Overall indoor quality of a non-renewed secondary-school building

Abstract

This work is the result of a field study about overall comfort aspects performed in a secondary-school building in the surroundings of Bologna during the winter season. The campaign aimed at describing overall comfort aspects of the school both from an objective and a subjective point of view, thus a questionnaire was administered to pupils during ongoing lessons to compare physical measurements and subjective responses. The monitored attributes and items concerned typical indoor quality aspects: acoustical, thermal, indoor air, and visual quality. As weak points emerged from the data analysis, possible solutions are illustrated, focusing on acoustics and optimizing effectiveness and cost.

Keywords

Educational building overall indoor quality acoustic characterization room acoustics building acoustics

Introduction

It is well known that a good indoor quality is a necessary requirement for an educational building.¹ Many studies have investigated indoor conditions of schools because of its close relation to students’ attention, productivity, and learning attitudes,² together with teachers’ health. The literature provides different kinds of studies: many studies focused on one or two environmental conditions,^3–6 while others were oriented to a holistic approach.⁷ For this study, the chosen perspective was to monitor all physical factors (acoustical, thermal, indoor air, and visual), focusing, in particular, on acoustics, with the aim of capturing the indoor aspects that could interfere with students’ attention. An attempt was made to weight the importance of each element asking students (13–14 years old) to complete a questionnaire about their level of satisfaction with the classroom indoor conditions: possible annoyance causes were investigated, as well as the way students interact with their environment to react to discomfort.

The school was built in the early 1970s; it has an articulated plan, spreads over three levels, and it is characterized by a concrete structure and ribbon windows shaded by projecting concrete elements. Given the modular structure of the building, classrooms have an almost constant size (floor area 58 m², volume 172 m³). This non-renewed secondary-school building is exploited as a case study: a repeatable measurement procedure is described and applied.

Investigation methods

Objective survey

Room acoustics (R.A.) measurements were performed in unoccupied and furnished classrooms in order to guarantee repeatability. The acoustic criteria were evaluated from impulse responses (IRs) measured using a maximum length sequence signal. All measurements were performed following the ISO 3382-2⁸ guidelines as a minimum target, and a statistical analysis was conducted for the acoustic criteria Early Decay Time, T₂₀ and C₅₀, in the octave frequency band from 250 to 4000 Hz. The IRs were measured using 3 sound source positions (teacher’s desk, center, and corner of the listening area; height 1.2 m) and 18 microphones positions (height 1.1 m) throughout the classroom. The values of reverberation time in occupied conditions were evaluated at a later stage accordingly to UNI EN 12354-6.⁹ Intelligibility measurements were performed using an NTI TalkBox loudspeaker that provided a human-like test signal with a sound power level of 60 dB(A),¹⁰ while an NTI XL2 analyzer evaluated the Speech Transmission Index (STI) parameter. The measurement setting comprised the sound source behind the teacher’s desk and six representative receiving points in the room. Building acoustics (B.A.) measurements included façade sound insulation and impact sound insulation and airborne sound insulation of a representative sample of building elements.¹¹ Environmental monitoring¹² took place during the heating period in five rooms, during morning classes: air temperature and relative humidity (RH) were measured with a DH206-2 Delta-Ohm data-logger with a time step of 1 min (accuracy: 0.3°C for temperatures; 2.5% for RH), put on a student’s desk in the center of the room (0.75 m above the floor). Measurements performed with the Thermal Microclimate HD32.1 completed the survey by measuring air temperature (t_air), globe thermometer temperature, RH, air velocity (v_air), CO₂ concentration, and desk illuminance (E_mean).

