Content uploaded by Jens Holger Rindel
Author content
All content in this area was uploaded by Jens Holger Rindel on Jun 23, 2016
Content may be subject to copyright.
1
Sound quality in dwellings in Norway – a socio-acoustic investigation
Anders Løvstad, Jens Holger Rindel, Clas Ola Høsøien and Ingunn Milford
Multiconsult, Oslo, Norway, anders.lovstad@multiconsult.no
Ronny Klæboe
Institute of Transport Economics, Oslo, Norway, Ronny.Klaeboe@toi.no
An extensive socio-acoustic survey has been performed in Norway. Based on field measurements of
sound insulation in 600 dwellings, a questionnaire survey was sent to nearly 4000 residents. 702
respondents evaluated the sound quality of their homes, 97 % of which lived in apartments in multi-unit
houses. Additionally, sound insulation were calculated for a selection of student houses, where 386
students responded to the survey. The field measurement results were well distributed over a relatively
wide range, even though most of the dwellings were recent constructions, and thus supposed to comply
with the presently governing requirements. The range in the measurement results allowed for the
establishment of exposure-effect relationships for annoyance caused by both impact and airborne sound
insulation. The correlation between subjective evaluation and various sound insulation descriptors were
evaluated. Furthermore, annoyance due to different sound sources and levels were assessed, as well as
effects of using light or heavy building structures, willingness to pay for improved sound quality, and
whether people limit themselves to ensure that neighbours are not annoyed.
1 Introduction
Sound quality and noise exposure affect the quality and well-being of residents. An extensive socio-acoustic survey has
been done in Norway [1], aiming to assess whether present sound insulation and noise level requirements are adequate
or in need of revision. The survey mainly comprises multi-unit houses, as there are no sound requirements in single unit
dwellings. Considering the population in Norway as a whole, only 17 % live in multi-unit houses [2]. Thus, the survey
does not represent the average annoyance of the population as a whole, but give a solid basis for evaluation of sound
insulation and noise level limits where they apply.
The present minimum sound insulation requirements between dwellings and towards common corridors or staircases in
Norway are R'w ≥ 55 dB and L'n,w ≤ 53 dB [3]. Thus, only frequencies from 100 to 3150 Hz are considered, although the
regulations include a recommendation to consider frequencies down to 50 Hz as given by the spectrum adaptation terms
C50-5000 and CI,50-2500. The inclusion of the low-frequency adaption terms are assessed, as well as employment of the
standardized parameters DnT,w and L'nT,w, as these can be calculated from the measurement data as described in
ISO 717-1 [4] and ISO 717-2 [5].
2 Method
The survey is based on field measurements carried out in 600 dwellings from 2002-2015. In projects with several equal
multi-unit houses, residents in all houses received the survey, assuming the same building performance in all buildings.
Questionnaires were sent to 3849 persons, of which 702 responded, giving a response rate of 18 %. In addition, about
1500 students at three campus sites received questionnaires, from which 386 responded. These two selections are
labelled the main and student selections, respectively.
The main motivation for including students were to enhance the data set for small units. As no measurements data were
available for the student selection, sound insulation data were calculated based on available construction data.
2
The questionnaire developed in COST Action TU0901 [6] was used as basis, and adapted to the recommendations
given in ISO/TS 15666 [7], using a 5-point scale. The questions used in social surveys on noise problems are paramount
for the quality of the results, and details concerning the way questions were asked and the wording of the possible
answers followed recommendations given in ISO/TS 15666 [7]. Annex B in the standard recommends wordings in
various languages including Norwegian, as used in this investigation. The questions and possible answers are of this
form: "Thinking about the last 12 month, when you are at home, how much are you annoyed of noise from … ? Not at
all – Slightly – Moderately – Very – Extremely".
The questionnaire consisted of 35 different questions in addition to background information about the respondents. The
questions can be grouped as:
annoyance due to noise inside and outside of the dwelling
noise sensitivity
annoyance due to specific noise sources or neighbours
restrictions on own behaviour not to disturb others
willingness to pay for a better sound insulated dwelling
To obtain statistically reliable results, a good spread in the sound insulation and noise source levels is essential. Figure 1
(a) and (b) show the distribution in measurement results for weighted apparent sound reduction index, R'w, and the
weighted normalized impact sound pressure level, L'n,w, respectively. Results are divided into light and heavy
constructions in order to assess their perceived sound qualities separately. As expected, the heavy constructions are
better than the light ones both for airborne and impact sound insulation.
