Interaction between annoyance, indoor noise levels and acoustic

classification of buildings Selma Kurra 1 Professor Dr. (rtd) Istanbul Technical University, Consultant to dB-KES Engineering Krizantem Sk. no: 70, 34330 Levent, Istanbul, Turkey

Ayça Sentop Dumen 2 Dr., Istanbul Bilgi University Kazim Karabekir Cad. No: 2/13, 34060 Eyup, Istanbul, Turkey

ABSTRACT Assessment of annoyance “at home” from environmental noises has been widely investigated so far and the ISO/TS 15666:2021 was developed to lead the socio-acoustic surveys. On the other hand, the rating of buildings’ acoustical performance considering the indoor noise levels, has also been well researched in building acoustics and the studies have ended up with the ISO/TS 19488:2021 covering the acoustic classification system for buildings. Basically the rating system needs to be supported by the subjective tests in the field or in laboratories, to acquire data about the annoyance/disturbance or satisfaction of residents. If the target is to design the healthy, comfortable and sustainable acoustical environment, both technical standards might be harmonized in the future, particularly with respect to survey methodologies and evaluations. In this paper, based on the dose/response relationships with respect to the indoor noise levels, an approach is proposed to translate the acoustic classes proposed in ISO/TS 19488, into the annoyance boundaries in terms of different scales (verbal/numerical and the HA% ) referred in ISO/TS 15666. The results from the previous laboratory and the field studies conducted by the authors, were used for verification of this approach.

1. INTRODUCTION

Investigation of subjective reactions against environmental noises has long being continued to derive the typical dose-response curves based on the wide range of surveys. On the other hand, the same depth of concern has been made on the subjective evaluations in buildings, i.e. satisfaction or dissatisfaction from indoor acoustic conditions and sound insulation. If the ultimate goal, is to attain a healthy, comfortable and sustainable environment both indoors and outdoors (i.e. open and confined environments where they hear, percept, like/dislike, get disturbed or satisfied from sounds to which they are exposed in daily life) it is important to find a way of collaboration (harmonization) between both area of research and practice. The main platform of collaboration can be constructed in common methodology and techniques in socio-acoustic surveys to be performed in the fields of environment and building acoustics e.g. data collection, questionnaire development, annoyance scales, acoustic measurements, data analysis and evaluations.

The two international technical standards should be referred namely ISO/TS 15666:2021 involving the assessment of annoyance from environmental noises and ISO/TS 19488:2021 concerning the

1 selmakurra@gmail.com 2 ayca.sentop@bilgi.edu.tr

worm 2022

acoustic evaluation in buildings with a classification system [1,2]. Certainly, evaluation and rating of buildings’ acoustic performance need to be supported by the physical and subjective tests either in the field or in the laboratory to acquire sufficient data regarding people’s reaction against the noise related problems.

This paper aiming to contribute to the above target, presents an approach to determine the relations with the annoyance descriptors, indoor noise levels and the acoustic classes.

2. BRIEF REVIEW OF LITERATURE (STUDIES) RELATED TO THE TOPIC

An outline of the status of research &development on the study topic is given below.

2.1. Assessment of annoyance from environmental noises Subjective evaluation of environmental noises, is based on the “dose-response relationships” derived from the field surveys for transportation noises which have always been the global noise adversely affecting the people, and for industrial facilities, wind turbines, neighbors, amusement centers and etc. The importance of dose-response relationships, have been well understood in noise management during design, planning and noise control. The noise data has been presented simply by L Aeq (time) dB in the past and by the time-weighted standard descriptors L dn or L den at present. The subjective responses against noise in the earlier field and experimental studies, have been measured by using a psychometric scale, in terms of the group annoyance scores or the percentage of annoyed people and, clustered in noise levels. Currently the percentage of highly annoyed people (HA%), which is a single number value, has been a common descriptor much suitable to compare the results of different surveys. It was also evidenced as the best correlated descriptor with the noise exposure levels. HA% is determined from the individual annoyance scores according to the selected annoyance scale, distributed over the noise levels measured outdoors. Environmental noises are categorized as 45-49, 50-54, 55-59, 60-64, 65-69, 70-74, > 75 according to END [3]. A comprehensive review of these studies is given in [4].

The noise/response data obtained from the worldwide surveys by numbers of researchers [4-8], were first gathered, analyzed and discussed by Schultz (1978) and the regression functions were derived from the individual response data for various noise sources [5-7]. The systematic reviews and meta-analyses of environmental noise data was also performed by WHO to acquire the best final curve fitting through the regression models and placed on the EU documents [8, 9]. Generally, it is proved that statistical linear regressions give better correlation when the group average data is employed and the logistic or polynomial regression functions yield higher regression coefficient for the HA%.

