The social acceptance of mass produced residential buildings among Hungarian young adults

ABSTRACT Off-site manufacturing may play a significant role in improving both process and output productivity in the construction industry and achieving UN Sustainability Development Goals 11 and 13. However, uniform mass production can lead to a monotonous built environment that may be alienating to people. Past examples show that such buildings and neighbourhoods can rapidly become obsolete, which makes them unsustainable. In this study, two online surveys were conducted with over 100 participants focusing on prejudices towards mass manufactured residential buildings and diversity requirements for homes among Hungarian young adults. In contrast to previous studies, respondents evaluated example scenarios with standardised buildings not as observers but by imagining how they would feel or act in situations involving these buildings. The first survey examined the bias of respondents towards mass manufactured buildings with Likert, Thurstone, and semantic differential scales. The second examined how respondents would feel if they visited a home similar to their own by varying visiting frequency and degree of similarity. T-tests showed that prejudices were positive, except concerning mass manufactured buildings’ diversity, and respondents were only negative about having to see identical dwellings to theirs on a regular basis. The results suggest that in Hungary, non-uniform but standardised residential buildings can serve as a suitable tool to provide affordable housing to young adults affected by the housing crisis.

emissions (World Green Building Council, 2019). This has to be reduced radically to meet the expectations of climate change agreements (United Nations, 2016). In addition, low productivity rates in the sector hold back the economies of developed countries (McKinsey Global Institute, 2017), which also motivates significant changes.
Off-site manufacturing may contribute to achieving a more sustainable and efficient construction sector. Standardisation and mass manufacturing residential buildings could have many advantages from the perspective of the economy and the environment (Kamali & Hewage, 2016;Li, Al-Hussein, Lei, & Ajweh, 2013). With off-site manufacturing, problems that affect most developed countries, including housing crises, shortages of skilled labour, and low productivity rates, could be effectively mitigated (Construction Industry Training Board, 2018;Jaillon & Poon, 2009;McKinsey Global Institute, 2017). Moreover, by applying off-site construction techniques, not only can the amount of construction waste be reduced, but other strategies that will help to reduce further embodied carbon, including topology optimisation and design for reuse (Orr, Copping, Drewniok, Emmitt, & Ibell, 2018), could be implemented.
On the other hand, multiple challenges are associated with prefabrication, including high capital costs, which makes it hard to enter the market; lack of skills and knowledge for new design and construction processes (Chen & Samarasinghe, 2020;Lu et al., 2020); and environmental and management issues associated with transport and logistics (Shih, Huang, Liu, Wu, & Wang, 2005). In addition, for many homebuyers, the term 'standardised building' still evokes images of either temporary or cheap and low-quality housing solutions, despite the existing counterexamples, like Le Havre (UNESCO World Heritage Centre, 2005) and Zlín (Kasper & Kasperova, 2018). Nadim and Goulding (2011) also found that unsuccessful past examples pull back the EU construction industry from embracing offsite production. Moreover, Kaasalainen and Huuhka (2016) showed that off-site manufacturing could lead to predictable design solutions and unintentional standardisation even without explicitly standardising layouts. Such monotonous living environments have proved to alienate people in the long run (Kronenburg, 2013). Therefore, despite the numerous potential benefits of mass manufacturing residential buildings, questions remain about the social acceptance of these buildings and neighbourhoods. These questions include how willing people are to compromise on uniqueness and whether they want to live in mass produced buildings at all. This paper attempts to answer these questions by presenting results from two online surveys that investigated prejudices towards mass manufacturing buildings and the social acceptability of standardised homes among Hungarian young adults.
In Hungary, young adults are particularly affected by the current housing crisis (Habitat for Humanity Hungary, 2019), forming a well-defined target group for future mass manufactured housing solutions. By investigating their main needs, particularly considering the required uniqueness of homes, the findings of this paper help to limit the solution space of mass produced housing solutions to those that consider customer satisfaction. These findings can also serve as a reference for authorities and developers to ensure that new buildings will keep their values in the long run.
As part of this study, only psychological factors are considered, and the effects of potential external controls (e.g. affordability and availability) are excluded. Therefore, the classic framework, the Theory of Reasoned Action (TRA), will be used to model social acceptance, which assumes that actual behaviour is predicted by behavioural intentions and subjective norms (Fishbein & Ajzen, 2010).

