Effects of pedagogical ideology on the perceived loudness and noise levels in preschools.

High activity noise levels that result in detrimental effects on speech communication have been measured in preschools. To find out if different pedagogical ideologies affect the perceived loudness and levels of noise, a questionnaire study inquiring about the experience of loudness and voice symptoms was carried out in Iceland in eight private preschools, called "Hjalli model", and in six public preschools. Noise levels were also measured in the preschools. Background variables (stress level, age, length of working career, education, smoking, and number of children per teacher) were also analyzed in order to determine how much they contributed toward voice symptoms and the experience of noisiness. Results indicate that pedagogical ideology is a significant factor for predicting noise and its consequences. Teachers in the preschool with tighter pedagogical control of discipline (the "Hjalli model") experienced lower activity noise loudness than teachers in the preschool with a more relaxed control of behavior (public preschool). Lower noise levels were also measured in the "Hjalli model" preschool and fewer "Hjalli model" teachers reported voice symptoms. Public preschool teachers experienced more stress than "Hjalli model" teachers and the stress level was, indeed, the background variable that best explained the voice symptoms and the teacher's perception of a noisy environment. Discipline, structure, and organization in the type of activity predicted the activity noise level better than the number of children in the group. Results indicate that pedagogical ideology is a significant factor for predicting self-reported noise and its consequences.

Introduction

The aim of preschools is to foster children's education and good behavior and to intensify their language development and skills. For that purpose, three main factors have to interact:

The children have to hear the speech sounds, syllables, words, and sentences satisfactorily,

The voice of the teacher has to be clear and durable, and

There must be good acoustics in the classrooms for sound to be carried without distortion and to attenuate noise.

Too high noise levels in the teaching premises can place children's education at a high risk. Noise in a learning environment is a mixture of background noise and activity noise that by nature are different. Background noise is a stationary noise generated by classroom equipment, electrical devices, air conditioning, heating, lights, fans, and noise from external traffic.

Activity noise is unpredictable and unstable both in its timing, duration, frequency (Hz), and level (dB). It is generated by human beings, by their activities, body sounds, behavior, and their use of items such as movable furniture. Activity noise seems to annoy people more than background noise.[1]

Several research projects have indicated that the level of noise in preschools lies at around 70 dB(A). The lowest mean value has been shown to be 67 dB LAeq[2] and the highest mean values have been around 74-76 dB(A).[34] The noise level has been found to be even greater at the level of the children's ears,[5] as most of the noise is originated by the children themselves.

Recommendations for noise levels and reverberation times

Guidelines for classroom acoustics can be found in UK in Building Bulletin 93 and Standard S12.60 of the American National Standards Institute.[6] According to it, the maximum level of background noise in a classroom caused by equipment, etc. should be 35 dB LAeq, 1h, the same as in the Icelandic Building Legislations.

In Iceland there are no specific laws or standards that limit noise in learning environments although they exist for other workplaces, as under the Icelandic Occupational Health and Safety at Work (OHSW) laws.[7] In places such as offices where important discussions are conducted or a high degree of concentration is required, the noise level should be no more than 50 dB LAeq, 8h. This is consistent with research that proved that students can listen effectively only when noise levels do not exceed 40-50 dB(A) in occupied classrooms.[8] Thus, typical noise levels found in preschools, as discussed above, exceed recommended noise levels by 20-25 dB(A) and thus, may have marked effects on communication and also harmful physiological and psychological effects on teachers and pupils.

Apart from noise, reverberation of sound in a space affects speech intelligibility. The average reverberation time recommended by Icelandic standards in preschools is 0.5 s that is in line with ANSI Standard S12.60. However, American Speech Language Hearing Association (ASHA)[9] recommends that for young listeners, reverberation time should not exceed 0.4 s if they are to obtain optimum communication efficiency. Speech recognition for adult listeners is not reduced until reverberation time exceeds approximately 1 s[10] because adult listeners need fewer auditory clues to understand meaning, as they have more preexistent knowledge and can thus, reconstruct language from fewer phonemes. However, if there is simultaneously some noise present, as is mostly the case in real-life situations or if the listener has a mild hearing loss, speech perception decreases under acceptable levels of reverberation.[11]

Noise and poor acoustics as risk factors for teachers’ voices

Voice problems are common for teachers[1213] and preschool teachers are especially at risk of developing voice disorders.[14151617] For instance, approximately 37% of the preschool teachers have been shown to have voice disorders, compared to 12% of nurses who talk as much as preschool teachers but in a different kind of work environment.[16] Additionally, research findings have suggested that one-fifth to one-third of teachers are absent from work as a result of voice problems during a working year.[1819] There has been an alarming tendency of a considerable increase of voice disorders among teachers in the last two decades.[20]

The causes of voice disorders have been found to be more environmental than genetic.[2122] Probably the most common external factors identified as having a damaging effect on teachers’ voices are activity noise and poor acoustics,[21123242526] as the speaker has a tendency to automatically raise his/her voice amid noise exceeding 40 dB(A), a phenomenon that is called the Lombard effect.[272829]

Noise is also an indirect risk factor for a voice disorder: it arouses annoyance and stress-related symptoms.[30] Stress, in turn, affects voice use[3132] and increases the incidence of voice symptoms.[25]

Noise effects on children's speech perception, learning ability, and well-being

In Iceland, over 90% of children aged 1-5 years attend preschools with almost all of them for 8 h a day and 5 days a week.[33] As young children are developing language skills, their vocal behavior is louder and contains more unstructured sounds and causes more noise than the behavior of older children.[334] Since children spend most of the time they are awake in preschools, the noise may have multiple damaging effects on children.

