The sample consisted of 449 Dutch children in the seventh and eighth grades of elementary school. The age of the children varied between 9 y 5 mo and 14 y 5 mo. Seven schools participated in the research, with a total of 18 classes. The results indicated that 43% of the children had difficulty getting up in the morning. Furthermore, 15% of the children reported sleep problems and 25% did not feel rested at school. Time in bed and sleep quality show no relationship with concentration. Sleep quality, feeling rested at school and less distinct bedtimes were clearly related to school functioning. Another result was that children who had no difficulty getting up displayed more achievement motivation. Being open to the teacher’s influence and achievement motivation depended mainly on sleep characteristics. Not getting bored at school, self-image as a pupil and control over aggressive behaviour were also influenced by gender, age, neuroticism and neurosomaticism.


A good night’s sleep is commonly accepted to be a substantial factor in allowing adequate daytime functioning of children. Children have to go to bed early to be able to concentrate at school and be alert. To understand the significance of sleeping well, a distinction must be made between sleep duration and sleep quality; two aspects of sleep that are uncorrelated ( Totterdell et al. 1994 ) and may have different effects on daily functioning. In most studies sleep duration is defined as the period of sleep from bedtime until awakening in the morning. Thus, it actually can be understood as ‘time in bed’. For quality of sleep the literature frequently cites criteria such as early onset, fewer interruptions and fewer early awakenings.

For European children aged 11–12 y the mean time in bed varies from 9 to 10 h and there is little relation between time in bed and perceived health, morning tiredness or psychosomatic symptoms ( Tynjäläet al. 1993 ). Contrasting findings were found in a Japanese study by Gau and Soong (1995). They showed that a shorter time in bed and a greater difference between self-reported required sleeping time and actual sleeping time were correlated with sleepiness during the day, tiredness and difficulty getting up. They also found a negative correlation between the academic pressure with which adolescents are faced and hours of sleep. To explain the adolescents’ sleep delay, Carskadon et al. (1993 ) introduced the idea of a biologically mediated phase preference delay that is linked to pubertal development and cannot be attributed merely to psychosocial factors. This has been validated by the finding that early starters at school exhibit significant sleep deprivation and daytime sleepiness ( Carskadon et al. 1998 ; Epstein et al. 1998 ).

The effects of sleep duration on cognitive performance have been investigated mainly in adults. The effects of sleep deficiency are task specific. Longer tasks, low demand tasks and tasks requiring intensive concentration will be affected. Achievement motivation will also decrease ( Webb 1992). Moreover, a meta-analysis of 19 studies indicates that cognitive performance is affected more by sleep deprivation than motor performance and that mood is affected much more than either cognitive or motor performance. The effects on cognitive performance and mood are more prominent with partial sleep deprivation (continuously ≤ 5 h in a 24-h period) than with temporary, but complete sleep deprivation ( Pilcher and Huffcutt 1996). The effects of sleep delay on the academic performance of students are presented in Wolfson and Carskadon (1998). Students, who perform less well, seem to obtain less sleep, have later bedtimes and more irregular sleep/wake schedules. A recent study of Randazzo et al. (1998 ) shows that a sleep restriction of 5 h during a single night in a sleep laboratory imposed upon children aged between 10 and 14 y leads to shorter sleep latencies and significantly worse performance in verbal processing, creativity and abstract thinking. Measures of rote performance, less complex cognitive functions and figural creativity were not affected. In most sleep-duration studies, however, sleep duration is confined to a relatively extreme deprivation of sleep and it is questionable, therefore, whether fluctuations in sleep duration under normal conditions have similar effects.

With regard to the quality of sleep, Rimpelä and Rimpelä (1983) report that ≈ 10% of Finnish youngsters have difficulty falling asleep and awake frequently at night. For elementary school children, tiredness in the morning seems to be connected with poor school achievement. The more difficulty children have falling asleep and the more frequently they awake at night, the more tired they feel the next morning. Tynjäläet al. (1993 ), their European study of 11–12-year-olds, showed that Finnish children have the most difficulty (33%) and Spanish children the least difficulty (16%) falling asleep. They found that difficulty falling asleep is related to perceived health, the experience of morning tiredness and, in particular, psychosomatic symptoms. Dahl (1996) concluded that inadequate sleep in children caused by fragmented sleep, sleep deprivation or poor quality of sleep results in difficulties with focused attention, irritability, emotional instability and a low threshold for frustration and distress.