Background and literature
The background and the related literature will be sorted into three different main topics: a short overview of past examples of standardised residential buildings, how mass customisation could change the sector, and a summary of past research related to the preferred complexity of the built environment.

History of standardised residential buildings
The earliest examples of prefabricated residential buildings were wooden framed houses. One of the first pioneers of this industry was Manning, whose 'Portable Colonial Cottages' were transported in dozens to Australia from England at the end of the nineteenth century (Herbert, 1972). In the United States, Sears Roebuck and Company sold more than 100,000 prefabricated wooden buildings that could be purchased from type catalogues and assembled by the clients themselves (Kronenburg, 2013).
The idea of standardising and industrialising the construction of buildings became increasingly popular at the beginning of the twentieth century (Gann, 1996) when architects like Le Corbusier and Buckminster Fuller were inspired by how effectively the automotive industry implemented mechanised fabrication and related technological advancements (Cobbers & Jahn, 2014). However, it was the housing crisis after World War II that made it financially viable to mass produce buildings (Kronenburg, 2013) with a newfound efficacy. The mass produced buildings of the era are extremely varied. Many countries, including the United Kingdom, started repurposing former factories producing weapons or aircrafts to manufacture temporary buildings. There are numerous examples of such buildings, like the AIROH and the Uni-Seco (Blanchet & Zhuravlyova, 2018). Prefabrication also became useful in creating affordable permanent buildings; for example, prefabricated components were used in the famous BISF houses and the Orlits (Blanchet & Zhuravlyova, 2018). In some other cases, standardisation and prefabrication were used to achieve the highest architectural quality. A prime example being Le Havre city centre which is currently a UNESCO World Heritage Site (UNESCO World Heritage Centre, 2005), rebuilt after World War II using prefabricated concrete components. As the aim was to create something beautiful that serves as a real substitute for the original urban texture, despite standardisation and prefabrication, the buildings present a great variety of architectural solutions.
In the 1950s, the Soviet Union bought the Camus building system's license, used in Le Havre, and developed it further to solve the extreme housing crisis resulting from WWII (Anson, Ko, & Lam, 2002). To create a large amount of housing in the shortest time possible, most ornaments and unique design solutions were removed (Isaac, Bock, & Stoliar, 2016;Panteleyeva, 2016). There is no doubt that the Soviet housing programme was one of the most significant housing projects of the twentieth centuryapproximately 170 million people still live in such homes (Meuser & Zadorin, 2015). However, a combination of too much standardisation, constructional deficiencies and high maintenance costs resulted in a negative prejudice toward mass produced buildings in the affected countries (Engel, 2019).

Mass customising residential buildings
Today, the housing sector is under pressure to provide enough and at the same time create buildings that seem attractive and unique (Leishman & Warren, 2006). As Leishman and Warren (2006) argue, these objectives can be met by increasing the use of standardised elements and facilitating customisation. Moreover, by applying a combination of standardised and customised elements, designers can ensure consumer satisfaction, cost-effectiveness, and high product quality (Pine, 1993).
The application of the aforementioned design strategies, often referred to as mass customisation, may soon become viable due to automation technologies, like single-task construction robots and service robot systems, emerging in the construction sector (Bock, 2015). According to Salvador, Martin De Holan, and Piller (2009), mass customisation requires three key steps: solution space development, robust process design, and choice navigation. Larsen, Lindhard, Brunoe, Nielsen, and Larsen (2019) give an overview of the state-of-the-art literature by defining four research areas related to this field. These are mass customisation, off-site construction, construction supply chains, and customer satisfaction. Customer satisfaction was examined, for example, by Forsythe (2015) and Torbica and Stroh (2001). They found that customer satisfaction significantly correlated with the quality and design of homes and the homebuilder companies' service. Franke, Schreier, and Kaiser (2010) showed that customisation helps in increasing personal attachment. This finding was supported by Tseng and Piller (2003), who argued that in contrast to standardised solutions, people were willing to pay more for customised products. However, as Hofman, Halman, and Ion (2006) added, this was not without limits. If the cost of customisation exceeded a specific limit, people preferred not to customise.
According to Berman (2002), it is a considerable sacrifice for homebuyers to choose mass customised solutions instead of the fairly flexible traditional architectural solutions. Moreover, despite the perceived customisation capability, off-site manufacturing processes may lead to predictable design solutions (Kaasalainen & Huuhka, 2016) and, therefore, unintentional uniformity.