Fidell et al. and Schomer et al., introduced a new descriptor of annoyance, so called “prevalence of highly annoyed population, P HA %” based on their theory that the annoyance is proportional to loudness times duration [7,10,11]. They proposed a variable indicating the position of their “duration adjusted loudness function”, along the L dn axis in the dose- response diagrams. Prediction of P HA % with the 95% prediction interval, has been also explained in ISO 1996-1:2016 [12].

Variation of annoyance degrees according to type of noise source and with specific parameters has been widely investigated and nowadays, the significance of the non-acoustic factors is emphasized to explain the vast variation of the results from different surveys. This rather complex issue might be acknowledged to the standard in the future.

The methodology to be implemented in the field surveys for environmental noise assessments was presented in ISO/TS 15666: 2021 (first issued in 2003) [1]. The two shared annoyance questions; verbal or 5-point (1-5) and numerical 11-point (0-10) along with the wording in different languages, are recommended by the ICBEN team (Guski et al) for reporting of residents’ reactions to noise on combined social and acoustical surveys [13-15].

The conversion between the annoyance rating scales in order to determine the HA%, was concerned by various researchers to align the cut-off points indicating high annoyance when the different annoyance scales are employed in surveys [16-20]. Discussion has ended up by accepting

the weight of 0, 4 to be applied to the “score 4” on the 5-point scale to provide the same cut-off 28% as on the 11-point scale (the upper part of the scale), in the revised ISO/TS 15666:2021[2].

2.2. Assessment of annoyance in buildings from noise Subjective evaluations in buildings are performed with various purposes mainly satisfaction from indoor acoustics and sound insulation. Some studies published so far, investigated on the annoyance from airborne and impact sounds at home, or from specific noises, like footfall, music or living sounds coming from the neighbors transferred by the partitions.Some of them presented their results as those from the environmental noises: i.e. dose- response curves relating the annoyed or highly annoyed people and the measured indoor noise levels [21-29]. Rindel gave an estimation about the 4 %/dB increase of HA% with the indoor noise levels

It is possible to apply dose-response relationships comprising with the indoor conditions, however, a simple conversion of noise levels from outdoor to indoor is difficult due to the differences in facade constructions and in spectral change of the noise characteristics transferred to indoors. Some studies have stressed the importance of low frequency reduction of the facades. Others have revealed the median differences as 16, 13, and 22 dB(A) for open, tilted, and closed windows respectively, as high as 31 dBA for the closed windows [30,31]. The classification scheme for the indoor levels in L den , has been proposed based on the large scale experimental data. However, L den indoor, can not be correlated with the insulation descriptors due to the difference in the frequency ranges.

Experimental annoyance studies in simulated environments (i.e. specific laboratories) were conducted widely in the past and mostly focused on sleep disturbance due to noise. Some studied general annoyance from traffic noise or annoyance during different activities at home. The outcome of such studies were generally presented in the form of the observed effects relative to noise levels [32-36] similar to dose-response functions.

On the other hand, the listening tests have been conducted for comparison of the loudness responses against different noises heard/perceived and to evaluate the insulation efficiencies [37]. However it is difficult to judge about annoyance in listening rooms with short samples of sounds, since annoyance is highly dependent on exposure duration, as well as the numerous other factors. Some good reviews on these studies can be found in literature [38].

The ISO/TS 19488: 2021 that was developed after the EU/COST Action TU 0901 project, presents an acoustic classification scheme according to a certain criteria for the airborne and impact sound insulations, and the indoor noise levels originated by mechanical service equipment [56]. The standard specifies the criteria for six classes, from A to F for dwellings. (A is the highest and F is the lowest class). The categories are described, e.g. C implies “protection against significant disturbance given normal behavior of neighbors who are considerate of other occupants” and D corresponds to “the disturbance more than occasionally”. The standard does not declare criteria for surveys and neither recommends a methodology for annoyance surveys. However, during the project referred above, a questionnaire template has been prepared by employing the verbal and numerical scales recommended in ISO/TS 15666:2021 [39]. Various questions are asked in addition to subjects’ annoyance from neighbor noises and outdoor noises from facades, such as regarding personal, social and building related aspects. However this questionnaire has not been standardized so far. Although the acoustic categorization in the standard, refrains from the inclusion of indoor noise levels, the maximum acceptable indoor noise levels, so called background noise levels, have always been paid attention in noise control in buildings and the noise limits have been presented according to the space and building usage, by WHO, OECD, ICBEN as well as in numbers of the national regulations. [40,41].