Here, we investigate the relationship of time in bed and quality of sleep with concentration and functioning at school in elementary school children in grades seven and eight. In accordance with Van der Wolf (1995), we assume that a child is functioning well at school if he or she reacts with focused attention on the teacher, shows a satisfying achievement motivation and does not get bored. On the basis of the literature, we expect a positive relationship between time in bed and quality of sleep and concentration and functioning at school. Since there is also a strong relationship between quality of sleep and psychosomatic and neurotic symptoms ( Morrison, McGee and Stanton 1992; Stores 1996; Manni et al. 1997 ), these variables were used as control variables.



Data concerning school functioning, concentration and neuroticism and neurosomaticism were collected using standardized questionnaires. We constructed a questionnaire for the sleep data. Quality of sleep has been measured on the basis of the subjects’ self-ratings on a visual analogue scale. In accordance with Totterdell et al. (1994 ), children were assessed on aspects of sleep latency, number of awakenings at night, sleep latency after awakenings and perception of sleep quality. These are also criteria for insomnia as provided in the DSM-IV.

To achieve a sufficient sample size, the questionnaires were completed in the classroom. The researchers administered the questionnaires.

All questionnaires were presented in the first trimester of the school year. For each class this occurred in the morning. To avoid weekend and holiday effects, data were not collected on Mondays or shortly after the midterm break.


The participants in our study were 449 children from the seventh and eighth grades of elementary schools in Amsterdam. Ninety-nine per cent of the children in the selected schools completed the questionnaires. The ages varied between 9 y 5 mo and 14 y 5 mo, with a mean age of 11 y 3 mo (SD=8 mo). The total group consisted of 228 boys and 219 girls. Grade seven consisted of 226 children and grade eight of 223 children. Seven schools with a total of 18 classes participated in our research. The schools were situated in middle-class neighbourhoods, therefore most of the children were white. There were no differences between the schools in terms of starting time.

In 278 cases both parents were employed; one of the parents was employed in 145 cases and 25 parents were unemployed. In 99 cases the parents were divorced.


Measures of time in bed and quality of sleep were obtained by means of a questionnaire constructed by the researchers, containing questions about children’s sleep behaviour (Appendix A). Time in bed was based on the time children went to bed until the moment of getting up the next morning. We used separate bedtimes for time in bed during the night before the data collection and time in bed in general, because we assumed that bedtimes in general were ideal bedtimes as set by the parents, and bedtimes in the night before the data collection represented more actual bedtimes. The reported bed times and rise times are real times, not categories.

Quality of sleep was measured on the basis of a scale consisting of four closed questions with three ordinal response categories ranked from 1 to 3. These questions referred to sleep latency, frequency of awakenings during the night, the degree to which the children fell asleep again after awakening, and a question about their subjective sleep perception. The one-dimensional solution with the optimal scaling program HOMALS ( Gifi 1990), which provides scale values that can be analysed at interval level, correlates 0.98 with the sum of the rankings. Since the interpretation of this sum is more obvious than the HOMALS solution, we used the sum score of the rankings as our measure of sleep quality. The lower bound of reliability of this quality-of-sleep scale was alpha=0.72. The minimum score was 4 (sleeping badly) and the maximum score was 12 (sleeping well).

In addition to questions about time in bed and quality of sleep, the questionnaire also contained questions about children’s sleep customs, whether they had difficulty getting up the following morning and whether they felt rested at school. The last two questions could only be answered by ‘Yes’ or ‘No’ to force an explicit answer.

Concentration was measured by means of the Bourdon-Vos test ( Vos 1992). The Bourdon-Vos test is a time-limited paper and pencil test that measures the level of continuous selective attention with a form of 33 rows filled with 24 figures of three, four and five unsystematically placed dots. Working row by row, the task of the subject is to strike out all figures with four dots. Tasks of this type have a long-standing tradition in the measurement of selective perception ( Johnston and Dark 1986). Two aspects of this test were included in our research: the accuracy and the speed at which children make this test. The test can be taken both individually and at class level. Vos (1992) found no differences between the test results of boys and girls. Vos (1992) reported a strong internal consistency (alpha=0.98).