Subjective survey

A questionnaire was designed starting from the inherent literature and paying attention to the respondents’ target.^13–15 Some simple but effective principles were followed in order to improve response quality and reliability. Items were kept as simple as possible, all questions’ options were completely labeled and they were often supplemented by pictures and colors, trying to keep pupils motivated to answer. Before its administration, the questionnaire was first validated and then submitted to the teachers’ supervision. The final version contained 45 questions organized in five sections: the first was about general information and overall impressions about the indoor environment, while the remaining four covered acoustical, thermal, air quality, and visual attributes, respectively. The acoustical quality section investigated noise sources, their intensities, their frequency of occurrence, and their degree of annoyance. Besides pupils had to evaluate room reverberation and speech intelligibility.^7,16 The thermal quality paragraph concerned the perception of the thermal environment:¹⁷ a seven-point scale was used both for the thermal sensation¹⁸ and thermal preference, while a separate judgment was asked for acceptability and students’ reactions to discomfort. Air quality questions investigated whether the air was dry, dirty, characterized by any smell, and if the students are used to open windows to ensure air changes. At last the lighting section regarded the quality of light (both natural and artificial) on the blackboard, on the multimedia board, and on respondents’ desks. Students were asked to fill in the questionnaires thinking about the whole heating period and the conditions that they experience in their classroom during lessons. The sample size consisted of 105 13-year-old pupils, on average, with 52% of females and 48% of males.

Results and discussion

Acoustic measurements

The old Italian regulation¹⁹ requires for schools a mean reverberation time smaller than 1.2 s in the octave bands from 250 to 2000 Hz in unoccupied conditions, while the regulation²⁰ introduced, as design criteria, an optimal reverberation time that depends on the room volume and sound frequency as shown in equation (1). UNI 11367²¹ gives a further reference as a function of volume as well (see equation (2)); in each frequency band values should never exceed 1.2 × T_ott. Italian law does not suggest any other speech quality criteria

\begin{array}{l} T_{o t t} = k_{f} (- 0.2145 + 0.45 \log V) \\ k_{125} = 1.7, k_{250} = 1.4, k_{500} = 1.2, k_{1000} = 1.1, k_{2000} = 1.0, k_{4000} = 1.0 \end{array}

(1)

T_{o t t} = 0.32 \log V + 0.03 mean value at 500 - 1000 Hz

(2)

Useful but not mandatory indications are found in UNI 11367, where the speech clarity (C₅₀ > 0 dB) and the STI (STI > 0.6) are considered. Table 1 gives a summary of measured R.A. criteria and provides an estimation of the reverberation time in occupied conditions following EN 12354-6.⁹ All of the classrooms have the same volume of about 172 m³ and should require a mean unoccupied reverberation time of 0.87 s or 0.75 s following the two mentioned equations (1–2) at 1000 Hz, but it never happens. The limit values of reverberation time are not respected, other acoustic criteria are not satisfactory, and a non-uniform spatial distribution of energy criteria was found. Albeit measurements results pointed out a T_{250, 1000 Hz} = 1.35 s (σ = 0.12), pupils’ opinions collected with questionnaires are quite positives in this regard. Only 10% found the room too reverberant, 53% did not know how to judge this aspect, and the remaining believed the room not reverberant. Moreover, 83% of students assessed to be able to hear and understand well or very well the teacher’s voice, though measured STI value without noise was 0.51 (σ = 0.05). Measured values correspond to a situation between “poor” and “fair” conditions for speech intelligibility (class G according to UNI 11367²¹) even though a class D (STI > 0.62) may be ensured in classrooms.¹⁰ The C₅₀ values are unsatisfactory as well: values are below zero and uneven spatially distributed; at least positive values of clarity should be guaranteed. In authors’ opinion, pupils’ answers are due to the fact that students are not accustomed to the acoustic dictionary and concepts. Hence, in a further survey, it may be useful to undergo a sensory evaluation, and to ask students to describe their acoustic environment with individual elicited attributes, trying to correlate better students’ listening experience and measured criteria. Regarding building acoustic requirements, the DPCM 5/12/1997²² establishes limit values for school buildings and UNI 11367²¹ sets performance classes. It is worth noting that the latter complements the DPCM taking into account internal partitions too, as it should be expected dealing with learning spaces. The measurements’ results are summarized in Table 2, together with the limit values; the arithmetic mean is underlined when it does not comply with the DPCM.

Table 1.

