NUS RE3901 Individual Assignment 2009
National University of Singapore
RE3901 Behavioural Studies in Real Estate
Academic Year 09/10 Semester 1
Housing preferences and attribute importance among low-income consumers in Saudi Arabia
A review and evaluation
Riding on a strong oil industry, Saudi Arabia has seen strong growth in its economy with a GDP growth rate of 4.2% in 2008 (Saudi Arabia, 2009). Coupled with this growth, the housing market in Saudi Arabia is also growing quickly (Al-Sayari, 2007). This demand is causing housing costs to increase at a rate that is higher than that of incomes (Salama & Alshuwaikhat, 2006), making the housing needs of the lower income groups in Saudi Arabia a critical issue that needs to be addressed.
The selected paper (Opoku & Abdul-Muhmin, 2009) focuses on the real estate consumption activity of the lower income population in Saudi Arabia in terms of their housing preference. Emphasis is put on the housing type that they prefer and the tenure options chosen. In doing so, the researchers hope to gain a better understanding of the needs of the lower income consumers who make up 60% of the Saudi population (CDSI, 2007). This information is useful for both the Saudi government in its government housing programs and also real estate developers who are targeting this large market.
Similarly in land scarce Singapore, housing demand is high and prices have been rising. At the same time, the income gap has been widening, leading to a growing concern that the lower income segments of Singapore’s population will find it hard to keep up with the growth (Lee, 2007). Therefore, the results of this paper could also be useful for policy makers in Singapore in understanding the general preferences of lower income consumers and design their housing policies to cater to their needs.
Both of these reasons make this paper a relevant and important study that needs to be further analysed and evaluated.
2. Behavioural aspects
The paper focuses on the psychological and sociological perspectives of behavioural studies. The researchers first identified the preferences of the low income consumers before attempting to link the results to the perception, motivation and reasoning aspects of their psychology. From a sociological perspective, the behaviours exhibited by the consumers are also explained using social factors such as family, religion and culture of Saudi society.
2.1 Psychological perspective
From a psychological perspective, this paper offers valuable insights into the mindsets of the lower income groups in Saudi Arabia. We are able to learn about the general preferences of the target group of consumers and also the specific factors that are attractive to them in making the decision to purchase a home. From this, we can then observe the results for any distinguishing behaviour that characterises the lower income consumers.
The investigation into their preferences is also augmented with explanations of possible reasons why the results are so by the native researchers. Thus, we are able to better understand why the lower income consumers behave in the particular way with regards to their perceptions of certain attributes of a home and the reasoning behind their choices. From this, we are able to link their external responses to the internal mental processes unique to this group of consumers.
2.2 Sociological perspective
The research also looks at the sociological perspective of behavioural studies by investigating the social factors influencing the preferences of an individual in the lower income group. These social factors include family traditions, religion and societal culture in Saudi Arabia.
In the paper, the factors of family, religion and culture are used to explain the results found. We can thus learn about how the cultural environment of Saudi Arabia affects the behaviour of the local consumers and how the individual interacts as a member of their social groups.
3. Objectives of paper
The paper explores the housing preferences of low income consumers in Saudi Arabia. The focus is mainly on investigating their preference for alternative dwelling types like small houses, duplex houses and apartments. Together with the housing type preferred, the tenure options of either buying or renting for each of the housing types are also looked into.
A second main focus of the research is regarding the relative importance of various house purchase factors. Attributes like the design of the building, price and proximity to relatives are treated in this case as discrete factors in the purchase decision of the consumer. The paper aims to identify any distinct trends in the preferences of individuals belonging to the low income group in order to arrive at a generalisation of the group as a whole.
Lastly, the paper also seeks to categorise the responses according to socio-demographic factors such as gender, marital status and family size. This is done in order to investigate if such factors have an impact on the housing decision criteria of the individual.
The research methodology is mostly quantitative and can be categorised into three main areas. The first of which is the design of the measures that were used to gauge the preferences of the target consumers. The second is the methods through which data was collected. Lastly, the third is the methods used in the analysis of the data.
4.1 Design of measures
In designing the method of measuring the housing type preference of the lower income consumers, real property options were taken from a real estate developer that had sponsored the research. The options included different dwelling types, tenure options, neighbourhood environments and prices. The respondents were then asked to choose from these property profiles based on their current and expected future financial and family situation. This is a good way of assessing the housing preference of the consumers as they feature real property options that a normal consumer would consider in the market.
The method of assessing the relative importance of various housing attributes is through a rating of individual factors on a scale of one to five. The factors were selected through a study of current literature, interviews and focus group discussions. The interviews and focus group discussions were done with students of one of the authors and with students of an MBA class taught by the same author. These students are stated to be mostly working professionals and do not belong to the lower income group. Thus the perspectives of the students may be different from those of the lower income consumers, leading to attributes that may be left out or over emphasised. The factors generated were then reviewed with the participation of the same students, once again exposing the research design to potential bias.