To determine children’s school functioning we administered the School Perception Questionnaire ( Van der Wolf 1995). This questionnaire of Likert type items is intended for children in grades 6, 7 and 8 of elementary school and indicates whether the child is functioning well at school. The following subscales were used in this research: (i) teacher’s behaviour as perceived by the child (8 items, e.g. Our teacher is one of the funniest persons in our school); (ii) being busy and not getting bored in the classroom (8 items, e.g. I often have to wait until the other children are ready with their job); (iii) self-image concerning school performance (7 items, e.g. I am worried about my report card); (iv) achievement motivation (12 items, e.g. When I am learning, a. I often think at other things or b. I am concentrating; Setting to do my homework, a. I like it or b. troubles me often; Studying hard, a. I like, b. I don’t like much, c. I hate); (v) control over aggression (8 items, e.g. When I am angry or fall into rage, I start to run). A higher score on the scale means a more prominent presence of the characteristic concerned. Reported alpha values for these scales are 72, 70, 75, 86, and 79 ( Van der Wolf 1995). In this study Cronbach’s alphas for the five subscales are 0.74, 0.70, 0.85, 0.76 and 0.86.

To control for psychosomatic and neurotic symptoms the Amsterdam Biographic Questionnaire for Children (ABV-K) was administered. From this questionnaire, we used the measures of neuroticism (occurrence of psychoneurotic complaints) and neurosomaticism (occurrence of functional somatic complaints). A higher score means more neurotic and neurosomatic symptoms. Standardized scores are known for boys and for girls. For boys of our age group the reliability of the subscale neuroticism is alpha=0.82 and for girls alpha=0.80, for neurosomaticism alpha=0.74 for boys and alpha=0.77 for girls ( van Dijl and Wilde 1982). Cronbach’s alpha values in the current study for neuroticism are 0.80 (boys) and 0.82 (girls) and for neurosomaticism 0.73 (boys) and 0.76 (girls).

Statistical methods

The results are presented in four sections: (i) time in bed and quality of sleep; (ii) concentration, school functioning and neurotic and neurosomatic complaints; (iii) relationship between sleep and concentration; and (iv) relationship between sleep variables and school functioning. Using hierarchic regression analyses, we examined whether time in bed and quality of sleep can be regarded as adequate predictors of concentration and functioning at school.

To prevent over-interpretation because of the large sample size in this study, we used, analogous to Wolfson and Carskadon (1998), an effect size criterion in addition to a statistical significance criterion for presenting our correlational results. Therefore, we restrict the discussion to significant correlations of an effect size between ‘small’ and ‘medium’ according to the norms of Cohen (1988). Using the percentage of variance explained, a Pearson product moment correlation of r=0.20 as the effect size is almost the midpoint between Cohen’s small (r=0.10) and medium (r=0.30) effect size. For phi we use the same criterion.


Time in bed and quality of sleep

In general, time in bed varied between 502 and 720 min, with a mean of 616 min and a SD of 40 min. Time in bed during the night before the data collection was shorter: it varied between 295 and 735 min, with a mean of 580 min and a SD of 60 min. The correlation between time in bed in general and time in bed the night before the data collection is r=0.63. Most children (83%) slept alone and had their own room. In most cases (91%), the parents decided when their children had to go to bed.

Concerning the quality of sleep, the children reached a mean score of 8.3 (SD=1.79), with a minimal score of 4 and a maximal score of 12. A total of 15.1% scored one or more SD below average (≤ 6 or mediocre quality of sleep), and 7.3% scored two or more SD below average (≤ 5 or bad quality of sleep).

Of the total number of children, 43% reported difficulty getting up in the morning and 25% indicated that they did not feel rested at school. Children who reported more difficulty getting up in the morning felt less rested at school (phi=−0.31, P < 0.000).

Table 1 shows that there were no gender differences on the sleep measures with respect to quality of sleep, time in bed in general, getting up in the morning, bedtime at weekends and feeling rested at school. However, boys displayed a shorter time in bed during the night before the data collection (M=572 min; SD=42 min) than girls (M=587 min; SD=58 min). Boys went to bed later on schooldays in general (M=21.00; SD=33 min) and the day before the data collection (M=21.35; SD=53 min) than girls (M=20.51; SD=31 min, and M=21.21; SD=52 min). There was also a small, but significant, relationship between age and time in bed: older children displayed a shorter time in bed in general (r=−0.232, P < 0.000) and during the night before the data collection (r=−0.214, P < 0.000).

Table 1.   Means and SD for total group (N = 449), means for boys and girls, and t-test and significance of t for sex differences on age in months, Bourdon-Vos, School Perception Questionnaire, Amsterdam Biographic Questionnaire for Children (ABV-K) (N = 449), and sleep variables. Means and SD of the norm group for the School Perception Questionnaire, the percentile scores for the total group of the Bourdon-Vos and the decile scores for boys and girls for ABV-K are between brackets. Bedtimes and get-up times are given in h and min Thumbnail image of

The correlations between quality of sleep, and difficulty getting up (r=−0.23; P < 0.001), or feeling rested at school (r=0.30; P < 0.001), are relatively low. There was no clear correlation between time in bed and quality of sleep.