Bradley and Sato[35] found that preschool-aged children need much better listening conditions than older children. The ratio between the level of a desired signal and the level of background noise [called signal-to-noise ratio (S/N ratio)] in a classroom should be +15 dB for 11-year-old children, while for 6-year-old children, it should be 7 dB higher than that for the older children. For disabled children, the S/N-ratio has to be even higher.[3637]

Noise affects children detrimentally by impairing speech perception[8383940] and hence, risking children's speech development[41] and performance.[42] Neuman et al.[40] found out that noise decreased speech perception more in children than in adults. Children do not generally reach adult-like performance on recognition tasks in noisy or reverberant conditions until approximately 13-15 years of age;[394043] in particular, the ability to identify consonants that is responsible for 60% of the intelligibility of speech[44] in such conditions may not mature until the late teenage years.[43]

The perception of speech may be further impaired by otitis media. Between 76% and 95% of all children have been found to have at least one episode of the illness with effusion by the age of 6 years.[39] Additionally, one-third of all children suffer from persistent otitis media during their first 3 years of life, a vital time for speech and language development. Since effusion in the middle ear attenuates perceived sound, these children may not be as effective listeners as adults, especially in combined conditions of noise and reverberation.[4345]

Poor listening conditions affect not only speech perception but also children's well-being and learning opportunities by increasing listening effort,[46] decreasing concentration,[1] and impairing attention.[47] Children from reverberant and noisy classrooms perform less well in a phonological processing task,[48] have retarded semantic[49] and recognition memory[50] as well as short-term memory of spoken items.[48] Noise may also impair cognitive development in children, especially reading comprehension.[50] Furthermore, noise affects children's voices. Children attending the noisiest kindergartens or large groups have been found to produce voice hyperfunctionally, and have breathy or hoarse voice quality.[5]

A psychological reaction that noise usually evokes is annoyance. In particular, intermittent noise in classrooms seems to annoy children[51] and influence children's relationships with their equals and teachers that are less positive than those of children from classrooms with good acoustics.[48]

The purpose of the study

Since noise levels inside classrooms have been found to depend on the number of children and the activities in which they are engaged,[34] one could assume that ideology with different educational and pedagogical methods may be a factor in determining the sound environment in preschools. The purpose of this research was to find out if the influence of working practices in preschools due to different pedagogies is related to the voice well-being of teachers, their perceived loudness, and noise levels in preschools. Two different kinds of preschools in Iceland, public preschools and “Hjalli model” preschools, were compared with each other. The “Hjalli model” preschools follow their own ideology with specific policies for instance, regarding toys, guidance on behavior and gender separation. Possible effects of teaching boys and girls separately on noise levels were not studied in this research. In addition to the type of preschool in question, additional variables that may affect voice use and sound environment were also studied.

According to Statistics Iceland,[33] in 2012 the total number of children in the public preschools in this study was 659 (mean 110, range 51-146) and in the “Hjalli model” preschools it was 869 (mean 109, range 32-152). Differences in the working practices of these two preschool types are presented in Table 1.

Table 1

Working practices in public preschools and “Hjalli model” preschools

Working practices	Public preschools	“Hjalli model” preschools
Average number of children in groups	20 or the whole department together with 2-3 teachers except 2-3 times per week in a small group with one teacher for a short time	6-10 in a group at a time with one teacher in planned activity for the majority of the day. The rest of the day, in smaller groups (2-4 children) with one teacher supervising 4-5 groups in different activities
Gender	In the same group	In separate groups
Toys	Mostly commercially sold	Open-ended play material such as blocks and clay and no use of traditional toys
Environment	Visually rich	Visually minimal
Instructions and guidance for children	Mostly given orally, also pictures used	Given as far as possible nonverbally, e.g., traffic lines on the floor for avoiding bumping and work areas marked with mats and shelves

The specific questions of this study for the two types of preschools were as follows. Are there any differences between the teachers’ opinions of:

Prevalence of voice symptoms and the timing of these;

Stress;

Perceived loudness and sources of noise; and

Hearing problems of children? Activity noise levels were also measured in one public preschool and one “Hjalli model” preschool to examine the possible differences in activity noise levels.

Methods

Schools

The participants came from all “Hjalli model” preschools (eight) and for comparison, six public preschools were selected with similarity of the type and age of the buildings, number of teachers, and the number and age of the children. The majority of all the buildings were built of concrete (six out of eight “Hjalli model” preschools and all the public preschools buildings) where nine were built after the year 2000 (five “Hjalli model” preschools and four public preschools).

The average number of teachers: “Hjalli model” - 23 teachers (6-32); public preschools - 24 teachers (11-37). The number of children: the average number of children in the “Hjalli model” preschool was 111 (20-150) and in public preschools was 110 (52-153). The children in “Hjalli model” were 2-6 years old whereas the children in the public schools were 1-6 years old.