Preferred complexity in the built environment
The preferred complexity of the environment has been investigated extensively by researchers since the 1960s. By showing pictures of various objects to participants, Wohlwill (1968) and Kaplan, Kaplan, and Wendt (1972) attempted to determine the relationship between rated complexity and self-rated preference. Wohlwill (1968) found an inverted U-shape relation between self-rated complexity and preference with pictures taken of the geographic environment and modern art. After a point, increasing complexity resulted in a decreasing preference. However, Kaplan et al. (1972) showed that the relationship was positive linear in natural images of the built and natural environment, suggesting that it was impossible to reach the tipping point with natural objects.
Since then, many researchers have been investigating how complexity could be defined and measured more rigorously. Rapoport and Hawkes (1970) defined complexity as 'detectable variations within an established system of expectations'. They suggested measuring it in terms of the rate of usable information per unit time and space, a metric that contained not only the quantity of information but the pattern of its structure. Stamps (2002) suggested that 'entropy', the measure of disorder in information, defined as aggregate architectural variation, would be an ideal candidate for this metric.
In contrast, Valtchanov and Ellard (2015) showed that the contour information of images, a complexity-like metric measured in fractal dimensions, was a strong predictor of preference. In their experiments, people preferred pictures in which the fractal dimension of the image was around the typical values found in images of nature. This was in line with the argument Salingaros (2010) made that the main reason why monotonous repetition disturbed people was that these buildings could not blend into the complex natural background. Similarly, Taleb (2012) also stated that the problem with modern architecture was that, as most top-down processes, it missed the complexity of nature and natural processes.
Most experimental research has been carried out by showing drawn images (Heath, Smith, & Lim, 2000;Stamps, 1994Stamps, , 1999Stamps, , 2002, rendered 3D-models (Lindal & Hartig, 2013), or photographs (Kaplan et al., 1972;Valtchanov & Ellard, 2015;Wohlwill, 1968) of imaginary or unfamiliar places. Consequentially, participants had to evaluate situations that did not affect them, so it can be assumed that the methods utilised lacked the respondents' emotional involvement. This could be a significant limitation because, as Brunson (1993) showed, personal attachment affects social acceptability. A good example of such an attitude is the 'not in my backyard' phenomenon in which people support a particular solution to an issue, but only when it does not affect them personally. Therefore, the results of these studies cannot be used to predict how people would evaluate the uniformity and limited diversity of mass produced buildings if they had to live amongst them. This paper aims to predict how society might react to the spread of mass manufactured buildings and determine what criteria people have about their homes' diversity and customisability.

Materials and methods
This study aims to investigate whether it is socially acceptable to build mass produced homes to deal with housing crises during the next decades in Hungary and how contemporary experimental design and construction techniques could be best implemented for this use case. Therefore, it was hypothesised that: MH. Hungarian adults are negative about mass produced buildings because of the uniformity associated with them.
This hypothesis has been detailed further to formulate such subhypotheses that were suitable for statistical analysis and focused on various aspects of this main issue.
H1. The majority of people are not positive about mass manufacturing residential buildings.
H2. The main problem people have with mass produced buildings is with regards to their presumed uniformity.
H3. People would pay less for an attractive, completely standardised mass manufactured building than a unique one, but they would pay the same price if the façades were different to each other.
H4. People would accept the theoretical possibility of owning a standardised dwelling. Still, the vast majority would not want to have their homes visually similar to those they regularly see.
Each hypothesis added more detail to our understanding of the general attitude and the possibilities of making mass produced buildings more attractive. In addition, by examining the main problem from multiple angles, the results of the investigations could also be cross-validated, and, therefore, the findings could be considered more robust. However, tackling all hypotheses together would have resulted in a survey that would have taken too long to fill out, so the hypotheses were sorted into two surveys. The first survey addressed the first three hypotheses, and the second one focused on the last hypothesis.
In these surveys, Hungarian young adults, between the ages of 18 and 40, answered various questions about their attitude towards buying mass manufactured buildings and living in environments with reduced diversity. Several equally weighted methods were used, including Thurstone scales, semantic differential scales, and imaginary scenario evaluations on Likert scales.
Despite its known limitations, opportunity sampling was used due to its relative easiness and cost-efficiency. The surveys were advertised on various online platforms, including the research project's webpage, mailing lists, and targeted advertisements on social media sites. Responses were collected for three weeks for both surveys. For Survey 1, the data collection took place between the 14th of May and the 4th of June 2020, and for Survey 2, between the 16th of March and the 6th of April 2020. As an incentive, the option to participate in a prize draw of a £50 Amazon voucher was offered to participants. All questions were optional, and responses were fully anonymised before the analysis. To evaluate and visualise the results, Jamovi and SPSS were used. Both surveys were subject to an ethical review by the University of Cambridge Engineering Department's Research Ethics Committee and received approval on the 3rd of December 2019.