3. METHODOLOGY TO ESTABLISH A CORRELATION BETWEEN THE

ANNOYANCE SCALES, INDOOR LEVELS AND ACOUSTIC CATEGORIZATION

In this study, the relation between the indoor noise levels and annoyance responses was investigated with respect of the acoustic classification scheme proposed in ISO/TS 19488. The categorization of indoor noise levels corresponding to the annoyance degrees in different scales was

made by conversion and regressions. Since there is not much available experimental data in this respect, the results of the earlier laboratory study conducted by the first author et al., was utilized in the analyses. Additionally, the result of the field study involving the annoyance from noises in residential buildings, which was performed by the co-author, was introduced for verification of the model if possible and to evaluate both outcomes in combination.

The steps of the study with the outcomes, are explained below:

3.1. Outline of the earlier laboratory study and results: New data analysis The earlier simulated experimental study conducted in Japan aimed to determine the annoyance from transportation noises at home, the variation of annoyance from different sources (i.e. road, railway, aircraft with different number of pass-bys) and with activity type, (i.e. reading and listening) in the simulated living room under the controlled noise conditions along with the visual simulation [35,36]. The actual recorded signals were processed by a transfer function considering the effects of façade and room acoustics. six levels of indoor noise between 30-55 dB (A) with 5 dB intervals were randomly given into the room for subjects’ evaluation in 30 minutes of duration. The number of sessions was 48 and the number of subjects was 192 distributed almost evenly according to the noise levels. The 7- point annoyance scale was used in the questionnaire that was prepared by keeping the similar structure as in the field studies at that time. The results were presented as the individual responses at each score, the group average values and the percentage of highly annoyed subjects (HA%). The overall annoyance and the source-specific annoyances were determined separately and in combination. Figure 1 shows the regression line between the indoor noise levels and the group average values using 7-point annoyance scale for the 3 types of transportation noise sources. The regression coefficient (R 2 ) is 0.85. The calculated annoyance scores from the linear regression equation at each noise category, are shown on the diagram. The slope of the regression line, is compatible with the results of other experimental studies which give slightly higher annoyance than the field studies, as proved in literature.

3.2. Steps of the study The procedure applied in the study is explained below. a. Conversion between the annoyance scales

The common annoyance rating scales used in socio-acoustic surveys in the past and present are 1- 3, 1-4, 0-7, 1-7, 1-9, 1-5 and 0-10 point scales, of which the last two have been standardized in ISO/TS 15666 . These are Likert type scales with the scores are positioned with equal intervals on the total scale length (i.e. equal intervals between each choice in the questionnaire) [42]. T he values (scores) on the scales are interpreted as discrete points, i.e. ordered-categorical data (i.e. ordinal data), not the interval data.

Overall annoyance & noise levels

7

Group average annoyance scores (1-7)

6,8

6

6,1

5,39

5

4,67

4

3,95

3

3,24

y = 0,1434x - 2,4957

2,52

R² = 0,8534

2

1,8

1

1,08

20 25 30 35 40 45 50 55 60 65 70

Indoor sound presure level, L Aeq, dB (30 min)

Figure 1: The dose-response relationship derived from [35,36]. The source-specific group averages using the 7-point annoyance scale are clustered on the indoor noise levels ranging 35-55 dB (A) (Three dots at each noise level, indicate the group averages for three types of the noise source).

Translation of the annoyance rating scales from one to the other, which is necessary to combine and evaluate the survey results, was made in this study by applying the linear interpolation on the annoyance scores. Figure 2 gives the graphical form to compare the 1-5 point and 0-10 point scales, however adjusted to 0 (as the initial score) for the 5 point scale. Table 1 gives the converted scores.

Figure 2: Conversion between 5-point and 11-point scales.

Tab le1: Corresponding scores on 5 and 11 point scales.

numerical verbal verbal numerical 11-point 5-point 5-point 11-point 10 4 4 10 9 3.6 3 7.5 8 3.2 2 5 7 2.8 1 2.5 6 2.4 0 0 5 2 4 1.6 3 1.2 2 0.8 1 0.4 0 0 b. Correlation between scales and classes

The acoustical classification scheme presented by ISO/TS 19488 is also a Likert type unipolar scale on which ‘performance degrees of building’ or ‘sound insulation properties of building elements’ are ordered with equal intervals. Hence, the acoustic classes could be substituted with the numbers (1-6 corresponding to A to F from best to worse). The interpolation was applied to determine the annoyance scores (AS), so that the lowest annoyance score matched the highest category of acoustic performance separately for each type of scale. Figure 3 gives the correlation between the scores and classes for the three scales with the relative linear regression equations. The calculated scores in relation to the acoustic classes, are listed in Table 2. The 5-point scale starts at “1” as it is used in practice (contrary to the Figure 2) to avoid confusion in the further evaluations, and the regression equation consists the intercept term. c. Relations between annoyance degrees on different scales, indoor levels and acoustic classes