0
5
10
15
20
25
30
< 40 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 > 70
No. of measurements
Weighted apparent sound reduction index, R'w[dB]
Light
Heavy
0
5
10
15
20
25
30
< 40 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 > 70
No. of measurements
Weighted normalized impact sound pressure level, L'n,w [dB]
Light
Heavy
a)
b)
Figure 1: Distribution of measurement results for (a) R'w and (b) L'n,w.
3
3 General annoyance and restrictive behaviour
Which noise sources contribute the most to the total sound level in and around a building depends on several factors,
e.g. where and how the building is built, the neighbours, the residents' sensitivity to noise. Figure 2 shows the
annoyance due to different sound sources, separated into extremely, very, moderately and slightly annoyed. It is found
that around 2/3 of the residents are annoyed by noise to some extent. 10 % of the residents answer that they are
moderately or more annoyed by speech or music. More surprisingly, footfall noise is reported to be comparable and
even slightly more annoying than traffic noise as more than 40 % are annoyed to some extent, and 20 % are at least
moderately annoyed from these noise sources.
It was not within the scope of the project to evaluate traffic noise specifically, so noise exposure levels from traffic have
not been correlated with the responses obtained. However, there is no reason to believe that these are particularly low.
Complaints due to impact sound are well known, especially for lightweight constructions.
0
10
20
30
40
50
60
70
Speech Music Footfall Traffic noise Total
Cumulative proportion [%]
Extremely annoyed Very annoyed Moderately annoyed Slightly annoyed
Figure 2: Cumulative proportion of annoyance due to different sound sources.
0
10
20
30
40
Speech Music Footfall
Percentage [%]
Annoyed (extremely+very+moderately) Worried about disturbing others
Figure 3: Annoyance due to different sound sources.
4
Negative effects of noise and lack of sound insulation are commonly related to health issues and reduced quality of
living for the receiver being exposed to these high noise levels. Another interesting aspect is whether one restricts
oneself in ones doings in fear of being a nuisance to and annoy neighbours. In Figure 3, results from questions
regarding annoyance and questions on whether one is worried about disturbing others are placed together.
As many as 1 out of 3 report that they are worried that airborne noise sources such as TV, music, speech may annoy
their neighbours, while only around 10 % reports to be moderately or more annoyed. This indicates that people are
likely to limit themselves to a certain extent in consideration of their neighbours for these types of activities. With
footfall noise this is not the case, and the percentage worried about disturbing others is lower than the percentage of
those annoyed by it.
4 Willingness to pay
Strict sound insulation requirements and noise limits will reduce annoyance, but in most cases housing units will be
more expensive to build in order to acquire such qualities. When asked if the current regulations on noise and sound
insulation are too strict or too lenient, only 1 % of the respondents in the main selection thinks the regulations are too
strict, while 38 % answered that the requirements are too lenient, as shown in Figure 4a.
Around 50 % of the 702 replies in the main selection would not pay extra for better sound insulation, but around 20 %
were willing to pay from 6 000 to 12 000 NOK extra per year, as Figure 4b shows. Figure 4c assesses the opposite
question, less sound insulation for a reduced monthly payment. 86 % replied that this would be out of question; only
10 % would accept lower sound insulation against a lower payment around 12 000 NOK per year.
9 %
10 %
18 %
16 %
47 %
61 %
38 %
1 %
1 %
1 %
2 % 10 %
86 %
9 %
10 %
18 %
16 %
47 %
1000 NOK/month 500 NOK/month
200 NOK/month 100 NOK/month
0 NOK/month
61 %
38 %
1 %
Adequate
Too lenient
Too strict
1 %
1 %
2 %
10 %
86 %
100 NOK/month 200 NOK/month
500 NOK/month 1000 NOK/month
Not interested in poorer quality
1 %
1 %
2 %
10 %
86 %
100 NOK/month 200 NOK/month
500 NOK/month 1000 NOK/month
Not interested in poorer quality
a) c)b)
Figure 4: a) The respondents’ opinion about the national sound insulation requirements for dwellings, b) Willingness to
pay for better sound insulation, and c) willingness to accept poorer sound insulation for a compensatory amount.