Survey 1
The first survey consisted of background questions, Thurstone scales, 7-point semantic differential scales, and Likert scales. The purposes of using all these question types are summarised in the following paragraphs.
H1attitude toward the uniqueness of homes and mass manufactured homes Thurstone scales helped evaluate Hypothesis 1 by quantifying and measuring participants' attitudes toward mass manufacturing residential buildings and the uniqueness of homes.
First, two interval scales were created that could measure these two attitudes. These scales contained 11 sentences about each examined attitude. These statements were chosen to be equally distanced from each other on a scale where the first sentence was the most negative and the 11 th was the most positive statement describing the attitude.
To create such a scale, one hundred sentences were evaluated from 1 to 11 by a small jury of 12 based on how favourable the statements were to the examined concept. The jury was chosen from volunteers, following Holt and Walker (2009), representing all age groups and both genders equally (50% males and females, 50% under and over 40 years). After the jury finished the evaluation, the statements were assigned to 11 groups based on the evaluation's median. Group 1 contained the least favourable statements to the concept (which had their median = 1), and Group 11 contained the most favourable ones (which had their median = 10.5 or 11). To find those statements that represented best their groups and to eliminate the effect of outliers, only the range between the two interquartiles around the median was examined. The statements that the jury most agreed upon were the ones with the lowest interquartile range. From each group, the best sentence was put in the survey to create the interval scale.
In Survey 1, all these 11 statements were listed, and respondents were asked to indicate whether they agreed with them or not. Figure 1 shows the statements and the proportion of the responses.
Each score was calculated as the sum of the statements' level they agreed to divided by the number of sentences they agreed to. The distribution of the respondents' scores described the overall attitude of people towards the examined concepts. This was analysed with one-sample Student's t-tests to understand whether the two variables, the Attitude toward mass manufacturing homes and the Attitude toward the uniqueness of homes, were significantly positive or not.

H2prejudice against mass manufactured buildings
Hypothesis 2 was examined by using 7-point semantic differential scales (Rózsa, Nagybányai, Nagy, & Oláh, 2006). Respondents were asked to think of how they would imagine mass manufactured residential buildings if they had heard that one of the largest contractors in their country had started producing them. With the pairs of adjectives shown in Table 1, the prejudices against mass manufactured buildings were measured. One-sample Student's t-tests were used to decide whether the bias was positive or not.

H3the value of uniqueness
Hypothesis 3 was investigated with 7-point Likert scales created of approximated price ranges to measure how much money people would pay for an attractive flat in a mass manufactured building compared to an apartment in a traditional building. The items of the scale were the following: Much less (< 70%), Significantly less (≈ 80%), A little less (≈ 90%), The same price (≈ 100%), A little more (≈ 110%), Significantly more (≈ 120%), and Much more (> 130%). The question was asked two times. In the first version, the mass manufactured building was completely identical to the previously built ones. In the second question, the building's façade could be customised. One-sample Student's ttests were used to analyse the estimated value of mass manufactured homes, and pairwise Student's t-tests were used to show whether custom façades would significantly increase the estimated value.