The interaction between the annoyance scores at different scales and the independent variables of acoustic classes and noise levels, are analyzed based on the experimental indoor noise &response data given in Figure 1 [34]. The figure shows the original regression equation derived from the laboratory study for overall annoyance question rated by using the 7-point scale. The original dose- response relation chart given in Fig.1, was revised according to the 11-point scale after converted the original scores of 7-point to the standard scale of 11-point (Figure 4). Then, the annoyance scores implying the average values grouped on the indoor noise levels, were integrated with the acoustic class & annoyance scores diagram, to obtain the indoor noise level categories corresponding to the acoustic classes. The indoor level categories ( L ) calculated by using the new regression equation are shown on the graph, with the relative acoustic classes (AC) on the horizontal axis. The interval of the indoor noise level categories, is 8.4 dB(A). The calculated indoor levels ( L ) corresponding to the

—————— a 234567890 ‘i-ccemamnnenmeacan

annoyance scales ( AS ) by using the regression equations, are shown on the Table 3. The increments of noise levels with respect to the annoyance scores, vary as 4,2 dB/score, 7 dB/score and 10,4 dB/score, for the 11-point, 7-point and 5-point scales respectively.

1 1 1 1 0 2 2 1 3 3 2 4 4 3 5 5 4 6 5 6 7 6 7 8 9 6 10

Annoyance scales & acoustic classes

10

y = 2x - 2

Annoyance scale (all points)

9

R² = 1

8

7

y = 1,2x - 0,2

R² = 1

6

5

4

3

y = 0,8x + 0,2

2

R² = 1

1

0

1 2 3 4 5 6

A B C D E F

A B C D E F

Acoustic classes

AS5 AS7 AS11

Figure 3: Comparison of the annoyance scales with respect to acoustic classes (AC 1 to 6 represent Class A to Class F . AS5, AS7, AS11 denote the three scales; 5-, 7- and 11-points respectively. Table 2: Annoyance scores corresponding to the acoustic classes obtained by simple conversions of the discrete points.

Acoustic classes

11-point scale 7-point scale 5-point scale

AC (1-6)

AS11 AS7 AS5 A(1) 0 1 1 B (2) 2 2.2 1.8 C (3) 4 3.4 2.6 D (4) 6 4.6 3.4 E (5) 8 5.8 4.2 F (6) 10 7 5

Annoyance & Leq indoor

(experimental data)

0 1 2 3 4 5 6 7 8 9 10

Group annoyance scores converted to 11-

y = 0,239x - 5,8262

R² = 0,8534

point scale

20 25 30 35 40 45 50 55 60 65 70

24,4 32,7 41,1 49,5 57,9 66,1

Figure 4: Average group annoyance values converted to 11-point scale and the relationship with the indoor sound pressure levels (the class levels shown below).

Indoor noise levels, L Aeq (30 min), dB

Table 3: Calculated noise le vels corresponding to the annoyance scores f or 3 annoyance scales (AS)

Leq,indoor, dBA AS 11 AS 5 AS 7 0 24.4

Annoyance

scores

1 28.6 24.4 24.4 2 32.7 34.8 31.3 3 36.9 45.2 38.2 4 41.1 55.6 45.2 5 45.3 66.1 52.2 6 49.5 59.1 7 53.7 66.1 8 57.9 9 62.0 10 66.1

The relationships between the indoor noise levels and the annoyance scores were determined for each scale. The regression equations obtained from the above analyses, are given in Table 4.

A nomogram to demonstrate the interrelations between different variables employed in this approach is presented in Figure 5. It provides the indoor noise levels with respect to the acoustic classes based on the converted scores on the three scales. The second graph from the left, is the basic dose- response function derived from the experimental study, (i.e. the indoor noise levels vs annoyance degrees). It should be noted that the basic dose & response data represents the total annoyance values using the group average scores in 7- point scale and the indoor noise levels used in the experiment. The same approach can be implemented if a new experimental (or field data) data, in the similar format, is available. If the new function based on the 11-point scale, is provided, the nomogram is reduced to two charts, so as to include the conversion of only two standard scales by eliminating the older scale of 7-point.

Table 4: Linear regressio ns between acoustic classes and indoor levels for three a nno yance s ca les.

Annoyance scales AS 5 AS 7 AS 11 Annoyance scores vs. acoustic classes AS5=0.8AC+0.2 AS7=1.2AC-0.2 AS11=2AC-2

Annoyance scores vs. indoor sound pressure levels AS5=0.0958 L indoor -1.3305 AS7=0.1434 L indoor -2.4957 AS11=0.2395 L indoor -5.8262

AS7= 1.5AS5-0.5 AS7= 0.6AS11+1 AS5= 0.4AS11+1

Scale relations

Figure 5: A nomogram to determine the interactions of the variables used in the study

The proposed indoor noise level categorization associated with the acoustic classes is shown on the matrix given in Table 5. If the noise level categories will be used for the regulatory purposes, they should be presented, as the acoustic class levels to be “equal or lower than..”. i.e. A: ≤ 24.4, B ≤ 32.7 , C ≤ 41.1, D: ≤ 49.4 , E: ≤ 57.9 and F: ≤ 66.1 L eq , indoor dB(A).