Room acoustics criteria measured and estimated^a in all classrooms (mean value, standard deviation, and range).

Source	250 (Hz)	500 (Hz)	1000 (Hz)	2000 (Hz)	4000 (Hz)
T
T_{20, mean} (s) [σ] (range)	1.17 [0.13] (0.91; 1.48)	1.17 [0.13] (0.92; 1.42)	1.35 [0.12] (1.14; 1.59)	1.47 [0.12] (1.24; 1.74)	1.30 [0.09] (1.16; 1.53)
EDT _mean (s) [σ] (range)	1.10 [0.19] (0.62; 1.61)	1.16 [0.14] (0.84; 1.51)	1.34 [0.13] (1.07; 1.61)	1.47 [0.09] (1.23; 1.82)	1.30 [0.09] (1.12; 1.55)
C_{50, mean} (dB) [σ] (range)	−5.48 [2.72] (−12.90; 1.10)	−5.87 [2.85] (−12.94; 2.04)	−5.55 [1.95] (−10.53; −0.65)	−6.28 [1.81] (−9.85; −1.72)	−5.74 [1.76] (−8.58; −0.15)
STI_mean [σ]	Without noise	0.51 [0.05]	Class G	Poor–fair
T_{60, OCCUPIED} (s)^a	1.07	1.08	0.94	0.95	1.02

EDT: Early Decay Time; STI: Speech Transmission Index.

Table 2.

Building acoustics measurements and reference values.

	Measure #1	Measure #2	Measure #3	Mean	DPCM 5/12/97	UNI 11367
D_2m,nT,W (dB)	32	26	27	28	⩾48	⩾38	⩾43
${L^{'}}_{n, W}$ (dB)	71	68	69	69	⩽58	⩽63	⩽53
${R^{'}}_{W, FLOOR}$ (dB)	52	50	53	51	⩾50	⩾53	⩾63
${R^{'}}_{W, WALL}$ (dB)	44	43	45	44	⩾50	⩾53	⩾63

Possible acoustic interventions

Since acoustic performances are not satisfactory, some solutions should be envisaged. The solutions that are presented in the following, even if tailor-made for this case study, are relevant for classrooms and educational buildings of comparable typology. In order to validate the described interventions two three-dimensional (3D) simulation models were set up and tuned on the basis of the measurements performed: one in Namirial Acoustics©²³ for B.A. and the other one in Odeon©²⁴ for R.A.

The B.A. measurements together with in situ inspections revealed that requirements were not always met due to unsuitable or outdated constructive solutions. The sound insulation performances of party walls and of the façade are impaired by the partition–frame junction, by the false aluminum sheet pillars with an uneven filling of polystyrene beads, by the symmetrical double glaze windows (3 + 1 + 3) and by the discontinuity of the panels of extruded polystyrene foam. Besides, floor has adequate mass but lacks in a resilient layer. Hence, it was considered appropriate to install on party walls a soundproof lining (0.95 cm plasterboard + 3 cm rock wool panel, about 10 kg/m²) and on floors a screed of 4 cm on a layer of resilient material (about 77 kg/m² overall). A complete replacement of fixtures would also be helpful to reduce acoustic bridges and coincidence phenomena. These solutions were all included in the numerical model of the whole building and they were all found useful.