Concentration, school functioning and neurotic and neurosomatic complaints

Table 1 presents the children’s results on the Bourdon-Vos test for concentration, the School Perception Questionnaire for school functioning and the ABV-K for neurotic and neurosomatic complaints.

Because standardized scores at classroom level are not yet available for the Bourdon-Vos test, the standardized scores for the individual test had to be used. The mean speed on the Bourdon-Vos for the research group (16.3 s.) was slower than that of the standardized group, which were based on individual scores (14.5 s.). Children in the research group also turned out to be less accurate than the standardized group. Our results show a large standard deviation, which is due mainly to the fact that 5% of the children made 35–71 mistakes, possibly due to concentration problems as a consequence of classroom testing.

The scores on the School Perception Questionnaire are comparable with those of the standardized group. No centile scores are available for this questionnaire. Boys and girls scored significantly differently on waiting and getting bored, self-image as a pupil and aggression control in the classroom.

On the ABV-K the mean scores of the children in the research group are comparable with those of the standardized group. As usual on this test, girls scored higher on neurotic complaints and neurosomatic complaints than boys. For this reason the ABV-K has been standardized separately for boys and girls. The centile scores of the boys and the girls indicated that the research group achieved average scores on this questionnaire.

Indirectly related to the topic of our study, interesting results could also be noted between neurotic and neurosomatic complaints and the different aspects of sleep. Again, no relationship was found between time in bed and neurotic and neurosomatic complaints. Quality of sleep, however, showed a significant correlation of r=−0.416 (P < 0.000) with neurotic complaints and r=−0.366 (P < 0.000) with neurosomatic complaints. The significant correlations of feeling rested at school and difficulty getting up in the morning with neurotic complaints were r=−0.295 (P < 0.000) and r=0.252 (P < 0.000), respectively, and those with neurosomatic complaints were r=−0.314 (P < 0.000) and r=0.252 (P < 0.000).

Relationship between sleep and concentration

The relationship between time in bed and quality of sleep, on the one hand, and children’s concentration, on the other hand, was studied by means of two hierarchical regression analyses: one for speed factors and one for measures of accuracy. Assuming that age and gender were important variables in predicting differences in concentration, these were entered first as predictors. Subsequently, measurements of neurotic and neurosomatic complaints were introduced to investigate the effects of psychological control variables. Next, time in bed (both general and the night before), quality of sleep, feeling rested at school and difficulty getting up in the morning were introduced in the regression equation to determine their effect on concentration.

The results are presented in Table 2. As is clear from this table, there is but a small relationship between age and accuracy. None of the other variables show any relationship with speed or accuracy as measured by the Bourdon-Vos test. The combination of variables into similar sets and the hierarchical introduction does not improve the relationship. The introduction of sleep characteristics, such as time in bed, quality of sleep, feeling rested at school and having difficulties with getting up in the morning, causes a change in the square of the multiple correlation (the proportion of variance explained) by 1.9% for speed and 1.1% for accuracy only. Therefore, on the basis of these results it may be concluded that sleep characteristics, such as time in bed, quality of sleep, feeling rested and difficulty with getting up in the morning, do not have a negative impact on concentration as measured by this test. A similar finding emerges when bedtime and time of awakening are correlated with the speed and accuracy scores. The predominant conclusion must be that neither time in bed nor quality of sleep lead to differences in concentration as measured by the Bourdon-Vos test.

Table 2.   Hierarchical regressions for concentration scores as measured by speed and accuracy with child variables, child characteristics, time in bed, and quality of sleep Thumbnail image of

Relationship between sleep and school functioning

Five hierarchical regression analyses were carried out to determine whether sleep characteristics influence school functioning: one for each of the subscales. Again assuming that age, gender and child characteristics on the ABV-K are important they are controlled for by entering age and gender at step 1, and child characteristics on the ABV-K (neurotic complaints and neurosomatic symptoms) at step 2. Time in bed, quality of sleep, difficulty getting up in the morning and feeling rested at school were entered at step 3. Outliers were not removed from the analyses because there were no good theoretical reasons to do so. The distribution of the residual scores appeared to be normal for each of the variables analysed. The results are given in Table 3.

