A A A Can you really put a price on a good night’s sleep? Lukáš Zelem 1 Department of Materials Engineering and Physics, Faculty of Civil Engineering, STU Bratislava, Radlinského 2766/11, 810 05 Bratislava, Slovakia Vojtech Chmelík 2 Department of Materials Engineering and Physics, Faculty of Civil Engineering, STU Bratislava, Radlinského 2766/11, 810 05 Bratislava, Slovakia Andrea Vargová 3 Department of Building Construction, Faculty of Civil Engineering, STU Bratislava, Radlinského 2766/11, 810 05 Bratislava, Slovakia Arnon Vandenberghe 4 Department of Computer Science, Faculty of Engineering Science, KU Leuven, Kasteelpark Arenberg 1 bus 2200, B-3001 Leuven, Belgium Monika Rychtáriková 5 Department of Architecture – Campus Brussel and Gent, KU Leuven, Hoogstraat 51, 9000 Gent/Paleizenstraat 65, B1030 Brussel, Belgium ABSTRACT This article focuses on investigation on to what extent would inhabitants of dwellings be willing to pay extra price for an apartment with increased sound insulation. A typical 3 room apartment in Bratislava was chosen as a case study. Online listening tests were used as a tool to understand the preferred price-quality ratio. The primary financial value of a case study apartment was estimated on real estate market in the third quarter of 2020. Later, nine (9) different variants of sound insulation of partition walls separating dwellings were considered. These walls were chosen on the basis of different con- struction system, two heavy-weight: (1) brick and (2) concrete, and (3) light-weight double walls-based gypsum boards. Each construction base was divided into three categories of defined by weighed sound transmission index R w (53 dB, 55 dB and 59 dB). The four types of sounds (pink noise, cough, quarrel, party noise) were filtered through the wall spectra, e.g. sound transmission index R (dB) of chosen walls. These sounds were presented during the online listening test along with a total price for the particular apartment. The test subjects were asked to decide, which apartment they would choose based on an increased acoustic comfort and increase price of the apartment. 1 lukas.zelem@stuba.sk 2 vojtech.chmelik@stuba.sk 3 andrea.vargova@stuba.sk 4 arnon.vandenberghe@student.kuleuven.be 5 monika.rychtarikova@kuleuven.be worm 2022 1. INTRODUCTION Several factors influence the price of dwellings, but the most prominent are location of the real estate property and its size. In Slovakia different evaluation methods for price estimation exist, such as method of positional differentiation, comparative method and combined method [1]. However, when selling-buying apartment, the market price typically dominates. Few studies have been performed in the past that related to involvement of the sound insulation aspects into value of the apartment. [2] These were however very theoretical/ academic studies, that have served as an initial look into this topic. When buying apartment, sound insulation of internal walls and floors, such as partitions with neigh- bours is typically not the first thing that buyer thinks about. Insufficient sound insulation is therefore often discovered only after the contracts are signed and inhabitants start to live in their new homes [3] “Insufficient sound insulation” belong to main complains in newly build dwellings in Slovakia. There are even cases when partition walls between the two apartments were build out of porous bricks with 8 cm thickness with almost no sound insulation, in order to increase the floor area of apartment which depredated the living comfort and value of apartments dramatically [4]. Our study has therefore two main goals. First, (based on listening test experiment) to understand the appreciation of improved acoustic comfort expressed by extra budget, that a buyer would be willing to pay. And second, is to make the potential dwellers aware of sound insulation, by letting them listen to neighbour’s noise through different walls constructions with different insulation properties. 1. DESCRIPTION OF THE CASE STUDY In this article we focus on a typical 3 room apartment in Bratislava with floor area of ca 76 m 2 . The initial price of the apartment was based on the typical price per 1 m 2 of apartments in Bratislava, in the fourth quarter of the year 2020 [5]. Nine (9) alternatives of the case study apartment were created by modifying the partition wall between dwellings. Therefore 9 different partition walls were considered from which: Three walls were based on ceramic bricks, three were from concrete and three were light-weight double walls based on gypsum boards (Figure 1). The three walls within each material base were carefully chosen, to fall into 3 dif- ferent sound insulation categories, defined through weighed sound transmission index R w . The total price of each of the 9 apartment variants depended on the particular partition wall. The price of the wall was calculated as a sum of the costs on building material including auxiliary structures (such as formwork in the case of monolithic walls) and cost of the work. 2. DESCRIPTION OF THE 9 PARTITION WALLS The main criterion for the selection of partition walls was the weighted sound reduction index R w , and their construction / material type. Based on the R w , three sound insulation categories were chosen. The walls in category I had R w = 53 dB, in the category II it was R w = 55 dB and category III represented walls with the R w = 59 dB. For each category there were three types of constructions (material) used for partition walls. The first type was based on reinforced concrete, the second type