R.A. measurements revealed excessive reverberation times and poor clarity and speech intelligibility, as expected, due to the lack of absorbing surfaces inside the room. Considering the classrooms’ layout, it is only possible to act on the ceiling and on some upper portion of walls. In order to understand which sound-absorbing material to choose and which is its optimal distribution, several numerical simulations were compared. Dealing with R.A. quality, the target is: on one hand to shorten the reverberation time (T₂₀), on the other to ensure a minimum speech clarity (C₅₀) and to maximize the useful-to-detrimental sound ratio (U₅₀).²⁵ Trying to pan the most common acoustic paneling, three solutions were chosen: (#1) polyester fiber panels, 5 cm thick, (#2) ceiling plasterboard, and (#3) acoustic baffles. A pre-sizing of the absorbing material required to get the target reverberation time values showed that solution #2 would need the whole ceiling area, while both solutions #1 and #3 require about an half of the ceiling surface. Thus, a trial of different configurations for the solution #1 was performed following literature references,²⁶ and the results were compared exploiting numerical simulation with Odeon©.²⁴ Table 3 shows the simulated configurations (abbreviations in brackets): full ceiling covered (#1-FC), half-front ceiling (#1-HFC), half-back ceiling (#1-HBC), ring (#1-R), anti-ring (#1-AR), partial ring and upper walls (#1-RUW), and half-back ceiling and upper walls (#1-HBUW), #2, and #3. The 3D model of the classroom was first tuned taking measured values as reference; then, the above configurations were tested. In Figure 1, the plotted values refer to the same receiver positions of the measuring campaign, and the sound source was located at position T, behind the teacher’s desk. Spatial averaged and mean values of acoustic criteria are plotted for each of the 10 configurations (see Figure 1). It is worth mentioning that both an ambient noise spectrum with a total power level of about 40 dB(A) and a source power level of about 66 dB(A) were included into the model, according to the spectrum suggested in ISO 9921.²⁷ Plots clearly show that adding more absorbing material systematically decreases the reverberation time and increases C₅₀ values, compared to the actual condition (indicated with NO INT in Figure 1). However, the addition of absorbing material to the ceiling also causes a decrease of calculated sound levels leading to a decreased speech intelligibility. Which means that adding sound absorption has both beneficial and detrimental effects and a compromise is necessary: the optimum may correspond to high SPL and U₅₀ and to clarity and reverberation time values that comply with the UNI 11367²¹ target.

Table 3.

Sketches of simulated configurations.

FC: full ceiling covered; HFC: half-front ceiling; HBC: half-back ceiling; R: ring; AR: anti-ring; RUW: partial ring and upper walls; HBUW: half-back ceiling and upper walls.

Figure 1.

Comparison of some acoustic criteria obtained from numerical simulations; plot of the mean of the spatial averaged results.

Although reverberation time is not expected to vary abruptly among the configurations for the material #1 (FC excluded), C₅₀ values are affected by the location of the absorbing material. Thus, it is possible to improve the conditions for speech by optimally locating the available absorbing material. Moreover, the condition that optimize both C₅₀ and speech-to-noise ratio can be determined by finding the configuration that corresponds to the maximum useful-to-detrimental ratio U₅₀.²⁵

It clearly appears that the #1-FC configuration causes a remarkable loss of SPL, and even if both C₅₀ and U₅₀ are good this solution require an excessive and unnecessary quantity of absorbing material. In the remaining #1 configuration it is interesting to notice that, even if the material amount is the same, both clarity and U₅₀ values vary up to 2 dB. The best compromise may be found for the #1-R configuration because of its higher values of clarity and useful-to-detrimental ratio.

Solutions #2 and #3 are interesting too: they guarantee comparable reverberation times, but the latter shows reduced speech-related performances.

All things considered that the most suitable solutions are #1-R and #2, but the latter (ceiling plasterboard) is much more durable and requires little maintenance, which are not negligible aspects in schools.

The best solutions analyzed and their costs are shown in Table 4. The costs of B.A. interventions would be about 8 800 € per classroom, instead R.A. solutions range from about 2200 to 6000 € per classroom. To rank the interventions priority, the authors decided to take into account pupils’ annoyance judgments concerning the most recurring noise sources (i.e. students in the room (SR), students in neighboring rooms (SNR), students in aisles (SA), and external road traffic (ERT)). Results are plotted in Figure 2. The higher scores are those of SR and SA because they are considered louder (52%, 35%) and annoying (22%, 22%) according to the largest amount of students. However, it is interesting to point out that students’ perception of annoyance is much more correlated with the frequency of occurrence of noises rather than with their intensity. Overall, it may be advisable at least to implement the current situation with party walls’ lining, new ribbon windows, and plasterboard ceiling lining, in order to reach reasonable performances and to take into account the main pupils’ claims.