Table 3.   Hierarchical regressions for measures of school functioning with child variables, child characteristics and sleep characteristics [time in bed (gen) = time in bed in general, time in bed (b.d.) = time in bed before data collection] Thumbnail image of

In contrast with concentration, Table 3 shows statistically significant influences of sleep characteristics on the school functioning scales. Inclusion of the sleep variables led to an improvement in the percentage of variance explained (R2) of 5.9% for teacher’s influence, 3.6% for being busy and not getting bored at school, 4.2% for self-image as a pupil, 17.1% for achievement motivation and 8.1% for control over aggression. Of these sleep characteristics, time in bed (both general and the day before) had no direct relationship with school functioning, except for a standardized regression coefficient 0.13 for time in bed in general with achievement motivation. Difficulty with getting up in the morning did not show a relationship with school functioning either, apart from a regression coefficient of −0.22 with achievement motivation: as children have more difficulty getting up, they are less motivated to do their best at school.

Quality of sleep had a direct positive relationship with four of the five aspects of school functioning: as the quality of sleep was higher, children were more receptive with regard to the teacher’s influence, had a more positive self-image as a pupil and displayed more achievement motivation and control over their own aggression. No relation could be found with not getting bored at school. A somewhat similar pattern of relationships was seen for feeling rested at school, which also had a positive influence on school functioning. Children who feel better rested display a more positive self-image as a pupil, more achievement motivation, have more control over their aggressive behaviour in the classroom, are less bored and are more receptive to their teacher.

Also note that being receptive to the teacher’s influence and achievement motivation appeared to depend mainly on sleep characteristics, while not getting bored, self-image as a pupil and control over aggressive behaviour appeared to be also influenced by gender, age, neurotic control variables or combinations of these variables. Thus, these control variables played a different part in the various aspects of school functioning. Apart from gender, indications of more neurotic complaints were associated with getting bored at school, a more negative self-image concerning school functioning and more manifestations of aggressive behaviour.


Similar to most studies, the sleep duration of the children in our study has to be understood as ‘time in bed’. It should be noted in this context that most measurements of sleep duration in sleep research are subjective in nature. Although these are not as reliable as polysomnographic records, they have been highly correlated with polysomnographic estimates, at least in adults ( Tepas and Mahan 1989). Epstein et al. (1998 ) showed high correlations between children’s and parents’ answers on questions concerning sleep habits and sleep times, illustrating the validity of sleep questionnaires for children. For young children Sadeh (1996) found significant correlations between objective sleep measures from actigraphy and reported sleep measures from parents.

In this study, the mean time in bed (10 h 15 min) is somewhat longer than mentioned in the literature ( Tynjäläet al. 1993 ). As expected, the time in bed the night before the data collection was shorter (9 h 40 min). This is supposed to be the actual time in bed, whereas time in bed in general reflects the ideal bedtime as set by parents. Although small, we did find some different results depending on the relationship with general time in bed or time in bed the day before the data collection. Use of various parameters for estimating sleep time will probably provide more insight into the factors that contribute to adequate functioning during daytime. Examples of such parameters are general or actual bedtimes, as well as getting-up time and start time of schools ( Carskadon et al. 1998 ; Epstein et al. 1998 ), or difficulty with getting up in the morning.

We obtained surprisingly low correlations between quality of sleep and difficulty getting up in the morning and feeling rested at school. Perhaps the correlations of these variables with quality of sleep could have been improved by using physiological variables (e.g. EEG, ECG). The correlations between physiological and subjective measurements of quality of sleep are rather poor, however ( Hofman 1994; Perlis et al. 1997 ).

A number of interesting conclusions can be drawn from our results concerning concentration and school functioning. For example, it is striking that children’s concentration does not appear to be related to time in bed and quality of sleep. Concerning time in bed, Epstein et al. (1998 ) also found no correlation between time in bed and difficulty with concentrating and paying attention. Based on their meta-analysis of studies on the effects of sleep deprivation, Pilcher and Huffcutt (1996) state that cognitive performance is affected by sleep deprivation. One of their findings is that partial sleep deprivation (< 5 h in a 24-h period), which can be best compared with lack of sleep in a normal situation, has the strongest negative effect on cognitive performance and mood. In addition, cognitive performance decreased as the length of the task increased. A possible reason for this lack of relationship between time in bed and quality of sleep with children’s concentration could be that the time in bed of the children in our research is much longer than 5 h, so that no deprivation of sleep occurred. However, a child that requires 11 h of sleep who routinely obtains 8 h every night would likely sustain a significant sleep debt over time. This is very different to a one-night sleep restriction to 5 h, but is likely to be clinically relevant and to affect performance measures.