was based on ceramic brick masonry and in the third type were a lightweight gypsum board wall. worm 2022 worm 2022 In case of the heavy-weight constructions, the first two categories (I and II) were designed as single- layer constructions and category III as double walls (Table 1). In case of light-weight walls, obviously, all walls were based on mass-spring-mass system, i.e. double walls. Figure 1 shows the sound trans- mission index of all walls, based on Insul 9.0 software [6] calculations. The chosen walls represent typical partition walls, that could be potentially used in Slovak dwellings. Table 1 Schemes of dwelling walls used in test with a parameters of airborne sound insulation. Category Type C 50-5000 (dB) Concrete Brick Gypsum Concrete Brick Gypsum Category I R w = 53 dB -1 0 -8 Category II R w = 55 dB 0 -1 -8 Category III R w = 59 dB 0 0 -11 Concrete Brick Gypsum 90 90 90 80 80 80 70 70 70 60 60 60 R (dB) R (dB) R (dB) 50 50 50 40 40 40 Category I Category I Category I 30 30 30 Category II Category II Category II 20 20 20 Category III Category III Category III 10 10 10 50 80 50 80 1250 2000 3150 5000 1250 2000 3150 5000 125 200 315 500 800 125 200 315 500 800 50 80 1250 2000 3150 5000 125 200 315 500 800 f (Hz) f (Hz) f (Hz) Figure 1 Sound transmission index R (dB) of walls used in listening test experiment 3. SOUND STIMULI Four different noise stimuli were selected for listening tests. The three of them were specific sounds that commonly occur at our households – i.e. party noise, cough and quarrel. For sake of comparisons, also pink noise was included in the study (as a stimulus No.4). Stimuli were prepared using frequency filter based on transmission spectra (Figure 1) in the frequency range between 50 Hz and 5 kHz. Stim- uli were later digitally calibrated/flattened based on their RMS value. The basic level of RMS value for all stimuli was 60 dB. The frequency spectra of all stimuli are separately shown in Figure 2 (left, middle, right). Stimuli spectra of party noise, cough and quarrel are shown the Figure 2 always in comparison with pink noise stimuli. 90 90 90 Cough Quarrel Party 80 80 80 Pink noise Pink noise Pink noise 70 70 70 60 60 60 L P (dB) L P (dB) L P (dB) 50 50 50 40 40 40 30 30 30 20 20 20 worm 2022 10 10 10 20 50 80 20 50 80 20 50 80 1250 2000 1250 2000 1250 2000 31,5 31,5 31,5 125 200 315 500 800 125 200 315 500 800 125 200 315 500 800 f (Hz) f (Hz) f (Hz) Figure 2 Sound pressure level of stimuli shown in the third octave bands. Each stimulus had certain unique feature. The low frequency content of sound in households is repre- sented in the “party noise” (Figure 2- left). The impulsive sound with dominant frequencies in the range of 200 Hz – 1.25 kHz is given in the sound stimulus “cough” (Figure 2 – middle) and the “quarrel” sound represents the sound of verbal communication (Figure 2 – right). 4. LISTENING TEST The online listening test was offered in two languages, English and Slovak. After reading the instruc- tions and choosing the suitable headset, calibration of the sound level in the headphones was per- formed as following. The test was calibrated individually for each test subject, and it was based on a calibration matrix (Fig. 3). The purpose of the calibration was to adjust the volume level so that the test subject could hear the stimuli in the considered octave bands. The rows of the calibration matrix represented octave bands from 125 Hz to 4 kHz, the bars represent the volume level from -15 to +15 dB in 5 dB step. Calibration Figure 3 Example of a filled calibration matrix (left) and main test window (right). Which apartment would you buy? = 1/288. ‘Apariment A Apartment 8 After the volume adjustment a window with information on a test protocol was displayed and test subject could start the test and perform it at comfortable speed. In total 39 people participated at the test, 24 male and 15 female. Listening test protocol In order to obtain the price/comfort ratio, resp. to understand how much a buyer would be willing to pay for an apartment which is acoustically better insulated a pairwise comparison method was chosen [7]. In the listening test, participant could see two possible prices of an apartment (depending on the price of the partition wall) and could listen to two random neighbour’s noise stimuli as filtered through the respective walls. In this way, he or she got an impression on the given sound insulation and was asked to choose the one of the apartments (by clicking on the computer screen). The asked question was: “Which one of the two apartments would you buy?” Altogether, 4 stimuli and 9 types of partition walls were tested. Combinations were created based on one stimulus and were compared to each other. The whole test was performed twice in one test session (test and retest), i.e. 288 questions in total. The experiment was originally planned to take place in the acoustic laboratory, but due to pandemics, it has been decided to go at least for an online listening test format. The test has been programmed with PsychoPy3 [8] and distributed online by means of Pavlo- via.org [9] PsychoPy3 runs locally in the Python environment, but to effectuate the coupling with Pav- lovia, all scripts are written in the JavaScript language. The test is organized as a sequence of windows in the time domain, with a flow from one window to another based on user interaction. 