Table 4.

Costs of the acoustic improvement solutions for each classroom.

Type	Improvements	Required quantity	Unit cost	Total amount
B.A.	Party walls’ lining	71m²	36.95€/m²	2 553€
B.A.	Floating floor	2.6m³	371.71€/m³	967€
B.A.	Windows	22.4m²	113€/m²	2 531€
B.A.	Frames		123€/m²	2755€
R.A.	Ceiling lining—#1-R	32m²	68€/m²	2176€
R.A.	Ceiling lining—#2	58m²	97.8€/m²	5 673€
R.A.	Baffles—#3	20m²	298€/m²	5 960€

B.A.: building acoustics; R.A.: room acoustics.

Figure 2.

Pupils’ answers: noise sources loudness (blue), frequency of occurrence (red), and annoyance (green).

Environmental measurements

The environmental monitoring took place in the heating season during morning classes; some results are resumed in Table 5. The microclimatic parameters were analyzed in order to evaluate Fanger’s indices,²⁸ and the results were compared to pupils’ impressions. The metabolic rate was set at 1.2 met due to students’ sedentary activity, while clothing insulation values were derived from questionnaires (mean value 1.05 clo). The calculated Predicted Mean Vote (PMV) values showed almost neutral mean values in all the classrooms, but the extremes determined different categories:²⁹ category II for classrooms 3D and 3E, category III for 3A, and category IV for 3B and 3C. Figure 3 shows the distribution of both actual thermal votes (Actual Mean Vote (AMV)), preferences, and calculated PMV, sided by acceptability judgments. AMVs tend to be distributed slightly more on the cold side, while in this survey PMVs overestimate thermal votes. As expected, the preference votes are almost symmetrically distributed if compared to AMVs, leading to a preferred neutral environment. It is interesting to focus on the wide acceptability range that emerged from the survey; claims were mainly expressed by students seated near windows and door. Up to 36% of pupils considered acceptable a cool environment (−2), 60% a slightly cool one (−1), 80% a slightly warm room (+1), and 50% a warm one (+2).

Table 5.

Indoor conditions during the day of microclimatic monitoring.

	3A	3B	Rooms	3D	3E
	3A	3B	3C	3D	3E
Location and orientation	First floor; NW	Second floor; NE	First floor; SW	Second floor; NW	Ground floor; SE
Number of students [boys, girls]	21 [10; 11]	22 [14; 8]	22 [9; 13]	19 [9; 10]	21 [8; 13]
t_air (°C)	20.6	19.6	20.2	20.4	21.3
RH (%)	53	55	49	46	50
t_op (°C)	20.6	19.5	20.1	20.4	21.4
v_air (m/s)	0.01	0.01	0.01	0.01	0.01
PMV_mean [range]	−0.5 [−0.69; 0]	−0.33 [−1.04; 0.35]	−0.33 [−0.84; −0.13]	−0.13 [−0.34; 0.5]	0.06 [−0.5; 0.15]
AMV (questionnaires)	−0.14	−0.55	−0.41	−0.68	−0.86
E_mean (lux)	241	231	627	535	540
CO₂ (ppm)	1936	n.d.	1480	1543	1972

RH: relative humidity.

Figure 3.

Pupils’ answers about thermal environment (AMV) compared with their preferences (PREF) and measured PMV.

CO₂ concentration was found considerably high, and only in two out of five classrooms the measured values did not exceed the basic requirements of 1500 ppm, but it is to say that windows were never opened during the measurement session, probably due to cold outdoor temperatures. Questionnaire responses followed the measurements trend: only 32% of the population is satisfied of air quality, 64% judged the air as bad and heavy, but at the same time almost all respondents affirmed to react actively to such discomfort opening the windows.

Illuminance requirements (⩾300 lux) were satisfied only in three out of five classrooms with the artificial light system switched off (it usually supplies these lacks and the non-homogeneous illuminance of rooms). This may be a cause of the large amount (30%) of unsatisfied students, comparable to the amount of those who complained about the thermal environment.