Table 5: The matrix for the indoor noise level categories fixed at each acoustic class and corresp onding annoyance scores on different scales.

Acoustic classes (1-6) A B C D E F Indoor noise classes, L eq , dBA 24.4 32.7 41.1 49.4 57.9 66.1 11-point scale (AS11) 0 2 4 6 8 10 7-point scale (AS7) 1 2.2 3.4 4.6 5.8 7 5-point scale (AS7) 1 1.8 2.6 3.4 4.2 5 d. Investigation of high annoyance values using the individual response data

The percentage of highly annoyed people, HA% as explained in Section 2.1, is calculated from the individual response data. High annoyance concept in this study was evaluated with respect to both annoyance scores and percentages of the people highly annoyed subjects, by establishing the relations between indoor noise levels and acoustic classes.

As known, the basic criteria for HA% is to take the number of respondents distributed on the uppermost three scores on the 11-point scale, (i.e. 8+9+10) covering the 27.7% of the length of the total scale (3/11). To meet the above criteria, the lowest score in 11-point scale corresponds to 4.2 on the 5-point scale and 5.8 on the 7-point scale, as found through interpolations. These critical scores reflecting high annoyance were used to find the associated indoor noise levels using the regression equations given in Table 4. Figure 6 shows the comparison of these regression lines for the three scales with respect to the noise levels. As a result, the indoor noise level at which the high annoyance starts, is found as L eq,indoor =57.9 dB(A) which corresponds to Class E .

Acoustic classes corresponding to indoor levels for 3

annoyance scales

10

9

8

8

7

5.8

6

5

4.2

4

3

2

1

Indoor so und pressure levels, L eq ,dBA 0

20 25 30 35 40 45 50 55 60 65 70

24,4 32.7 B 41 C 49.4 D 57.7 E 66.1 F

Class noise levels, L eq ,dBA

High annoyance descriptor

High annoyance descriptor

Figure 6: Det ermination of the critical indoor noise level that evok e s “high annoyance” for 3 scales.

Annoyance scale (0-10) Annoyance scale(1-5) Annoyance scale(1-7)

To find the percentage of highly annoyed respondents from the r esults of the experimental study, the approach proposed by Brink et al. was used in this study [19,20] . Since the critical score calculated was 5,8, not a n integer, determination of the exact number in cate g ory 5 needs to apply a weight to the number o f respondents scoring 5. This weight, which implie s the fraction of the respondents scoring 5, is d etermined by following the steps as explained below ( Figure 7):

1. Each score (on 7 point scale) is accepted as the mid points of the category bands. 2. The upscal ing was applied at each score, between 0-100 . ( T he interval: 100/6 = %16.6 ) 3. The upscaled upper bounds and lower bounds of the categ o ries, are calculated. 4. Upsc aled i ntervals are calculated. 5. F above for the category in which the critical score lays, is c a l culated for weighting. Since U= 75% , L=58% and x= 72.73%, the formula in Figu re 7 yi elds F above = 0.134. This is t he weight to be a pplied to the number of res p ondents in category 5. T he total number of the subjects for

HA%, are calculated: (1) All subjects giving score 7+, (2) all subjects giving score 6+, (3) all the subjects giving score 5 after weighted by 0.134.

Similar method applied for the 5-point scale, gives F above = 0.36 implying that the number of the subjects for HA% should be determined by taking the total response number in category 5 + total number in category 4 weighted by 0.36 . For the 11- point scale; since the critical score is 8, the full numbers of the subjects giving scores of 10+9+8 will be taken into account on the HA% assessment.

Criteria for HA%

72,73% 27,27%

0 16,6 33,3 50 66,6 83 100 1 2 3 4 5 6 7

Scores (7-point)

Upscaled %

8,3 25 42 58,8 75 91

Upscaled intervals %

F above = U - x

U - L

x: 72,73% U: upper bound of category (%) L: Lower bound of category (%)

Figure 7. Implementation of the Brink method to find the weighting number for 7-point scale [30].

The calculated HA% values for the experimental data, are correlated with the indoor noise levels. Figure 8 gives the dose-response relationship using the HA%, with the derived polynomial regression equation. It was found that the critical indoor noise level 57.9 dB(A) as determined above, corresponds to 66% HA which is extremely high for the communities to be protected from noise. However much lower percentages of HA relative to the indoor levels, given in Table 6, can be proposed for the regulations. For example, HA%= 16% has been commonly accepted criteria in national regulations corresponding to Class C.