Survey 2
H4the required uniqueness of homes Hypothesis 4 was studied in detail with the second survey. This survey examined the respondents' preferences for the uniqueness of their own dwelling by asking how they would feel if they had to see the interior/exterior of a home similar to theirs. The level of similarity and visiting frequency were varied from question to question.
This survey consisted of three blocks: first, there were the same background questions as in the first survey; second, questions about the participants' preferences about the interior of their homes; and lastly, similar questions about the exterior of their homes. Between the interior and the exterior block, the possibility to terminate the survey was offered to respondents.
In the interior block, respondents were asked to imagine their own homes and another building with a described level of similarity and answer questions like, 'How would you feel if you had to visit the other flat at a given frequency?'. For each similarity level, the respondents had to imagine how they would feel if they had to visit the other building regularly (REG), if they had to visit it a few times (FEW), or if they just knew that it existed without seeing it (NEV). The same questions were asked repeatedly, but the questions' order was randomised to mitigate the effect of fatigue or getting practised (Holt & Walker, 2009). The evaluation happened on a 7-point Likert scale with the following levels: −3: extremely negative, −2: rather negative, −1: slightly negative, 0: neutral, 1: slightly positive, 2: rather positive, 3: extremely positive.
To describe the level of similarity, a definition and two images were used. These pictures, shown in Figure 2, were taken of a rendered 3D model developed for this purpose to minimise unintentional differences. Before answering the questions, respondents had to indicate whether they had understood the definition or not by choosing from 'Yes', 'Maybe' and 'No'. Those answers in which participants didn't choose the 'Yes' option were excluded from the analysis.
The specified similarity levels were the following: . Almost identical flats (AIF), where only movable furniture was different, but coverings and places of walls and doors were the same. . Flats with the same room arrangement (SRA), where coverings and furniture were different, but the places of walls and openings were the same. . Similar flats (SIM), where furniture, coverings and the places of interior walls and doors were different, but the adjacent walls and openings remained the same. The structure of the exterior block was similar. Before each set of questions, four images and a definition were used to explain the terms. Pictures of different building types were taken from Google Street View to help respondents imagine how a building with the specified level of similarity would look compared to their primary residence. The described levels of similarity were the following: . Completely identical buildings (CIB), where massing, openings, ornaments, materials, and colours were the same . Almost identical buildings (AIB), where massing, openings, ornaments were the same, but materials and colours were different . Quite similar buildings (QSB), where massing, openings and ornaments were different, but proportions, colours, and materials were the same.
Respondents were asked to indicate their attitude towards seeing the other building multiple times daily (DLY), regularly (REG), a few times (FEW), or just knowing that it exists (NEV) on the same Likert scale as before.

Results and analysis
For the first survey, 100 responses were collected in total. 49% of respondents were male, 50% female, and 1% preferred not to answer this question. 60% of participants had an AEC industryrelated professional or educational background. 59% of respondents were from cities, and 41% were from smaller settlements. The majority, 66% of the participants lived in flats, and 33% in houses. In the second survey, 108 people completed the interior block, and 101 people finished the exterior block. Gender proportion slightly deviated from 50/50: 47% of the respondents were males, 52% female, and 1% preferred not to answer this question. 55% of participants had a relevant background. The majority, 68%, were from cities with over a million citizens, 30% were from smaller settlements, and 2% did not answer. 62% were from flats, 31% from houses, and the accommodation of 7% could not be classified as one of these.

Power analysis
Before the detailed analysis of the results, a power analysis was carried out to determine whether the planned statistical tools would have the necessary power or further data collection would be required. According to the power analysis for a two-tailed one-sample or a paired sample t-test, a minimum of N = 90 respondents were required to have the desired power π > 0.8, with a minimally interesting effect size d = 0.3 and α = 0.05. To ensure the validity of the findings, only those results which had an effect size large enough to be detected with the required power will be reported.

Reliability analysis
Reliability analysis was carried out to ensure that the different questions in Survey 1 measured the same underlying variable. In this analysis, it was found that the results for the different prejudices were coherent with the responses for the attitude toward mass manufacturing homes and the price people would pay for mass manufactured buildings (McDonald's ω = 0.811). However, by removing variables considering the price, ω has risen to 0.829, which indicated that the price people would pay for these buildings was also affected by subjective norms.

Survey 1
H1attitude toward the uniqueness of homes and mass manufactured homes One sample Student's t-tests, testing equality with the central value, 6.0, showed that both the Attitude toward the uniqueness of homes, t (97) = 4.02, p = .000; d = .40, and the Attitude toward mass manufacturing homes, t (96) = 5.47, p = .000; d = .55, were significantly positive. The effect size in the first test was small, and in the second, it was medium, based on Cohen (1988)'s convention. These results showed that respondents were positive about mass manufacturing residential buildings, although they would not give up their homes' uniqueness. In conclusion, the first hypothesis, that people are against mass manufacturing residential buildings, turned out to be false, as the general attitude was significantly positive.