All schools, except one of the “Hjalli model” preschools and one of the public schools, were functioning in buildings built as preschools. Collective agreements and salaries were similar in the “Hjalli model” schools and the public schools.

Participants

The total number of participating teachers (with and without teachers’ certification) in the “Hjalli model” schools was 154 and in the public schools was 159. The questionnaires were distributed among the teachers by the head teachers of each school. The participants took part voluntarily after getting information about the study. Employees who were not directly responsible for the children, such as ancillary staff (staff working in the kitchen and cleaners), were excluded from the study.

The participation rate was 70% for the “Hjalli model” preschool teachers and 85% for the public preschool teachers [Table 2]. Of the teachers, 13 were males and 230 females. There were no significant differences in the age or length of teaching career between the two groups of teachers.

Table 2

Demographic information on teachers in the two types of preschool

Factor	Preschool category
	Public	“Hjalli model”
Number of teachers	135	108
Mean (range) age in years	37.8 (17-66)	35.5 (18-66)
Mean length (range) of the working career in years	8.4 (0.1-38)	8.6 (0.1-40)
Average number of children per teacher	21	10
Percentage (%) of preschool teachers with certification	59	41

Questionnaires

The questionnaire on the voice used was based on the ones used in previous studies.[52] The environmental part of the questionnaire was worked out in cooperation with the teachers because of their knowledge of their own working environment. Information was collected via questions on participants’ voice symptoms and their timing; stress, noise sources; and the ability of children to hear. The questionnaires were handed out to the schools’ head teachers. To ensure the best response rate, the participants were made to fill in the questionnaire independently at a staff meeting in the schools.

The frequency of voice symptoms (dryness, tickle in throat, feeling of lump in throat, hoarseness without cold, voice endurance, voice failures, voice loss, and voice fatigue while reading, singing, and in conversation) was examined with a 5-point scale: “hardly ever,” “seldom,” “sometimes,” “often,” and “almost always”. The recall period was 1 year. Teachers who reported having voice symptoms “often” or “almost always” were considered to have a voice disorder. The questionnaire also asked when the voice symptoms occurred: during weekends, in the morning, while teaching, in the evening, or during the summer, autumn, winter or/and spring (more than one answer was allowed). Alternatives for sick leave because of voice failure were “never,” “seldom,” “sometimes,” and “often.” For statistical analysis, the participants’ answers were converted into dichotomous variables where “almost always” or “often” were recoded into 1 and the other options into 0.

The rating scale for experiencing stress was from 1 (no stress) to 8 (extreme stress). Points from 6 to 8 were considered to indicate severe stress. Participants were also asked in what circumstances they experienced stress, plus an open question on what caused it.

To classify the teachers’ experience of noise, the teachers were made to answer four questions. First, what spaces in the preschools were loud (e.g., teachers’ coffee rooms, children's locker/changing rooms, or corridors)? Second, at what time was noise too loud (e.g., the mornings, the time when children came to school, when children had lunch, or when children played freely)? Third, how loud were the following noise sources: human beings (children and adults), toys (e.g., building blocks), building elements (e.g., doors), furniture (e.g., chairs and tables), equipment (e.g., fans), and/or external noise (e.g., from traffic)? The fourth question asked how reverberant different spaces were perceived to be.

There was a 5-point rating scale for assessing loudness and reverberation (from 1 = “far too high” to 5 = “almost none”). Replies 1 and 2 (“far too high” and “too high”) were interpreted as giving poor oral communication conditions. Two questions were used to obtain the teachers’ observations concerning the children's hearing: had they noticed hearing problems in children (yes/no)?; how often they had observed that pupils could not hear the teacher (often, sometimes, seldom, or never)? Additionally, the participants were asked open questions on how to lower noise and reverberation. For statistical analysis, the teachers’ answers were converted into dichotomous variables where “far too high” and “too high” were recoded into 1 and the other options into 0.

Activity noise measurement

To perform noise measurements, one school from each category was selected. The school selected from the “Hjalli model” group of preschools was situated in a new building (built in 2006, size 1,020 m2) with 32 teachers and 148 children. The public preschool selected was in a church hall (built in 1980, size 2,000 m2) that was not built as a preschool, with 27 teachers and 108 children. Both buildings were of concrete in accordance with Icelandic building regulations. Some renovations were conducted in the church hall so that it could meet the needs of preschool activity, for example, soundproofing. However, the shapes of the rooms was not altered and they remained irregular-sized and this in addition to the narrowness of the corridors may have affected the work environment. Measurements showed that reverberation time (RT) levels for both schools were under 0.5 s. (200-500-1,000-2,000-4,000 Hz).

The noise measurements were made in a) children's locker rooms (two identical rooms in Hjalli model of 31.5 m2 and one room in the public preschool of 53.5 m2 and b) the classrooms where the children were having their lunchtime and playtime (two identical rooms of 35.5 m2 each in the Hjalli model and three rooms (one 20 m2 and two 55 m2 in the public preschool). These were the spaces where half the teachers or more (48-80%), according to the questionnaires, had experienced loudness of the noise as high or far too high.

A total of 67 measurements were taken in the “Hjalli model” school: locker rooms-27 measurements; the space where children have lunch-24 measurements; and the space where children play-16 measurements. A total of 35 measurements in the public preschool: locker rooms-five measurements; the space where children have lunch-10 measurements; and the space where children play-20 measurements. Further, classrooms and locker rooms were compared with each other in more detail.