The lack of effects on concentration could also be due to the test used to measure children’s concentration. This test is simple and takes only a relatively short time, which might not be long enough to show the negative effects of lack of sleep. This conclusion is in accordance with the finding of Randazzo et al. (1998 ) who showed that after a single night of restricted sleep imposed on children aged 10–14 y, verbal creativity and abstract thinking are impaired, but the less complex cognitive functions are not. Another possibility for the lack of effects may be due to the high degree of individual differences in concentration (particularly accuracy) and in sleep variables. A within-subject design may thus be necessary to detect the relationship between sleep performance and concentration.

Time in bed does not show a relationship with school functioning either. Low, but significant, correlations have, however, been found with bedtimes before the data collection, during the week and at the weekend. In contrast, quality of sleep is clearly related to different aspects of school functioning. After controlling for gender, age, neurotic complaints and neurosomatic complaints, quality of sleep appeared to contribute 3.6–17.1% of the explained variance of the different aspects of school functioning. Achievement motivation, in particular, was found to be affected by quality of sleep. This is in agreement with the conclusion of Webb (1992), who stated that suffering from sleep deficiency results in decreased achievement motivation. Likewise, feeling rested during school hours shows a positive relationship with school functioning and the negative relation between difficulty getting up in the morning and achievement motivation is also noteworthy. Here, our findings are in accordance with the results of Rimpelä and Rimpelä (1983) and Tynjäläet al. (1993 ). These researchers found a relationship between quality of sleep (morning tiredness, difficulties with falling asleep) and poor school achievement and psychosomatic symptoms.

Age, gender, neurotic and neurosomatic complaints have been introduced as control variables in our study. The results show a moderate relationship between age and accuracy and age and self-image as pupil. Not getting bored at school, self-image as a pupil, achievement motivation and control of aggressive behaviour are influenced by gender. Girls appear to be less bored at school, to have a higher achievement motivation and to show less aggressive behaviour. Boys present a better self-image as a pupil. These results are consistent with literature concerning gender differences on school functioning ( Harter 1998) and aggression ( Tremblay et al. 1996 ). Both neurotic and neurosomatic complaints show a negative relationship with quality of sleep and feeling rested at school, and a positive relationship with having difficulty getting up in the morning. They also relate to various aspects of school functioning. Like quality of sleep, neurotic complaints appeared to be predictors of school functioning.

Based on our results, we conclude that quality of sleep has a substantial impact on school functioning for children aged 10–14 y. Concurrently, Carskadon et al. (1993 , 1998) conclude that this age group is particularly vulnerable for sleep problems resulting from psychosocial influences and changes in bioregulatory systems controlling sleep. Our results, showing that 15% of the children report sleep problems, 43% have difficulties with getting up in the morning and 25% do not feel rested at school, confirm this. In children with problematic school functioning we would, therefore, recommend verifying whether these children also have sleep problems. The different effects of quality of sleep, sleep duration and time in bed on school functioning need further investigation.


Appendix A

Listing of questions on sleeping

1. Do you have a bedroom of your own?



2. What time did you go to bed yesterday? (try to describe this as accurately as possible)

3. What time do you usually go to bed on the days you have to go to school?

4. What time do you go to bed in the weekends?

The next questions are about the days on which you have to go to school.

5. Are you allowed to decide for yourself what time you go to bed?



6. Would you prefer to go to bed at another hour?

Earlier, at

Later, at


7. Does someone put you to bed?



8. When you’re in bed, does anybody come and say goodnight?



9. What do you do if you do not have to turn off the light immediately?


Play a game

Something else, namely…

10. Do you have to turn off the light as soon as you are in bed?



11. What time do you usually turn off the light?

You do not turn off the light yourself.

You turn off the light at…o’clock

12. When you’re in bed and the lights are turned off:

You fall asleep at once.

You stay awake for a while.

It takes you a long time to fall asleep.

13. Do you sometimes wake up during the night?

never, sometimes, nearly every night

14. If you wake up during the night:

Mostly you don’t notice.

You fall asleep again soon.

It takes you a while to fall asleep again.

15. Do you sleep well at night?



Yes, always

16. How do you wake up in the morning?

You just wake up (spontaneously).

By the alarm clock.

Someone calls you.

Other, namely…

17. What time did you wake up this morning? (describe as accurately as possible)

18. What time do you wake up in the morning (on days that you have to go to school)?

19. Do you find it difficult to get up in the morning?



20. When you are at school, do you feel rested?