5. RESULTS AND DISCUSSION In the Figure 4 we can see 4 graphs expressing the results per stimulus. In each figure it can be seen, how many times the apartment with specific dwelling wall was selected as the test subject's preference. The solid columns in (orange, blue and green) show number of chosen responses for each particular case. The transparent columns (black lines) indicate the absolute price per apartment. Each of the 4 figures shows the result per type of structure (concrete, brick, gypsum boards) and the category of sound insulation (I, II and III). Quarrel Cough 202 000 600 202 000 600 Price per apartment (€) Price per apartment (€) Number of selections (-) Number of selections (-) 200 000 500 200 000 500 198 000 400 198 000 400 196 000 300 196 000 300 194 000 200 194 000 200 192 000 100 192 000 100 190 000 0 190 000 0 Concrete Brick Gypsum Concrete Brick Gypsum Category I Category II Category III Price per apartment Category I Category II Category III Price per apartment worm 2022 Party Pink noise 202 000 600 202 000 600 Price per apartment (€) Price per aparment (€) Number of selections (-) Number of selections (-) 200 000 500 200 000 500 198 000 400 198 000 400 196 000 300 196 000 300 194 000 200 194 000 200 192 000 100 192 000 100 190 000 0 190 000 0 Concrete Brick Gypsum Concrete Brick Gypsum Category I Category II Category III Price per apartment Category I Category II Category III Price per apartment Figure 4 Comparison of the answers of the tested subjects and the purchase price for the apartment. In general, people are willing to pay more for appartement with excellent sound insulation (Category III indicated in green colour). This trend is present within each construction style. If we look into the results per stimulus, we can conclude that in case of human voices, i.e. “quarrel sound” at neighbours, most of the participants would be willing to pay extra for an apartment in case of lightweight gypsum board Category III. In this case also Category I and II score better than any single heavy weight wall. Masonry and concrete walls in Category III, i.e. double walls, score better then Lightweight Category I or II. This result is logical, since mass-spring-mass systems insulate mid- dle frequencies better than constructions with only one mass. In case of the “coughing sound” there is the least difference between the construction styles. In all cases, the Category I and II are almost identical and the preference for Category III is obvious. The most preferred wall is the double concrete wall; however, the double masonry wall and lightweight wall are not much worse. This result is also logical. Coughing is impulsive sound with short duration and high intensity and so very audible. As shown earlier in the Figure 2 there is an increase in the middle frequencies and so very difficult to insulate by only mass law. Double walls are preferred. From the results in can be concluded that “party noise with strong bass sound” produced by neighbours, would convince people to pay more for appartement with walls made out of heavy weight partitions. In this case concrete scores even better than masonry. In case of intolerant neighbour, people would prefer to buy appartement with concrete walls. Here even Category I (concrete) is more preferred than lightweight walls of the Category III. In case of masonry wall, Category I and II score the same as Lightweight wall of Category III. This is interesting, but also rather logical, because the low frequency content of party noise was here very high. For sake of comparison, artificial noise (pink noise) was included into the study. Results show very similar trend as it can be seen for the party noise. This is on one hand very logical, because the spectra of both are similar, but it is also interesting, because the two stimuli have very different temporal con- tent. worm 2022 6. CONCLUSION It can be concluded that sound insulation of dwellings has been identified by respondents as an im- portant apartment quality issue. Interestingly, in spite of the highest costs, the most chosen was Cate- gory III, with R w = 59 dB. Concrete and masonry walls were chosen more often when stimuli contained strong low frequency components (Pink noise, Party noise). Double walls based on gypsumboard were the most frequently chosen (preferred) when quarrel (human voices) stimulus was presented. Finally, it is important to understand that this study is just a preliminary study, that has mainly served for testing of the methodology. We presume, that for instance, slight differences between the party noise and pink noise would be detectable in case we would repeat the experiment in laboratory envi- ronment that allows for listening to very silent stimuli. 7. ACKNOWLEDGMENT The authors would like to take this opportunity to thank the VEGA 1/0205/22 support grant. 8. REFERENCIES 1. Decree No. 492/2004 Coll. Ministry of Justice of Slovak republic. 2. Vargová, A., Rychtáriková, M. (2016): Is the sound insulation of a facade a relevant factor in price estimation of an apartment DAGA 2016, Aachen. 3. Rindel, J. H. Sound Insulation in Buildings , CRC Press, 2017. 4. Majdúch, D. Všeobecná hodnota stavieb a pozemkov . Bratislava: STU, 2006. 5. Nehnutelnosti. United Classifieds s.r.o. [Online]. Available: www.nehnuteľnosti.sk. 6. INSUL - predicting sound insulation. [Online]. Available: https://www.insul.co.nz. 7. Thurstone, L. L. A law of comparative judgement. Psychological Review , 34 , 278–286 (1927). 8. Pierce, J. W., Gray, J. R., Simpson, S., MacAskill, M. R., Höchenberger, R., Sogo, H, Kastman, E. & Lindeløv J. Psychopy3. [Online]. Available: https://www.psychopy.org/index.html. 9. Pavlovia. [Online]. Available: https://pavlovia.org/docs/home/about. worm 2022 Previous Paper 736 of 769 Next