The sound level meter (Rion NL-05, Rion Science & Technology Shanghai Ldt, China) was placed on a shelf or in a safe place as close to the height of the children (about 1 m from the floor), as close as possible to the children's group and the microphone was directed toward the children. The meter was placed at a comparable place in all the rooms during the period of measurement. The measurement period was set at 15 min. The teacher controlled the start and end of the measurement and registered the number of children and adults attending the room and the activity during each measurement period.

The noise parameters measured for each 15-min period were the maximum sound level LAmax and the equivalent continuous sound level LAeq.

Data analysis

Descriptive statistics were used to describe small samples (playrooms and locker rooms). The Student's t-test, Chi-square test, and two proportions tests were used to compare the groups of teachers. Differences between the groups were taken as statistically significant with a P value of ≤0.05.

In order to see whether factors other than the school type affected the dependent variables, backward stepwise logistic regression analysis was used. Thus, stress, education, age, number of children per teacher, and working career in years were included in the analysis of all the dependent variables. Smoking was analyzed only in variables dealing with voice symptoms. The dependent variables were binary and so were the independent variables: school type, smoking, and education. In turn, stress, age, number of children per teacher, and working career in years were continuous. The logistic regression analysis was carried out only in those dependent variables that separated the two school types. Data were analyzed using the Statistical Package for the Social Sciences (SPSS 20.0 Inc., Chicago, Ill., USA).

Results

The highest response rate was >90% among both school categories for the voice symptoms and stress. The rate varied from 40% to 90% for the questions on noise. The response rate for the question on perceived reverberance in different rooms ranged from 56% to 81%.

Smoking and hearing

In “Hjalli model” preschools, there were a higher number of smokers (19.8%) than in the public schools (8.3%; P value 0.009). More public school teachers (11.3%) reported having problems with hearing compared with “Hjalli model” teachers (5.7%) but the difference was not significant.

Prevalence of voice disorders

The most frequent symptoms were throat symptoms such as the feeling of dryness or tickling in the throat followed by symptoms “voice does not last in noise” and “hoarseness without cold” [Figure 1]. Symptoms such as “dryness, tickle in the throat,” “feeling of lump in the throat,” “hoarseness without cold,” and “voice does not last in noise” were much more frequent among public school teachers than in the “Hjalli model” teachers. The proportion of teachers from the “Hjalli-model” preschools was lower in all aspects of reported voice symptoms except for “voice fatigue while singing.” No teacher from the “Hjalli model” preschools marked “almost always” or “often” in response to the question of whether they experienced “voice loss without having cold.”

Figure 1

Reported voice symptoms

According to logistic regression analysis, out of the voice symptoms of “feeling of a lump in the throat” and “hoarseness without a cold,” odds ratios of the school type were significantly lower than 1, meaning that the teachers from “Hjalli model” preschools were more unlikely to experience these symptoms “almost always” or “often” [Table 3]. “Teachers’ stress level” also had odds ratios that were significantly greater than 1 (in three out of four symptoms), meaning that the higher the stress level of the participants the more likely they were to experience voice symptoms “almost always” or “often.” In addition, the variables “teachers’ sick leave,” “working career in years,” and “age” were associated with voice symptoms. Additionally, of the “Hjalli model” preschool teachers, 94% had never taken sick leave owing to voice failure compared with 83% of the public school teachers. The difference was significant (P = 0.021, Chi-square test).

Table 3

Logistic regression analysis of variables associated with “almost always” or “often” responses from teachers regarding voice symptoms

Explanatory variables for voice symptoms	OR*	95% CI** for OR	P value	A	B
Dryness, tickle in the throat				0.783	0.326
Teachers’ stress level	1.611	1.289-2.013	<0.001
Working career in years	1.076	1.025-1.131	0.003
Sick leave	0.284	0.094-0.864	0.027
Constant	0.144		0.014
Feeling of lump in the throat				0.379	0.195
School type	0.258	0.096-0.695	0.007
Teachers’ stress level	1.416	1.099-1.825	0.007
Constant	0.082		<0.001
Hoarseness without cold				0.307	0.153
School type	0.367	0.146-0.925	0.033
Age	1.060	1.015-1.105	0.009
Constant	0.037		<0.001
Voice does not last in noise				0.636	0.319
Teachers’ stress level	1.761	1.336 - 2.323	<0.001
Sick leave	0.203	0.069 - 0.595	0.004
Constant	0.068		0.001

Backward Stepwise Logistic Regression, *OR: Odds ratio, **CI: Confidence interval, A = Hosmer and Lemeshow goodness-of-fit test, B = Negelkerke R Square, Variables entered in equation; school type (“Hjalli model” = 1), smoking (smoking = 1), teachers’ stress level (range 1-8), education (teachers with certification = 1), teachers’ sick leave (never = 1), age, number of children per teacher, working career in years

The most common periods during which teachers experienced voice symptoms were in the evenings, mornings, at the time of winter, and while teaching [Figure 2]. Half of the teachers from both the school categories reported experiencing voice symptoms while teaching. Teachers reported fewer voice symptoms during weekends and during spring and summer than at other times. Evenings and winters were the time periods when significantly more public school teachers than “Hjalli model” teachers reported voice symptoms [see Table 4]. In winter, teachers’ higher stress levels also increased the likelihood of suffering from problems in voice use (odds ratio significantly different from 1).