Overall comfort: students’ perception

The subjective approach aimed at finding out correlations between pupils’ perception and aspects that may be critical relying solely on physical measurements. Hence, students were asked how they feel in their classroom and to rank the four quality aspects taking into account their influence on the overall satisfaction. Moreover, each section of the survey ended with two questions about the satisfaction for each single aspect and if it affects learning activities. Purposely the word “comfort” was never used to avoid misleading answers or interpretations. The largest part of students (68%) said to feel “good” or “very good” in classroom, 31% answered with a neutral vote, and only 4% of them answered “bad” or “very bad.” The most voted aspect, that is, ranked as the most important, was the acoustic one (30% of votes), followed by the thermal (21%), the air quality (19%), and the visual (17%) aspect, as shown in Figure 4(a). What appears more interesting is the correlation of satisfaction and influence votes with overall satisfaction judgments, plotted in Figure 4(b). It clearly emerges that pupils’ votes are strongly correlated with the influence votes of both acoustic and visual aspects, while satisfaction votes have a different trend. The correlation with acoustic votes has to be linked to noise aspects because of collected responses about reverberation.

Figure 4.

(a) Overall satisfaction: pupils’ ranking of attributes and (b) votes of satisfaction and influence on learning performance.

Possible interventions for overall comfort

On the whole, students’ main environmental complains are about noise (55%), drafts (69%), and thermal asymmetry caused by cold surfaces (59%) and by the sun entering through windows (60%). Nevertheless, they are not used to use external curtains to shade their classrooms. Thus, a useful intervention may be used to replace frames to prevent draft but also to limit the solar radiation by installing solar control glasses (with low g-value). This solution would be useful also from the acoustic point of view; other possible interventions to improve acoustic quality of rooms are largely illustrated in section “Possible acoustic interventions.” A more expensive solution would be to install external movable shadings, mechanically oriented to grant the proper illuminance inside each classroom.

Besides, installing lighting control-integrated system of natural and artificial lighting in classrooms would guarantee saving energy and a greater comfort.

Taking into account the air quality it may be possible to install CO₂-monitoring devices with audible or visual indicators so that users can ventilate the room until the desirable level of CO₂. The introduction of a multi-zone mechanical ventilation system with heat recovery units and RH- and CO₂-monitoring devices would be a further solution, but cost and payback time should be carefully evaluated.

Conclusion

The overall indoor quality in educational buildings is an important issue. A measurement campaign performed in a non-renewed secondary-school building is exploited as case study. Measurement methods are outlined, and results are discussed both from objective and subjective point of view. The building acoustics requirements are affected by outdated constructive solutions, classrooms were found too reverberant, lacking in clarity and with a low performance in supporting oral communication due to the acoustically untreated surfaces. CO₂ concentration was found too high because air changes depend on users only, and natural light alone does not comply with illuminance requirements. Questionnaires revealed that the acoustic and visual aspects play a key role on students’ perception of their learning spaces. This is the typical outline of a non-renewed educational building. In this framework, acoustics is considered a noteworthy aspect that is often neglected both in the design and retrofit stages. Thus, improving solutions have been illustrated, comparing the alternatives on the basis of numerical simulations and a cost analysis. Albeit the suggested solutions may appear quite expensive, they would not in a perspective of whole building refurbishment. As a matter of facts, the school may need further interventions to get to reasonable energy efficiency and comfort standards. Thus, some environmental possible interventions are envisaged too.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

De Giuli

Pontarollo

De Carli

. Overall assessment of indoor conditions in a school building: an Italian case study. Int J Environ Res 2014; 8(1): 27–38.

Dockrell

Shield

. Acoustical barriers in classrooms: the impact of noise on performance in the classroom. Brit Educ Res J 2006; 32(3): 509–525.

Mydlarz

Conetta

Connolly

. Comparison of environmental and acoustic factors in occupied school classrooms for 11–16 year old students. Build Environ 2013; 460: 265–271.

Marinosci

Morini

Semprini

. Preliminary energy audit of the historical building of the School of Engineering and Architecture of Bologna. Energy Proced 2015; 81: 64–73.