Led eesesused a= Leqindoor dBA

Figure 8: Indoor noise level & HA% relationship from the laboratory data (HA% was calculated at each noise level as explained above). The total response: 192.

Table 6: Calculated HA%’s by using the polynomial regression equations for each category of noise level (from the experimental data).

Acoustic Classes A B C D E F Indoor noise class levels 24.4 32.7 41.1 49.5 57.9 66.1 HA% 0 2% 16% 38% 66% 100%

3.3. Confirmation of the relationship with the field survey data Outline of the field study:

A socio-acoustic survey [43, 44] was made between 2017 and 2019 in order to collect data about applicability of the Turkish Regulation on Protection of Buildings against Noise [41]. Six buildings were selected to represent different noise characteristics in relation with the urban characteristics. Four of these buildings (Building B1, B2, B3, B4) were located in residential complexes with green areas while others (Buildings B5, B6) being in the city centre. Recently, in 2022, two additional buildings in the city centre (Buildings B7, B8) were measured. All of the buildings were located in İstanbul except for Building B2, which was located in Ankara. Considering the nearby noise sources; Building A was facing a playground, Buildings B2 and B3 were in silent locations, Buildings B4, B5 and B8 were located next to a busy road and Buildings B6 and B7 were located on a street with moderate traffic. The buildings had reinforced concrete structural system and the façade materials were hollow brick (Building B1, B2, B3 and possibly B8), aerated concrete (B4, B6, B7), and glass (Building B5). All of the buildings had double glass windows. For the acoustical performance evaluation, the Regulation requires night-time (23.00 – 07.00) indoor noise measurement for bedrooms of residential buildings [41]. In each building, one of the bedrooms was measured in floors 1-3 in compliance with ISO EN 1996 –1, –2. Following the measurements, face-to-face survey was conducted with the residents. More than twenty people were selected randomly from Buildings B1- B6. Only one individual’s response was gathered from Buildings B7 and B8. 138 residents were interviewed at total and 132 of the responses were valid for analysis. A questionnaire of three pages was prepared and the interviews took 20 to 45 minutes [43]. During the interviews, residents were asked to evaluate their annoyances from outdoor and building noise sources using both the 5-point Likert scale and the numerical scale from 0 to 10. Traffic, speech, music and construction noises were evaluated as outdoor noise sources. For the current study, the maximum of the traffic annoyances, were used for the evaluations in order to provide comparable results with laboratory findings.

Results show that between traffic and speech annoyances, speech annoyance is more prominent in residential complexes while traffic is more prominent in buildings near the city centre. Table 7 shows the indoor noise levels, average outdoor noise annoyances and percentage of highly annoyed residents in selected buildings.

Table 7: The indoor noise levels and average outdoor noise annoyances in buildings.

B1 B2 B3 B4 B5 B6 B7 B8 Measured L eq indoor , dBA 30 19 20 29 33 26 26 32 Avg. outdoor noise annoyance (11-point scale) 3.4 0.4 1.2 0.9 4.1 3.0 3 7 Percentage of highly annoyed residents ( ≥8/10) 9.5 0.0 0.0 4.5 16.7 13.6 0.0 0.0 Combined data for determination of HA% : The field data reflects very low indoor levels & annoyance degrees, therefore can not be comparable with the experimental data. An attempt was made to integrate the results with the experimental data to determine the combined dose-response relationship. Figure 9 shows the variation of HA% at each category of noise levels. The regression coefficient becomes slightly higher for the combined data and the HA% corresponding to the acoustic class E, appears to be 6% lower.

Figure 9: Relationship between indoor noise levels & percentages of highly annoyed respondents obtained from the field study and the simulated laboratory experiment. Total number of cases: 322 (192 (laboratory), 130 (field).

HA% (Combined)

60

y = 0,0447x 2 - 1,9595x + 24,173

50

R² = 0,7423

HA % (from combined data)

40

30

20

10

0

10 15 20 25 30 35 40 45 50 55 60

L eq ,indoor,dBA

4. CONCLUSION

The annoyance from outdoor noises at home are presented by using various rating scales and the conversion of the scales are of importance in comparison of different survey results. On the other hand, the interrelation between annoyance and the acoustical performance classes used for buildings, should be considered in terms of the indoor noise levels measured or predicted in buildings. In this study involving with the interactions of the dependent and independent variables, the categorization of the indoor sound pressure levels were determined through regression analyses, in accordance with the acoustic classes (based on the experimental data). The results were verified with the field study and the HA% values corresponding to the acoustic classes were presented both for the experimental and the combined data. 5. REFERENCES