Figure 2

Periods in which teachers experienced voice symptoms

Table 4

Logistic regression analysis of variables associated with responses from teachers regarding the timing of voice symptoms

Explanatory variables for timing of voice symptoms	OR*	95% CI** for OR	P value	A	B
In the evenings				—	0.104
School type	0.298	0.149-0.598	0.001
Constant	2.529		0.001
During the winter				0.008	0.110
School type	0.407	0.206-0.805	0.010
Teachers’ stress level	1.223	1.021-1.464	0.029
Constant	0.793		0.599

Backward Stepwise Logistic Regression, *OR: Odds ratio, **CI: Confidence interval, A = Hosmer and Lemeshow goodness-of-fit test, B = Negelkerke R Square, Variables entered in equation; school type (“Hjalli model” = 1), smoking (smoking = 1), teachers’ stress level (range 1-8), education (teachers with certification = 1), teachers’ sick leave (never = 1), age, number of children per teacher, working career in years

Stress in participants

Significantly, more teachers (23.4%) in the public preschools experienced working with children as markedly stressful than teachers in the “Hjalli model” (13.8%) (P < 0.001, Chi-square test). According to the answers to the open-ended question, the most stressful situations were mainly when there was a shortage of staff and increased workload among teachers from “Hjalli model” schools, and noise and lack of discipline among teachers from the public schools. Teachers’ stress levels were also associated with experiencing voice symptoms more frequently [Table 3], experiencing children who were too noisy, and perceiving situations and occasions in the working environment as too noisy.

Loudness of noise sources and perceived reverberation

The proportion of the teachers from “Hjalli model” schools who experienced noise loudness as “far too high” or “too high” was lower than in the public schools for eight out of 10 items considered [Figure 3]. The biggest differences between the two preschool types were in the loudness (“very high”, “high”) of the noise from children (36%) and from building blocks (17%). The difference between school types for the item on how colleagues (staff/teachers) used their voices was close to significant (P = 0.051); however, the low response to this item decreased the reliability of this result.

Figure 3

The proportion of the teachers who experienced noise loudness as “too high” or “far too high”

Apart from school type, other factors were connected with the experience of finding noise sources loud. One of these factors especially affecting was age: noise sources were “too high” or “far too high” for older teachers (odd ratios greater than 1 in three out five items). For the variable “staff/teachers, e.g., how they use their voices,” qualified teachers were 3.6 times more likely to feel that the staff used their voices “too high” or “far too high” (odds ratio of 3.6). In addition, the variable “teachers’ stress level” explained the experience of finding noise from children and air conditioning “too high” or “far too high” (odds ratios significantly greater than 1).

Teachers from the “Hjalli model” preschools were generally less troubled by the loudness of noise than their counterparts in the public preschools [Figure 4]. The difference was not significant in the time periods when children came to or left school. Logistic regression analysis confirmed the above mentioned result (odd ratios for school type lower than 1). Logistic regression analysis also showed that especially during lunchtime and breaktime, noise was more often experienced to be far too high in public schools than in the “Hjalli model” schools. It was only during story time that the school type did not explain the experience of loudness; the explanatory factor was the teachers’ stress level. Other variables were also associated with noisy situations: stress and age with five variables, the number of children per one teacher with three variables, teachers’ education with two variables, and the length of the working career with one variable.

Figure 4

Time periods when the teachers experienced noise loudness to be “too high”

The proportion of teachers who considered the rooms to be “far too” or “too” reverberant was lower for all items in the “Hjalli model” preschools [Figure 5]. The only room where the difference was not significant was the teacher's coffee room. In the “Hjalli model” preschools, only a few teachers considered the rooms to be reverberant while over half of the teachers in public preschools considered rooms to be “far too” or “too” reverberant. Logistic regression analysis showed teachers’ ages and the length of the working careers also explained how reverberant the working places were experienced to be by the participants.

Figure 5

Proportion of teachers who experienced different rooms to be “too” or “far too” reverberant

Replies for the open-ended questions on how to lower the noise and how to reduce reverberation were similar in both types of preschool. The participants’ suggestions were: fewer children in a group, more discipline control, and reducing noise from furniture and furnishings, for example, by using mats on the floor and cushions under chair legs.

Hearing problems in children observed by teachers

The majority of the teachers did not observe children having difficulties in hearing speech: 73.7% from the “Hjalli model” and 66.9% from the public school teachers gave this answer; the difference was not significant. The “Hjalli model” preschool teachers were more rarely aware that their pupils could not hear them than the teachers from the public schools [Table 5]. The “Hjalli model” school personnel had somewhat frequently replied that they “often” found that a child had problems in hearing speech.

Table 5

Teachers’ knowledge of children having problems in hearing speech in the two types of preschool

Type of preschool	Teachers (n)	Proportion of observations (%)
		Often	Sometimes	Seldom	Never
Public preschools (6)	130	4.6	24.6	41.5	29.2
“Hjalli model” preschools (8)	101	6.9	12.9	33.7	46.5

Chi-square test P = 0.018

Activity noise levels

Noise levels were significantly lower in the “Hjalli model” school than in the public preschool [Table 6]. In the “Hjalli model” preschool, 51% of the equivalent noise level samples (LAeq15min) exceeded 70 dB(A) and 7% exceeded 75 dB(A), the highest equivalent noise level being 79 dB LAeq15min. In the public preschool, 89% of samples were above 70 dB (A) and 43% above 75 dB(A), and the highest activity noise level was 83 dB(A).