H2prejudice against mass manufactured buildings
By using the results of the 7-point semantic differential scale questions, one-sample Student's t-tests were carried out to determine whether the prejudices towards mass manufacturing residential buildings were positive or not. It was assumed that with approximately 100 responses, the test was robust to non-normality, and the variables could be considered continuous ones.
The results are summarised in Table 2. There were statistically significant (p < 0.001) differences from 4.0, the central value, in all cases, except Price and Attractiveness. The distribution of the Price variable was found to be bimodal, showing that people have a rather strong but either positive or negative view on this feature. It was assumed that the non-significance of the test considering Attractiveness meant that people had no clear prejudice about this feature. All other prejudices turned out to be positive except the one towards Diversity.
The results support the second hypothesis that people's main problem with mass produced residential buildings is their presumed uniformity. Moreover, the results indicate that this is the only problem people have with these buildings.

H3the value of uniqueness
It was investigated how much people would pay for a good mass manufactured building compared to a traditional one if they knew that it had already been constructed a few times, and how their approach would change if the façades were different. Student's t-tests were carried out with the same assumptions as before to determine if the preferred prices were significantly different from the traditional building's worth. The price was significantly less than 100% (M = 94.9, SD = 11.50) when the façades were the same (t (99) = −4.43, p = .000; d = −.443) and also less (M = 97.0, SD = 9.27) when the façades were different (t (99) = −3.24, p = .002; d = −.324). The effect size exceeded Cohen (1988)'s convention, in both tests for a small (0.5 > |d| > 0.2) effect size. A paired samples t-test showed that people were willing to pay significantly more when they learned that it was possible to have the mass manufactured building with a custom façade, t (99) = 2.30, p = .024; d = .230. When the building had unified façades, only 54% of the respondents would have paid the same or a higher price, with this proportion increasing to 75% if it had a custom façade.
Therefore, based on this analysis, we can conclude that Hypothesis 3 was only partly true. People would pay less for a mass manufactured residential building than for a traditional one, and they would pay less still, although just by a small difference, even if these buildings had a custom façade. However, such a small difference (3%) in preferred price means that, in practice, mass manufactured buildings with custom façades are likely to be sellable at the same price as traditional buildings.

Survey 2
H4the required uniqueness of homes As data were not normally distributed across all variables, one-sample Wilcoxon signed-rank tests were used to analyse the results of the second survey. The Wilcoxon signed-rank test is a non-parametric test suitable for ordinal data. It was assumed that visiting another flat or seeing another building was an entirely neutral experience. Thus, the null hypothesis was that the sample median equalled zero. The effect size (r) was calculated as the standardised test statistics (Z) divided by the square root of the sample size, and it was evaluated based on Bartz (1988).
Thus, by summarising the results in Table 3, we can state that it is clearly a negative experience to see completely identical buildings daily (p < .001, r = .51, moderate effect) or regularly (p < .001, r = .26, low effect). It is also found to be a negative experience to see almost identical buildings daily (p = .042, r = −.08, very low effect), and visit an almost identical flat regularly (p = .012, r = −.14, very low effect). Still, these results should be treated with caution because of the almost negligible effect size and the rather high p values. All other tested scenarios resulted in a significantly positive relationship or a statistically insignificant difference from the neutral case. As all distributions were unimodal, in those cases where the relationship was statistically insignificant, it was assumed that the attitude could be considered neutral. Mainly those scenarios were found to be positive where the other dwelling was seen only a few times, like in cases of almost identical buildings (p < .001, r = 0.40, moderate effect), quite similar buildings (p < .001, r = 0.47, moderate effect), flats with similar room arrangements (p < .001, r = .34, low effect), and similar flats (p < .001, r = .48, moderate effect). As for similar flats people were positive about seeing them even regularly (p < .001, r = .34, low effect).
Based on these results, people are completely neutral about the existence of buildings and flats with any of the described similarity levels if they just know that these buildings exist but do not see them. Even in cases of completely identical buildings (CIB NEV ) and almost identical flats (AIF NEV ), less than 10% of the respondents had a strong (i.e. rather or extremely) negative opinion.
In summary, the fourth hypothesis was supported. People would not mind standardisation if they did not have to face it frequently, and they prefer when rarely seen buildings resemble their own. However, people were the most positive about those similarity levels that had the least binding requirements. In the exterior block, it was QSB where massing, openings, and ornaments were different, but proportions, colours, and materials were the same. In the interior block, this was also the level with the least binding similarity requirement, SIM, where furniture, coverings and the places of interior walls and doors were different, but the adjacent walls and openings remained the same.