5 Airborne sound insulation
Annoyance levels due to airborne sound insulation between units indicated that the present requirement of R'w ≥ 55 dB
is set at a reasonable level. Similar annoyance levels were found for normalized and standardized descriptors. For music
with bass and drums, inclusion of the low-frequency spectrum adaptation term, C50-5000, gave slightly better
correspondence, but there was no significant difference. These results are further discussed in [1, 8].
Small flats were of specific interest as part of the mandate from the government was to investigate whether building
procedures could be simplified and be more area effective. Present building regulations require two doors between
corridors or common areas to comply with R'w ≥ 55 dB. In practice, the sound insulation comply with the regulations
only when there is a separate hall or entrance. Otherwise the sound insulation is significantly below the requirement
with only a single door between the living room and corridor. In the latter case measurement results are typically in the
range from R'w ≥ 35-45 dB.
5
0
10
20
30
40
50
60
70
80
90
100
30 35 40 45 50 55 60 65 70
Cumulative proportion [%]
R'w[dB]
Extremely annoyed Very annoyed
Moderately annoyed Slightly annoyed
Figure 5: Cumulative proportion of annoyance due to sound insulation from corridor, staircase etc. to housing units.
Solid lines are within 95 % of the range of the measured data.
Figure 5 shows cumulative annoyance due to sound insulation from corridors and staircases to housing units. The mean
apparent sound reduction index value, R'w, was 51,6 dB with a standard deviation of 7,0 dB. As mentioned above, there
is a leap in sound insulation between situations where one or two doors are used. With two doors R'w ≥ 55 dB is
achieved, while results are typically 10-15 dB below this requirement with a single door between the housing unit and
adjacent corridor or staircase. Thus, the measurement results are unevenly distributed, and therefore not the best basis
for evaluation. Still, the results indicate that satisfactory sound insulation are achieved at around R'w ≥ 50 dB, but using
one door between corridors and housing units are probably not sufficient in most cases. The degree of annoyance was
found to be independent of at which floor the units were located.
Studentutvalg, alle typer med direkte utgang, plage relatert til tale
osv. i gang, trapperom mv.
Extremely, very or moderately annoyed Slightly annoyed Not annoyed
25 %
6 %
69 %
5 % 5 %
10 %
86 %
12 %
84 %
a) c)b)
Figure 6: Proportion of annoyance due to sound insulation from corridor, staircase etc. to flats, in percent. a) Main
selection. b) Student selection with two doors towards common corridor. c) Proportion of the student selection with a
single door towards the common corridor.
Figure 6 shows the proportion of annoyance due to sound insulation between common staircases, corridors etc. and
flats. The proportion of annoyance for the main selection is shown in a), while b) and c) show results for units in the
6
student selection with two and one door towards the common corridor, respectively. The single doors in c) have a sound
insulation of Rw = 43 dB and are mounted in good wall partitions, giving a total sound insulation of R'w = 43-45 dB.
Presently, units in the main selection with a single door towards the common staircase or corridor have not been
analysed separately due to tight time schedule and budget in this project.
The results for the main selection and student selection with separate entrance are comparable, while there is clearly a
higher degree of annoyance with only a single door towards the corridor. Student houses commonly have long corridors
leading to several housing units, indicating that the traffic load might be higher in these cases than in other multi-unit
houses. An interesting further investigation would be to assess the correlation between annoyance and number of units
each corridor leads to, which has not been done presently.
These results indicate that the sound insulation requirement can be reduced to around R'w = 50 dB. Thus, there is still
need for a door number two between housing units and common areas.
6 Impact sound insulation
Impact sound insulation data and results for the main selection are shown in Table 1. These data are in conjunction with
Figure 1b, but deviate somewhat as these data are from the responses obtained while Figure 1 shows the measurement
database. As questionnaires were sent to all residents living in houses at the same address, the response data differs
from the measurement basis to a certain extent.