In the research, the stated preference methodology was used. The respondents were asked to make a hypothetical choice and this is then taken to be an accurate representation of their preferences. This is in contrast to the revealed preference methodology which is possibly more accurate as it makes observations from past data that has already been realised. In this case this data would be housing purchase choices of past purchasers. However, as stated in the paper, actual market data is not available in Saudi Arabia and that is the reason why the stated preference methodology was adopted.
4.2 Collection of data
The data for the research was collected through a self administered questionnaire survey distributed to the respondents. The advantage of using questionnaires is that it allows for clear results that are quantifiable and verifiable using mathematical tests. Questionnaires are also a quick way of gathering a large number of responses in a relatively short period of time.
However, there are a few inadequacies in the method of data collection. As stated in the paper, the data were collected as part of a company sponsored marketing research class project. The data were solely gathered through the students’ distribution of the questionnaire to their family and relatives. This convenience sample is a poor sampling of the low income consumer population as the questionnaires were distributed without regard to the respondents’ income. The convenience sampling method and the following sub-sampling approach also made it impossible to derive a response rate to the survey and thus an estimate of the non-response bias was unavailable.
Additional problems in the sampling method include the small sample size and low diversity in respondents. The sample group was limited to the small number of students and their relatives, leading to a final sample of only 557 responses available for analysis. The non-probability method of gathering responses from students and their relatives only also leads to a sample group of respondents that are not diverse. They are mostly young respondents in their twenties and a large proportion of them have a highest education level of a university degree which may not be representative of a low income consumer group.
A possible source of error in the survey is respondent error. As the respondents are all students and their families, they may be influenced by the knowledge that the results will be used in a publicly available research paper and indicate choices that reflect better on themselves. By deliberately giving false responses, the results may be skewed or inaccurate.
Another limitation of the data gathering process is due to the strict gender separation in Saudi public life. Because females are difficult to reach for data collection in Saudi Arabia, only 13 percent of the respondents were female. This leads to a possible bias towards the preferences of male low income consumers and will not be reflective of the entire population as both the preferences of males and females need to be considered during a housing purchase decision.
4.3 Analysis of data
In the analysis of the data, cleansing and discarding of data was practiced to select respondents that fit the low income criteria of the study and also to remove unemployed respondents like full time students as they are not able to purchase a home yet. Out of the 1412 usable questionnaires collected, 815 of them met the low income criteria and finally, only 557 responses were selected for analysis. This exposes the research to risk of discarding important data as close to two-thirds of the raw data were removed from analysis. To prevent this, better sampling methods could have been practiced and respondents chosen more carefully beforehand.
Descriptive statistics were then calculated to illustrate the results. Percentages of respondents selecting each of the dwelling types available and the other factors were tabulated before conclusions were drawn from these figures. By using these descriptive statistics calculated from the data, the results are made easy to interpret and understand.
A variety of mathematical tests were also used to test the robustness of the results. These included the chi-square test and the t-test. This allowed the strength of the data and the corresponding results to be evaluated. However, how this evaluation was done was not explained and the numerical limits imposed on the results were also not given justification. For example, it was not explained why a t-score of 1.99 is considered significant while a score of 1.71 is not.
5. Recommendations to improve paper
A key weakness of the paper was that the sample size was too small and narrow. Therefore, in order to improve the research methodology I propose that the number of respondents should be increased to improve the accuracy of the results. Also, a probability sampling approach could be adopted. In particular, using a cluster sampling approach will be effective in gathering the opinions of a variety of respondents from different geographical locations in Saudi Arabia. This is in contrast to limiting the sample group to students and their families from a particular academic course.
Instead of approaching the research from a purely quantitative perspective, the study could have included qualitative research methods. Interviews of individual low income consumers could have revealed more detailed explanations of the reasons why such preferences were observed in the low income group instead of just speculation. Qualitative methodologies are generally preferred in the areas of sociology and psychology because they reflect the individual phenomenological perspective apparent in these topics. In addition, as the research is exploring a new area, a qualitative approach is recommended as a first source of information about the research issue.
Additionally, a conjoint analysis of both revealed and stated preferences could be used in this research as combining the two types of data improves the identification of important attributes to the research targets (Earnhart, 2002). Perhaps the revealed preference data that is lacking in Saudi Arabia could instead be adapted from data found from other Arab countries in order to implement this method.
6. Results of study
Results of the study can be grouped into two main areas. Firstly, the dwelling type and tenure preferences of the low income consumers were observed. Secondly, the relative importance of house attributes were analysed both generally and specifically to socio-demographic factors.
6.1 Dwelling type and tenure preferences
Of the three housing options provided to the respondents, 43% indicated a preference for the small house. 29.7% and 27.3% of the respondents preferred the apartment housing and duplex housing respectively. The authors offered the explanation that this might be due to the need for privacy in Saudi society as the small house could be walled and would thus offer reduced exposure to the public.
As for the tenure options, over two-thirds of the respondents (68.5%) prefer buying over renting. When the tenure results were categorised by their dwelling type, the observation was that most respondents that selected the small house or the duplex house preferred buying(over 80%), while those that selected the apartment house mostly preferred renting (65.9%). This indicates a very strong relationship between tenure preferences and dwelling type. The explanation could be that most households would see the purchase of a home property as an indicator of success so those who are able to afford purchasing instead of renting would likely do so.