Table 6

Equivalent and maximum activity noise levels in one public preschool (35 measurements) and in one “Hjalli model” preschool (67 measurements). Each measurement lasted 15 min

Sound level	Public preschools (dB)	“Hjalli model” preschools (dB)	P value*
LAeq			<0.001
Mean (SD)	74 (4.5)	70 (3.6)
Min-max	62-83	59-79
LAmax			<0.001
Mean (SD)	91 (7.7)	85 (5.9)
Min-max	72-108	74-99

*Student’s t-test = Two-sample assuming equal variances

In the “Hjalli model” preschool, 9% of LAmax samples exceeded 95 dB (A) while the highest level was 99 dB LAmax. In the public preschool, 29% of the noise samples were over 95 dB LAmax and the highest was 108 dB LAmax.

Equivalent continuous and maximum noise levels when children were playing or were changing in the locker rooms are shown in Table 7. All the measured noise levels were higher in the public school than in the “Hjalli model” preschool. Equivalent sound levels did not exceed the 85 dB LAeq risk level for noise-induced hearing loss but in two measurements of the public preschool, the limit of 80 dB LAeq was exceeded.

Table 7

Means for equivalent continuous and maximum sound levels (range in parenthesis) in the classrooms and locker rooms in the two types of preschool

Location	Children (n)	Teachers (n)	LAeq (dB)	LAmax(dB)
Classroom
Public preschool	6-7	1-2	77 (70-83)	96 (72-108)
“Hjalli model” preschool	7-13	1-2	70 (67-76)	85 (77-96)
Locker rooms
Public preschool	20-30	5	73 (72-76)	88 (82-95)
“Hjalli model” preschool	15-24	2-3	70 (59-74)	85 (74-98)

Discipline, structure, and organization in activity seem to better predict the activity noise level than the number of children in the group. In the “Hjalli model” preschool, the noise levels were independent of the number of children in the group but in the public school, there was a significantly higher noise level in the play area with only a few (6-7) children, compared to the activity noise level in the locker room with many (20-30) children present.

Discussion

The main purpose of this research was to study how school pedagogical ideology with tighter discipline, less noisy toys, stricter rules, and more visual guidance for the children may lower the noise and hence, decrease the consequences of it. For that purpose, two schools applying different pedagogical ideologies were selected for noise surveys. Results indicated that educational and pedagogical ideologies are significant factors for predicting noise and at least part of its consequences.

Symptoms of voice disorder

The outcome of this study is in line with other research studies which have shown that preschool teachers have a high rate of voice symptoms[1516] and they are exposed to high noise levels in preschools.[244253] The results of this study also indicate that pedagogy has an effect on the prevalence and timing of voice symptoms as teachers from the “Hjalli model” reported significantly less voice symptoms, except for “while singing,” than the teachers in public schools. They also reported significantly less laryngeal symptoms than the public school teachers. These results can be interpreted as indicating less voice load in the private schools.

Timing of voice symptoms

The teachers experienced considerably fewer voice symptoms during the summer, during weekends, and in the spring. Undeniably, these time periods are when teachers spend less time indoors in the classrooms with their pupils. They might, therefore, be exposed to less noise originating from the children playing outdoors. It could also be assumed that poor indoor air quality may have some effects, the assumption of which gets support from study results that have shown that poor indoor air quality is correlated with teachers’ voice symptoms such as dry throat, pitch breaks, a lump in the throat,[23] and the occurrence of laryngitis.[25]

The “Hjalli model” teachers experienced fewer voice symptoms than the public school teachers during the winter that is the most heavily loaded working period and in the evenings that suggests less voice fatigue after a working day. Loud voice use in noisier classrooms during the day may be the reason why teachers in public preschools report more vocal symptoms in the evenings than those in the “Hjalli model” preschools. A greater voice load and more serious voice problems confirm the finding that considerably more teachers in the public schools had taken sick leave because of voice problems than teachers in “Hjalli model” schools.

Sick leave due to voice disorders

The results regarding the rate of sick leave did not confirm a general belief that teachers take sick leave very often because of voice symptoms (see, e.g., Van Houtte et al.,[54]). In this study, it was found that 94% of the “Hjalli model” preschool teachers and 83% of the public schools teachers had never taken sick leave because of voice failures. These rates suggest that teachers may not be aware of the serious consequences that voice disorders can have. The result is consistent with earlier findings according to which teachers usually have only little, if any, knowledge about voice disorders and more than half of teachers do not take sick leave even though they have obvious voice problems.[55] A handicap due to a voice disorder may manifest itself in other ways that are harmful for the working capacity.