1. ISO/TS 15666:2021 Acoustics-Assessment of noise annoyance by means of socio-acoustic surveys 2. ISO/TS 19488:2021 Acoustic classification of dwellings 3. Directive 2002/49/EC of the European Parliament and of the Council of 25 June 2002 relating to the assessment and management of environmental noise - Declaration by the Commission in the Conciliation Committee on the Directive relating to the assessment and management of environmental noise 4. Kurra S., Environmental Noise and Management-Overview from Past to Present , Wiley, 2020 5. Schultz T.J., Synthesis of social surveys on noise annoyance, The Journal of the Acoustical Society of America 64, 377 (1978); 6. Miedama H.M.E., Vos H., Exposure –response relationships for transportation noise. Journal of the Acoustical Society of America , 104, 3432-3445 (1998) 7. Fidell, S., Barber, D., and Schultz, T. J.(1991). “Updating a dosage-effect relationship for the prevalence of annoyance due to general transportation noise,” J. Acoustical Society of America 89, 221–233. 8. WHO Noise Guidelines for European Region , 2018 9. EU European Communities, 20 Feb., 2002, Position paper on dose response relationships between transportation noise and annoyance, Euronoise 2015 , Maastrich (2015) 10. Schomer, P., V. Mestre, S. Fidell, B. Berry, T. Gjestland, M. Vallet, and T. Reid (2012). Role of a community

tolerance value in predictions of the prevalence of annoyance due to road and rail noise, J. Acoust. Soc. Am ., 131(4), 2772-2786 11. Fidell S., Mestre V., Schomer P., Berry B., Vallet M., Reid T., A first- principles model for estimating the

prevalence of annoyance with aircraft noise exposure, appendix 12, J. Acoustical Society of America , 130 (2), (2012) 12. ISO 1996-1: 2016 (confirmed in 2021) Description, measurement and assessment of environmental noise—Part 1:

Basic quantities and assessment procedures Annex D Relationship to estimate the percentage of a population highly annoyed and the 95%prediction interval as a function of adjusted day-evening-night and day-night sound levels 13. Fields J.M., Jong R.G., de, Gjestland T., Flindell I.H., Job R.F.S., Kurra S., Lercher P., Vallet M., Yano T., Guski

R., Flescher-suhr U., Schuemer R., Standardized general-purpose noise reaction questions for community noise surveys: research and a recommendation. Journal of Sound and Vibration. , 242, 641-679 (2001) 14. Fields J.M., Jong R.G., de, Brown A.L., Flindell I.H., Gjestland T., Job R.F.S., Kurra S., Lercher P., Schuemer-

kohrs A., Vallet M., Yano T., Guidelines for reporting core information from community noise reaction surveys. Journal of Sound and Vibration , 206(5), 685-695 (1997) 15. Guski R, Schreckenberg D., Schuemer R.WHO environmental noise guidelines for the European Region: a

systematic review on environmental noise and annoyance. Int. J Environ Res Public Health . 14(12) (2017) 16. Gjestland T., Comments to be revised draft of ISO-DTS 15666, SINTEF 17. Gjestland T., A systematic review of the basis for WHO’s new recommendation for limiting aircraft noise

annoyance, International Journal of Environmental Research and Public Health , 15,2717, MDPI (2018) 18. Clark C., Gjestland T., Lavia L., Notley H., Michaud D., Morinaga M., Revising ISO/TS 15666 - The noise

annoyance standard, ICBEN 2021 ,June 2021, Stockholm, 19. Brink M., Giorgisand L., Schreckenberg D. and Evrard A.S., Pooling and Comparing Noise annoyance Scores and

High Annoyance”(HA) Responses on the 5-point and 11-point Scales: Principles and practical advice, International Journal of Environmental Research and Public Health, ,18,7339 MDPI (2021) 20. Brink M., Schreckenberg D., Vienneau D., Cajochen C., Wunderli J.M., Probst-Hensch N. and Röösli M., Effects

of Scale, Question Location, Order of response Alternatives and season on self- reported noise annoyance using ICBEN scales: A field experiment, International Journal of Environmental Research and Public Health, ,13 (11),1163 MDPI (2016)

21. Rindel J.H., Acoustical comfort as a design criterion for dwellings in the future, sound in the built environment,

Auckland, 21-23. Aalborg university, Denmark, 2002 22. Rindel J.H., Acoustic quality and sound insulation between dwellings. Journal of Building Acoustics. 1999, 5 pp.