In previous literature, it was hypothesized that socio-demographic factors like gender and martial status affected the purchase decision. However, only the income level was shown to affect the housing type preference of the respondents. The higher income consumers in the low income group are more likely to opt for a small house while the lowest income consumers in the group are more likely to opt for apartment house. This is consistent with the argument that consumers will spend within their means and choose the housing options that better fit their income level.
6.2 Importance of house purchase factors
When analysing the importance of the house purchase factors, the paper concluded that all ten factors were generally important to low income consumers with the exception of the factor of street location. Street location and exterior space were two factors that were ranked relatively low in importance to the respondents. The researchers explained that this might be due to the fact that the weather environment in Saudi Arabia is hot and not conducive for outdoor activities. Once again, the emphasis on privacy of the Saudi society is suggested to be another contributing factor to this.
Relatively however, the top three important factors are the financial factors, amount of private living space and the aesthetics of the house. The most relatively unimportant factors are that of proximity to relatives, outdoor space and street location. One should note that the top six important factors in the study actually have very small differences in their rating and could possibly be taken as equivalent in importance.
Socio-demographic factors were observed to affect the preferences of house purchase factors more than that of dwelling type and tenure preference. Females seemed to find the attributes of interior layout, private living space, aesthetics and exterior space more important than males. In addition, for married respondents, they seemed to consider both the local environment and proximity to relatives to be more important than when compare to singles. However, there were no differences in attribute preference found for families with children and families without children.
7. Future research
As stated in the paper, an area which has potential for future research is the analysis of the influences of men and women during different stages of the housing purchase process. Factors such as which of the decisions are predominantly made by the male or female would be interesting to observe in a Saudi Arabia where most women are not in the workforce. This would be useful to determine the relative importance of the different preferences of men and women and thus help in deriving a numerical weight for each of the different preferences.
Also suggested by the authors is that further research should be done in terms of the details of the housing attributes that are important to the low income consumers. For example, the number of bedrooms was found to be an important consideration in the quality of private space in the home but the present study did not examine the specific number of bedrooms preferred. Future studies could address this area by possibly providing more detailed options in the questionnaires.
In addition, qualitative interviews could be used in future studies to supplement the quantitative analysis of the data. This will allow for a more in depth look at the thinking process of the respondent and allow us to understand better the reasons for any decisions or preferences made. Any of the additional details of interest regarding the preferences could also be easily queried during these interviews and would serve to address the weakness of the current study mentioned above.
In the local context, it is also possible to repeat the study in Singapore and neighbouring countries to observe if there are any similarities between the low income consumers in different geographical regions. Possible results could include similar emphasis on financial concerns of housing purchase decisions. This information will be useful in Singapore for government policy makers to better tailor their strategies according to the needs of the lower income groups in a bid to help them achieve a higher standard of living in Singapore.
The research explores a new area that has not been studied before and has made a few interesting observations regarding the housing preferences of low income consumers in Saudi Arabia. As the research field is still new, there is a lack of historical data to back up most of the suggestions in the paper. However, the quantitative methodology adopted in the study allows convincing conclusions to be drawn from the research data alone.
While the findings presented in this paper are interesting, there exist several areas in which the study could have gone into increased detail, an example being the overly general housing attributes. The methodology of the paper could have been improved as well by incorporating more qualitative approaches that may yield more substantial input from the respondents. Conjoint analysis using another similar country’s housing data could also improve the accuracy of the results.
In conclusion, the area of research being explored in this study is of importance to many parties such as government policy makers and also private developer firms from all countries. Effort should thus be made to further the investigation into the housing preferences of lower income consumers in order to more effectively cater to their housing needs.
Opoku, R. A., Abdul-Muhmin, A. G. (2009). Housing preferences and attribute importance among low-income consumers in Saudi Arabia, Habitat International, 1-9.
Al-Sayari, H. (2007). Housing Finance in the Kingdom of Saudi Arabia. Conference on Housing Finance in the Kingdom (14, 1-2). Riyadh: BIS Review.
Earnhart, D. (2002). Combining revealed and stated data to examine housing decisions using discrete choice analysis. Journal of urban economics, 51, 143-169.
Central Department of Statistics and Information (CDSI). (2007). Distribution of disposable income in Saudi Arabia. Riyadh: Ministry of Planning and National Economy, CDSI.
Lee, H. L. (2007, August 19). National Day Rally 2007. Retrieved November 5, 2009, from Prime Minister’s Office: http://www.pmo.gov.sg/News/Messages/National+Day+Rally+2007+(English).htm
Salama, A. M. & Alshuwaikhat, H. (2006). A trans-discipllinary approach for a comprehensive understanding of sustainable affordable housing. Global Built Environment Review (GBER) , 5(3), 35-50.
Saudi Arabia. (2009, October 28). Retrieved November 5, 2009, from CIA – The World Factbook: https://www.cia.gov/library/publications/the-world-factbook/geos/sa.html