Discussion
This study attempted to further nuance the existing knowledge about the social acceptability of reduced complexity environments, especially by focusing on the attitude toward mass produced buildings and the underlying assumptions residents may have concerning these buildings. The practical aim of the study was to collect data on whether it is socially acceptable to use mass produced buildings to deal with housing crises or not and, if it is, then what requirements people have towards them. For this purpose, over a hundred Hungarian young adults, who are particularly affected by the current housing shortage, were questioned about their preferences and intentions.
It was hypothesised that this group would be negative about mass produced buildings because of the uniformity associated with them. This hypothesis was further detailed with the help of four subhypotheses. First, that people are not positive about mass manufacturing residential buildings (H1). Second, that the main problem people have with them is with regards to their presumed uniformity (H2). Third, that people would only pay the same price for a standardised mass manufactured building as for a unique one if the façades were custom (H3). Lastly, that people would accept the theoretical possibility of owning a standardised dwelling but would not want to have their homes visually similar to those they regularly see (H4).
Based on the results of the examinations of H1, the general attitude toward mass produced buildings was found to be positive. The investigations of H2 showed that the only threat to the popularity of these buildings is that they are expected to be monotonous. However, the investigations of H3 revealed that despite the positive attitude, people would pay less for a mass manufactured building than for a traditional one, even if the façade was custom. This is seemingly contradictory with the results of H1, especially because people would not expect mass produced buildings to be cheap based on the results of H2. A possible explanation could be that first, in the current Hungarian market, it is impossible to imagine a decent accommodation being cheap for the average citizen, and second, that although Hungarians would be happy to see mass-produced buildings on the market, they still think that traditional buildings are a bit more valuable. This is also underpinned by the findings of Berman (2002) and the results of H1 that showed that people are indeed positive about the uniqueness of their homes.
Another explanation, in line with the findings of Nadim and Goulding (2011), could be that social norms based on negative historical experience, still affect how much would people pay for such buildings. Even though based on H1 it seems, that young adults in Hungary no longer have aversion towards industrialised construction. In addition, H3 also showed that with custom façades, the difference between the price people would pay for a manufactured building and a traditional one was almost irrelevant from the market's perspective.
The investigation of H4 also revealed that the standardisation of residential buildings is theoretically acceptable for the vast majority of people, provided that it could incorporate all the functional needs and that the façades of the buildings remained varied. This is in line with the results of H1, H2 and H3. In addition, respondents preferred some similarity both externally and internally if they had to see the other building only a few times. This suggests that familiarity can be a distinct source of preference, which is in line with the findings of Vartanian, Navarrete, Palumbo, and Chatterjee (2021).
The results suggest that the variety people want from individual buildings could be achieved by using different materials and colours as finishes and letting residents personalise their home's furnishing. This finding indicates that standardising building structures and even layouts can be socially acceptable among Hungarian young adults.
Based on these findings, it seems that Hungarian adults are positive about mass produced buildings despite the uniformity associated with them. Therefore, both MH and the null hypothesis of MH, namely that Hungarian young adults are indifferent to mass produced buildings independently of the rate of uniformity, can be rejected.

Conclusions and future work
In conclusion, the presented investigations contribute to the existing literature on the preferred environmental complexity by showing that studying intentions and prejudices can reveal complex attitudes. Being aware of these attitudes is essential to predict real-life decision-making processes. For architectural research, the main outcome of this paper is that based on our results, the former aversion toward industrialised housing solutions no longer exists among young adults in Hungary. Regarding the practical implications for industry and practitioners, this also means that non-uniform standardisation can be a suitable tool to provide affordable housing to young adults affected by the housing crisis. Moreover, it is a socially acceptable practice for the construction industry to move towards a greener and more efficient way of delivering housing, even at the cost of compromising uniqueness.
This study had a couple of possible limitations which should be considered. First, the study used a convenience sample, and such samples can be biased in multiple ways. Second, this paper focused on mainly the building level and did not attempt to examine how people would accept neighbourhoods or streetscapes created partially or entirely of mass produced buildings.
Therefore, future investigations could support current work by using a truly random sample and by extending the investigations to the urban scale, for example, by examining the preferred variety of streetscapes and comparing the results of Stamps (2002) and Lindal and Hartig (2013). Future work could also cover how different nations and cultures differ in their attitudes toward mass produced buildings and what they perceive as attractively unique and attractively monotonous in the built environment.