291–301 23. Rasmussen B., Rindel J.H., sound insulation of dwellings-legal requirements in Europe and subjective evaluation

of acoustical comfort, proceedings DAGA 2003, Deutsche Gesselschaft für akustik. 24. Lovstad A., Rindel J.H., Hosoien C.O., Milford I., Klaboe R., Perceived sound quality in dwellings in Norway,

12 th ICBEN Congress , 18-22 June Zurich, 2017 25. Rindel J.H and Rasmussen B. Assessment of airborne and impact noise from neighbors. In: Proceedings of the

Inter-noise 97 , Budapest, 1739–1744 (1997) 26. Milford I, Høsøien CO, Løvstad A, et al. Socio-acoustic survey of sound quality in dwellings in Norway.

Proceedings of the Inter-noise 2016 , Hamburg, 21–24 August, 2016 27. Rindel J.H., Løvstad A., Klæboe R., 2016). Aiming at satisfactory sound conditions in dwellings – the use of dose

response curves. BNAM 2016, Stockholm, Sweden. Proceedings of the Baltic-Nordic Acoustics Meeting 2016 , Nordic Acoustics Association. 28. Hongisto V., Oliva D. and Keränen J., Disturbance Caused by Airborne Living Sounds Heard through Walls -

Preliminary Results of a Laboratory Experiment, Inter- noise 2013 , Innsbruck, Austria, 2013 29. Hongisto V., Subjective and Objective Rating of Airborne Sound Insulation – Living Sounds Acta Acustica

united with Acustica 100(5) 2014 30. Locher B. , Piquerez A. , Habermacher M. , Ragettli M. , Röösli M. , Brink M. , Cajochen C. , Vienneau D. ,

Foraster M. , Müller U. and Wunderli J. M. , Differences between Outdoor and Indoor Sound Levels for Open, Tilted, and Closed Windows, Int. J Environ Res Public Health . Jan. 15(1) 149 (2018) 31. Scamoni F., Scrosati C. The facade sound insulation and its classification; Proceedings of the Forum Acusticum ;

Krakow, Poland. 7–12 September 2014; p. 6. 32. Ohrstrom E. and Bjorkman M. Effects of noise disturbed sleep - a laboratory study on habituation and subjective

noise sensitivity. J Sound Vibration 122 (2), 277-290. (1988) 33. Rice, C.G. CEC joint research on annoyance due to impulse noise: laboratory studies, Proceedings of the 4th

International Congress on Noise as a Public Health Problem (1983). 34. Kurra S., Morimoto M., Maekawa Z., Transportation noise annoyance-A simulated environment study for road,

railway and aircraft noises, Part 1. Overall annoyance , Journal of Sound and Vibration , 220 (1) 251-278 (1999) 35. Kurra S., Morimoto M., Maekawa Z., Transportation noise annoyance-A simulated environment study for road,

railway and aircraft noises, Part 2. Activity Disturbance and combined results, Journal of Sound and Vibration, 220 (2) 279-295(1999) 36. Marquis-Favre C. and Morel Ju., A Simulated Environment Experiment on Annoyance Due to Combined Road

Traffic and Industrial Noises, International Journal of Environmental Research and Public Health 12 , 8413-8433 (2015) 37. Rychtarikova M., Roozen N.B., Muellner H. et al. , Perceived Loudness of Sound Transmitted through Light

Weight and Heavy Weight Walls, Advanced Materials Research, 649,101-104 (2013) 38. Vardaxis N.G., Bard D. and Waye K. P., Review of acoustic comfort evaluation in dwellings—part I: Associations

of acoustic field data to subjective responses from building surveys, Building Acoustics , 25(2) 151-170 (2018) 39. Rasmussen B., Machimbarrena M.,Fausti P.,Gerretsen E.,Smith R.S. Building acoustics throughout Europe

Volume 1: Towards a common framework in building acoustics throughout Europe , Chapter 6. Developing a Uniform Questionnaire for Socio-Acoustic Surveys in Residential Buildings 40. Environmental Protection Agency (1974). “ Information on Levels of Environmental Noise Requisite to Protect

Public Health and Welfare with an Adequate Margin of Safety ,” U.S. Environmental Protection Agency, EPA/ONAC 550/9-74-004, Washington, DC. 41. Bayazit, N. T., Kurra, S., Ozbilen, B. S., & Sentop, A, “ New Regulation on Noise Protection for Buildings and

Sound Insulation in Turkey “, 23 rd International Congress on Sound and Vibration, Athens Greece (2016) 42. Robbins N.B., Heiberger R.M., Plotting Likert and Other Rating Scales, Section on Survey Research Methods –

JSM 2011 43. Şentop Dümen A. and Tamer Bayazıt N. Enforcement of acoustic performance assessment in residential buildings

and occupant satisfaction, Building Research & Information , 48(8), 866-885 (2020) 44. Şentop Dümen, A. (2020). An Approach For Acoustic Performance Assessment Of Dwellings In The Context Of

Legislations And Subjective Evaluation. Istanbul Technical University (Unpublished PhD Thesis).