Noise levels

Noise levels measured from the two preschools in this study were in line with other research studies that show high noise levels in these environments.[243456] Such levels are too high for children to concentrate and listen effectively,[8] especially in the context of the current standards (50-65 dB) that are applied in adults’ workplaces where concentration is required.[7]

Noise levels in two types of preschools

Measurements of noise levels show higher values for the public school than for the “Hjalli model” school. Former measurements on RT and activity noise in the two types of schools already existed and they were carried out by the Icelandic OHSW. These measurements showed that RTs were within the recommended limits or under 0.5 s in both the types of school. OHSW measurements had been carried out in the same rooms as noise measurements in the present study. However, since the two preschools were in different types of buildings, the results regarding noise may only partly be explained by the different types of pedagogic ideologies.

High noisiness was also evidenced by the replies given by the teachers from both schools but especially by those from the public school. Teachers from both schools identified that far too much noise came from the children themselves, which is in line with the findings of Shield and Dockrell.[34] Interestingly, around one-fifth of the “Hjalli model” teachers and one-third of the public school teachers found that annoying noise was coming from their colleagues. This could be due to poor working practices as too many teachers in the rooms try to raise their voices above the noise.

Perceived reverberance

Considerably more teachers from the public schools than from the “Hjalli model” schools perceived the rooms and spaces in the buildings to be too reverberant or far too reverberant, including the toilets. The considerable differences between teachers in their perceptions of reverberation in corridors (10% of “Hjalli model” schools teachers against 54% of the public school teachers) could also be explained by a feature of the “Hjalli model schools, where there are arrows and a midline on the floors, showing the children how and where to go, which could be interpreted as indirect discipline. This might have reduced activity noise in these spaces. From clinical practice it is known that young children need visual guidance. In addition, the finding that more public school teachers than “Hjalli model” teachers experienced that floors caused noise also supports the explanation of the usefulness of visual instructions.

Too much reverberation impairs speech intelligibility[57] that, in turn, may increase children's restlessness and further noisiness since children cannot hear what is said. This may also partly explain the finding why many more teachers from the public schools than from the “Hjalli model” schools reported work with children as stressful. Open comments from the public school teachers also give support to these interpretations as the teachers highlighted that noise and maintaining discipline were the most stressful issues in their work.

Noisy situations

The outcome of this study further suggests that there are certain situations when loudness of noise is experienced at the highest level. These circumstances are when the children are having lunch, entering/leaving the school building for playing outside, playing with building blocks, and when the children are in their locker rooms. Although these situations were the noisiest in all the preschools, the teachers from “Hjalli model” schools seemed to experience significantly less noise under such circumstances than the public school teachers. Analyzing these events and reorganizing the working practices will make it possible to reduce the noise.

At lunchtime, many children who require more or less help are together in the rooms with few teachers to help them. Cutlery, plates and other table requirements, and movable chairs also cause noise. In the teachers’ free comments, they pointed out the need of reducing noise from furniture and furnishings by using mats on the floors and cushions under chair legs.

Locker rooms are small in Icelandic preschools and are relative to the number of children who need to use them. In the public school where noise measurements were made, a 53.5 m2 L-shaped locker room contained 85 spaces, each 22.5 cm broad, where the individual child could keep his or her clothing. Two to four teachers were helping 20-30 children to change their clothing, either before going out to play or after coming back from play. In the “Hjalli model” schools, the locker rooms for the children's clothing were sized 31.5 m2 each with 53 spaces for the children's clothes. Two to three teachers were helping between 12-24 children in each locker room. These working conditions may easily explain the high noise levels.

Group size of children

The free comments of the teachers from both types of schools showed that most of them preferred teaching groups with fewer children. According to their replies, teachers from the public schools work together with an average of 20.8 children, double the number of children compared with the “Hjalli model” teachers who work more personally with fewer children (on an average 10.2).

Indeed, it was the teachers from the public schools who complained of noise that is in line with the finding that the noise increased with an increased number of children in the space.[34] Moreover, the public school teachers complained of the lack of discipline. Presumably, fewer the directions given in an activity, the louder the noise. The difference may indeed be a large one [20 dB(A)] between distinct activities.[34] The grouping of children could be partly responsible for the low noise levels in “Hjalli model” preschools where the children were never all in or out at the same time; they were always in small groups where one teacher was with his/her group(s) in a specific place. A fundamental tenet of “Hjalli model” ideology is that it is essential for children have the same teacher throughout the day in order to ensure successful education, discipline, and control.

Although the design of this study did not allow full statistical analysis of the relationship between the number of children and the perceived loudness of noise, the logistic regression analysis and noise measurements made in the two preschool types revealed that the noise levels might depend more on pedagogical practices than on group size. Notably, when measuring the noise levels in the playrooms the group size happened to be lesser in the public school (6-7 children) than in the “Hjalli model” school (7-13) but still, the noise level was higher in the public school compared with the “Hjalli model.”

Hearing problems in children

In this study, the hearing problems were not measured; instead, they were ascertained from the teachers. The results of the study showed that the majority of the teachers had not observed that children had difficulties in hearing speech in the preschool environment. However, based on the knowledge of speech perception in noise in children, one could suppose that more children should have had difficulties in hearing speech than the participants found in this study. A possible explanation is that the teachers presumed that the children heard better than they really did. In addition, difficulties in hearing may be due to sensorineural or conductive hearing losses. Conductive hearing losses in children are mainly due to otitis media. Children may also have temporary hearing losses due to exposure to noise (temporary threshold shifts) and permanent sensorineural hearing losses. Noise and reverberation may affect the mentioned two types of hearing losses in different ways.