Table 1. Data for measured impact sound insulation and the degree of annoyance due to impact sound related to the
mean value of impact sound insulation.
L'n,w
L'n,w + CI,50-2500
L'nT,w
L'nT,w + CI,50-2500
Number
473
439
411
402
Mean, dB
49,4
53,7
45,4
50,1
Standard dev, dB
4,1
3,0
5,0
2,0
Slightly annoyed
45,5 %
42,6 %
43,3 %
42,8 %
Moderately annoyed
22,2 %
19,9 %
20,0 %
19,1 %
Very annoyed
12,3 %
11,2 %
11,2 %
10,2 %
Extremely annoyed
4,5 %
3,8 %
4,3 %
3,8 %
The degree of annoyance is comparable for the four acoustic descriptors evaluated. More than 42 % are annoyed to
some extent, and around 20 % are moderately or more annoyed. This indicates that annoyance levels are higher than
presupposed, as the mean impact sound level L'n,w = 49,4 dB meets the requirement with almost 4 dB margin.
The normalized mean values are around 4 dB higher than the standardized values, which is explained by the room
volume correction [8]. The mean spectrum adaptation term CI,50-2500 is 4-5 dB, indicating that there is considerable
transmission at low frequencies, and that floor construction used exhibit limited sound insulation below 100 Hz.
Figure 7 shows the cumulative proportion of annoyance due to impact sound in horizontally separated housing units
(i.e. units on different floors). In a) the weighted normalized impact sound pressure level, L'n,w, is shown. It is clear that
there is no correlation between the subjective annoyance reported by the residents and the objectively measured values.
The same result was found for L'nT,w, but this is not shown here. When the spectrum adaptation term CI,50-2500 is added,
the cumulative annoyance correlate much better with the subjectively reported annoyance levels, as shown in Figure 7b.
This was the case both for normalized and standardized descriptors.
Other researchers have also pointed out the poor low-frequency properties of timber and other light-weight
constructions [9]. A certain proportion of the constructions included in the survey are heavy floors with thin floating
floors, typically consisting of 30-40 mm lightweight concrete, anhydrite gypsum plaster or similar and 15-20 mm stone
wool insulation with parquet as the finishing layer. These constructions typically exhibit spectrum adaption terms in
excess of 10 dB, so even if they comply with the regulations (L'n,w ≤ 53 dB) with good margins, impact sound levels
below 100 Hz can be significant and induce annoyance in adjacent housing units. Similar results have recently been
reported by Wolf and Burkhart [10].
7
0
10
20
30
40
50
60
70
80
90
100
40 45 50 55 60 65
Cumulative proportion [%]
L'n,w [dB]
Slightly annoyed Moderately annoyed
Very annoyed Extremely annoyed
0
10
20
30
40
50
60
70
Speech Music Footfall Traffic noise Total
Cumulative proportion [%]
Extremely annoyed Very annoyed Moderately annoyed Slightly annoyed
0
10
20
30
40
50
60
70
80
90
100
40 45 50 55 60 65
Cumulative proportion [%]
L'n,w + CI,50-2500 [dB]
Slightly annoyed Moderately annoyed
Very annoyed Extremely annoyed
0
10
20
30
40
50
60
70
Speech Music Footfall Traffic noise Total
Cumulative proportion [%]
Extremely annoyed Very annoyed Moderately annoyed Slightly annoyed
a)
b)
Figure 7: Cumulative proportion of annoyance due to impact sound in horizontally separated housing units (i.e. units on
different floors). a) the weighted normalized impact sound pressure level, L'n,w and b) L'n,w + the spectrum adaptation
term CI,50-2500. Solid lines are within 95 % of the range of the measured data, while the dotted lines around the slightly
annoyed curve show the 95% confidence interval for a change in the slope (ignoring associated change in intercept).
7 Summary
An extensive socio-acoustic survey comprising multi-unit houses has been carried out in Norway. It is found that 2/3 of
the residents are annoyed by noise to some extent. Impact and traffic noise caused most nuisance, as 20 % reported to
be moderately, very or extremely annoyed from these noise sources, which is twice as high as from speech or music.