Biserni

Garai

. First and second law analysis applied to building envelope: a theoretical approach on the potentiality of Bejan’s theory. Energy Reports 2015; 1: 181–183.

Valdiserri

Biserni

Tosi

. Retrofit strategies applied to a tertiary building assisted by Trnsys energy simulation tool. Energy Proced 2015; 78: 765–770.

Astolfi

Pellerey

. Subjective and objective assessment of acoustical and overall environmental quality in secondary school classrooms. J Acoust Soc Am 2008; 123(1): 163–173.

ISO 3382-2. Acoustics—measurement of room acoustic parameters. Geneva: International Organization for Standardization, 2008.

EN 12354-6. Building acoustics—estimation of the acoustic performance of buildings from the performance of elements—part 6: sound absorption in enclosed spaces. Brussels: Comité Européen de Normalization, 2003.

10.

IEC 60268-16. Sound system equipment—part 16: objective rating of speech intelligibility by speech transmission index. Geneva: International Electrotechnical Commission, 2011.

11.

ISO 16283. Acoustics—field measurement of sound insulation in buildings and of building elements. Geneva: International Organization for Standardization, 2014.

12.

ISO 7726. Ergonomics of the thermal environment—instruments for measuring physical quantities. Geneva: International Organization for Standardization, 1998.

13.

Borgers

Hox

Sikkel

. Response effects in surveys on children and adolescents: the effect of number of response options, negative wording, and neutral mid-point. Qual Quant 2004; 38: 17–33.

14.

De Leeuw

. Reducing missing data in surveys: an overview of methods. Qual Quant 2001; 35: 147–160.

15.

Haddad

King

Osmond

. Questionnaire design to determine children’s thermal sensation, preference and acceptability in the classroom. In: 28th PLEA conference, Lima, Peru, 7–9 November 2012.

16.

ISO/TS 15666. Acoustics—assessment of noise annoyance by means of social and socio-acoustic surveys. Geneva: International Organization for Standardization, 2003.

17.

ISO 10551. Ergonomics of the thermal environment—assessment of the influence of the thermal environment using subjective judgment scales. Geneva: International Organization for Standardization, 2001.

18.

ASHRAE Standard 55-2013. Thermal environmental conditions for human occupancy. Atlanta, GA: American Society of Heating, Refrigeration and Air Conditioning Engineers, 2013.

19.

Circ. Min. n. 3150 22/5/1967. Criteri di valutazione e collaudo dei requisiti acustici negli edifici scolastici.

20.

D.M. 18/12/1975. Norme tecniche aggiornate relative all’edilizia scolastica, ivi compresi gli indici minimi di funzionalità didattica, edilizia ed urbanistica nella esecuzione di opere di edilizia scolastica.

21.

UNI 11367. Acustica in edilizia—Classificazione acustica delle unità immobiliari, Procedura di valutazione e verifica in opera. Milano: Ente Nazionale Italiano di Unificazione, 2010.

22.

DPCM 5/12/1997. Determinazione dei requisiti acustici passivi degli edifici.

23.

Namirial Acoustics. Ancona: NAMIRIAL S.p.A, http://www.microsoftware.it/software-acustica.asp

24.

Christensen

. ODEON room acoustics software—version 12, industrial, auditorium and combined editions. Lyngby: ODEON, 2013.

25.

Bradley

. Predictors of speech intelligibility in rooms. J Acoust Soc Am 1986; 80(3): 837–845.

26.

Bradley

. Optimizing sound quality for classrooms. Proceedings of XX Meeting of SOBRAC, Rio de Janeiro, 21–24 October 2001.

27.

ISO 9921. Ergonomics—assessment for speech communication. Geneva: International Organization for Standardization, 2003.

28.

ISO 7730. Ergonomics of the thermal environment—analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria. Geneva: International Organization for Standardization, 2005.

29.

EN 15251. Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality, thermal environment, lighting and acoustics. Brussels: European Committee for Standardization, 2007.