If children have problems in hearing, they very probably increase the volume of their voices that, in turn, elevates noise levels. This serves as the basis of the finding that individuals increase or decrease pitch and loudness in accordance with the aural feedback they get.[282958] As far as is known, there is no research on whether or not children with a hearing loss are noisier than children with normal hearing; this issue requires investigation.

The effect of background variables

According to the logistic regression analysis, the school type did not always have the strongest effect on voice symptoms and on the experiences of perceiving the work environment as noisy. Their occurrence was also predicted by the following background variables: stress, education, age, and working career.

Apart from the school type, stress explained many of the analyzed variables. Indeed, stress has been found to change the acoustic features of the voice[3132] and increase the incidence of voice symptoms.[25] The relationship between stress and voice symptoms may also be reciprocal, as suggested by Dietrich et al.[59] and Rantala et al.[25] : the voice symptoms evoke anxiety or stress that, in turn, affect voice use. Noise may also be a linking element between stress and voice symptoms (see connection between noise and stress by Basner et al.[30]).

Another background variable connected, especially, with noise was education. The participants who were certificated as preschool teachers experienced noise to be too high more frequently than those who were not. Educated teachers may have received some information on the risks factors for voice disorders during their studies since voice problems are common among teachers. In addition, the teachers’ awareness of a long working career may have increased their concern about the risks. Awareness of the risks of voice disorders may help them to observe the possible threats in the school environment. The importance of the awareness is supported by the finding according to which people with no voice education reported the experiencing of less voice symptoms than those who had received short-term vocal training.[60]

Furthermore, age and working career were explanatory factors for voice symptoms. Indeed, participants between 40 years and 59 years of age have voice disorders most often.[13] Consequently, teachers with a long working career have probably more problems in voice use because of the cumulative effect of vocal loading and tissue damages due to biological aging.

Our results also showed that older teachers especially found noise to be too loud. Experiencing noise that was too loud and annoying may be connected with the phenomena of noise sensibility and hyperacusis. Sensibility to noise is also increased by fatigue[61] that, in turn, may be increased by aging. Furthermore, sensorineural hearing loss has a feature (hyperacusis); people with this disorder find loud sounds to be annoying as the synapses of the inner hair cells in their cochlea are damaged due to exposure to even quite low noise levels (84 dB) and even without recognizable and measurable hearing loss.[62]

The background variable smoking did not explain the voice symptoms in this study. A rather controversial finding was that there were more smokers among teachers of the “Hjalli model” preschool than among the teachers from the public preschool. Could it be possible that a quieter working environment has protected “Hjalli model” teachers from the risks of smoking? Even though smoking has generally been considered to cause risk of a voice disorder by provoking inflammation in mucous membranes as well as in the vocal folds,[63] the causality has not been confirmed in every study.[13] The bigger the vocal loading, the more likely it is for the voice symptoms to occur. Owing to the low number of smokers in this study, the result cannot be generalized.

Limitations of the research

The selection of schools for the activity noise measurements may include a confounding factor affecting the interpretation of the results. First, the acoustic environments of the schools may differ since the public preschool building was not originally intended as a preschool. Second, RTs of the rooms or other acoustic parameters were not measured in the present study. The former measurements carried out by the OHSW showing reverberation times below 0.5 s in both the types of schools were presupposed to be valid, as no renovation had taken place. Third, the size of the rooms were considerably larger in the public school than in the “Hjalli model” school, resulting in more individuals in the rooms at the same time and more distance between speakers and listeners. Owing to this, the results should be interpreted with caution.

The use of questionnaires as a measurement tool could also affect the results due to possible reporting bias. However, the use of subjective estimation in noise research is relatively common and according to Hilkhuysen et al.,[64] displays high correlations with physical metrics and this qualitative approach has been found to be a valid alternative to more traditional means of measurement.[65]

In this study, RT was not measured and instead the evaluations of reverberation were based on the participants’ auditory observations. The teachers may have confused noise with reverberation in their assessments. Although this confusion is typical in nonprofessionals, they may still evaluate the quality of acoustics quite well.[2] The finding that there was a large and significant difference between the two types of preschool regarding the perceived reverberation (i.e., if the room made a reverberant impression) may imply that the participants of this study could make rather reliable observations about the issue.

Practical implications

Without taking pedagogy into account, the average activity noise level in preschools exceeds the level recommended for concentration and speech communication in the current standards for adults. These levels prevent children from perceiving speech well, learning the language, and paying attention. Further, noise in preschools seems to be determined by the children's activity and behavior rather than the number of children in a group.

Thus, preschool teachers should be aware of the detrimental effects of noise on children and themselves. Further, information on hearing and what affects it in children should be disseminated among teachers. In addition, teachers should receive voice ergonomic education in teacher training and have pedagogic tools to lead activity toward quiet habits. Choosing less noisy toys and having groups with fewer children can lead to more positive and quieter learning environments by decreasing noise levels that in turn can improve children's focus and eagerness in the learning process.

Conclusion

This study indicated that pedagogical choices may affect the sound environment and teachers in preschools. The results indicate that there is a need to analyze the working practices and vocal habits of preschool teachers and children as well as to analyze the noise sources and their effect on communication.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.