8
However, as many as 1 out of 3 report that they are worried that airborne noise sources such as TV, music, speech may
annoy their neighbours, and are thus likely to limit themselves to a certain extent in consideration of their neighbours
for these types of activities. With footfall noise this is not the case, and the percentage worried about disturbing others
are lower than those annoyed by it.
Only 1 % of the respondents in the main selection thinks the regulations are too strict, while 61 % answered that the
requirements are adequate. Around 50 % of the main selection would not pay extra for better sound insulation, but
around 20 % were willing to pay from 6 000 to 12 000 NOK extra per year. On the other hand, 86 % replied that less
sound insulation for a reduced monthly payment is out of question.
Annoyance levels due to airborne sound insulation between units indicated that the present requirement of R'w ≥ 55 dB
is set at a reasonable level. The results indicate that sound insulation between housing units and common corridors or
staircases can be reduced to around R'w = 50 dB, but there is still need for a second door between housing units and
common areas.
Annoyance levels due to footfall noise are higher than presupposed, as the mean impact sound level L'n,w = 49,4 dB
meets the Norwegian requirement with almost 4 dB margin. No correlation between the subjective annoyance reported
by the residents and the objectively measured values was found if the spectrum adaptation term CI,50-2500 was not
included, both for normalized and standardized descriptors. The mean spectrum adaptation term CI,50-2500 is 4-5 dB,
indicating that there is considerable transmission of impact sound at low frequencies, and that the used floor
constructions exhibit limited sound insulation below 100 Hz.
8 Acknowledgements
This work was supported by “Direktoratet for byggkvalitet” (DIBK), National Office of Building Technology and
Administration, Norway. The initial preparation of the measured data from all the building sites were done by Sigrid
Husebø Øygard. Hanne Beate Sundfør at the Institute of Transport Economics assisted with the preparation and analysis
of the questionnaire.
References
[1] DIBK, Lydforhold i boliger. Evaluering av byggtekniske krav til lydforhold, (In Norwegian). Document code
127762-RIA-RAP-001, March 2016. Direktoratet for byggkvalitet (DIBK), National Office of Building
Technology and Administration, Oslo, Norway, 2016
[2] Statistisk sentralbyrå (SSB), Folke- og boligtellingen, boliger, (In Norwegian). Seksjon for befolkningsstatistikk,
Statistisk sentralbyrå, 2011.
[3] NS 8175, Lydforhold i boliger. Lydklasser for ulike bygningstyper, (In Norwegian), Standard Norge, 2012.
[4] ISO 717-1:2013, Acoustics - Rating of sound insulation in buildings and of building elements - Part 1: Airborne
sound insulation. Second Edition 1996, Third Edition 2013. Organization for Standardization, Geneva,
Switzerland.
[5] ISO 717-2:2013, Acoustics - Rating of sound insulation in buildings and of building elements - Part 2: Impact
sound insulation. Second Edition 1996, Third Edition 2013. Organization for Standardization, Geneva,
Switzerland.
[6] COST Action TU0901, Integrating and Harmonizing Sound Insulation Aspects in Sustainable Urban Housing
Constructions, European Cooperation in Science and Technology, 2014.
[7] ISO/TS 15666:2003, Acoustics - Assessment of noise annoyance by means of social and socio-acoustic surveys.
Organization for Standardization, Geneva, Switzerland.
[8] J.H. Rindel, A. Løvstad and R. Klæboe, Aiming at satisfactory sound conditions in dwellings – the use of dose-
response curves, Proc. Baltic-Nordic Acoustic Meeting, Stockholm, 2016.
[9] F. Ljunggren, C. Simmons and K. Hagberg, Correlation between sound insulation and occupants’ perception –
Proposal of alternative single number rating of impact sound, Applied Acoustics 85, 57-68, 2014.
[10] M. Wolf and C. Burkhart, Vermeidungsstrategien und Ansätze einer Vermeidung des Estrichdröhnens, (In
German), Proceedings of DAGA 2016, Aachen, 2016.
