The International Journal of Construction Management (2012) Vol. 12 No.

1, 1-21

AN APPRAISAL OF HOUSING SATISFACTION IN

SOUTH AFRICA LOW INCOME HOUSING SCHEME

Clinton O. AIGBAVBOA1 and Dr. Wellington D. THWALA2

Department of Construction Management & Quantity Surveying, Faculty of Engineering and

the Built environment, University of Johannesburg, Johannesburg, South Africa.
1
PhD Student. Email: [email protected]
2
Associate Professor. Email: [email protected]

Abstract
Post-occupancy evaluation techniques have been developed to provide a means for
evaluating occupant responses to changes in an environment and linking this response
to physical measures of that environment. POE has been used to evaluate the
performance of buildings after they have been built and occupied for some time. This
paper presents findings on the social and physical factors which influence residential
satisfaction in four different government Housing Subsidy locations in the Gauteng
Province of South Africa. Data obtained from the occupant survey were analyzed
through the use of descriptive statistics. Findings arising from the survey revealed that
the respondents were satisfied with their overall housing situation, but had complaints
about certain aspects of the housing unit. However, the respondents informed that
most of their housing needs were not being met. Also, a comparison is also made of
the perceived factors of dissatisfaction amongst the housing subsidy occupants. It is
recommended that a wider systematic coverage of the subject through investigation
and diagnostic POE and occupants’ need assessment should be carried out in housing
subsidy schemes in South Africa.

Keywords
Post occupancy evaluation, housing satisfaction, housing subsidy, Government policy

INTRODUCTION
The way a building functions when it is used is essential for both whether or not it is
regarded as a success and constitutes an asset for its owners and occupants (Lu, 1999).
A systematic evaluation of buildings in use is an effective way to produce this
knowledge in relation to the planning of new buildings and not least for the
development and change of existing buildings that are not satisfactory to the
occupants (Blakstad et al., 2010). However, there are many concepts, definitions and
methods that are relevant to a building’s quality, standard and condition. Most of
these are associated primarily with a building as a physical object and not with its
usability. An important approach to usability of building is that a building in itself has
no value, but has value only when it is used and satisfies the occupants.

Universally there are growing efforts to undertake the appraisal of occupied buildings
in response to the quest for more efficient buildings to meet occupants’ satisfaction
and sustainability challenges. This is because sustainable development is a primary
concern to the present world and it formed the main theme of the report “Our
2
Aigbavboa and Thwala

Common Future” produced by the Bruntland Commission for the United Nations
(World Commission on Environment and Development, 1987). The Bruntland
Commission report described sustainable development as “development that meets
the needs of the present generation without compromising the ability of the future
generations to meet their own needs”. The report also asserts that the environment is
where we live and development is what we all do in an attempt to improve our portion
within that abode, and as a result the two are inseparable. In all ways, the built and
natural environment exerts a huge influence on the quality of life of the inhabitants
(Akintoye, 2006). The core ideal behind this belief is to create an effective system of
resource distribution and utilization with a long-term perspective in mind (Aribigbola,
2006).

The potential of building performance studies extend beyond the benefits for
improvement to a specific building under investigation. It probes outcomes and makes
recommendations that open up opportunities to enable transfer of knowledge in future
projects (Lackney, 2001; Zimring, 2002; Lu et al., 2004; cited in Mastor et al., 2010).
An effective building appraisal study requires adoption of systematic procedures and
techniques, whereby the most commonly known is Post-Occupancy Evaluation (POE).
POE is different from other appraisal methods in that it emphasizes the needs of
building occupants (Preiser & Vischer, 2005). The strength of POE lies on its capacity
to promote the advancement of knowledge through lessons learned and feedback for a
better performance of the buildings. Past studies have established the importance of
POEs as determinants of crucial performance factors relating to sustainability such as
resource consumption, environmental conditions, and occupant satisfaction and
operator experiences. As a result there has been a firm call to make POE a mandatory
step in the design and commissioning of buildings, be it privately owned or in
subsidized low income housing schemes (Preiser & Vischer, 2005; cited in Issac et al.,
2009).

Low-cost housing provision has been a major focus of government in post-apartheid

urban South Africa, as the government attempts to address historical race-based
inequalities, poor municipal service provision and present-day rapid urbanization. The
South Africa White Paper on Housing formulated in 1994, which has undergone some
modification, prioritized the needs of the poor, has encouraged community
participation and the involvement of the private sector, and committed to deliver 1
million houses in five years (Jenkins, 1999); of which the delivery of 1 million houses
has since be surpassed. The African National Congress (ANC) Reconstruction and
Development Program document (RDP) of 1994, and the Constitution (The South
Africa Constitution, 1996) also commit to providing housing for the poor (the low-
income and disadvantage group). Since 1994, the low-cost housing program has
mostly involved building serviced townships on urban peripheries, which in itself
presents a myriad of environmental, social and political concerns. By the end of 2010,
government had approved 2.85 million houses, giving shelter to more than 14.0
million people, free of charge according to the Department of Human Settlement.
However, many problems with the process have become clear as the process has
unfolded. These include:

(i) New houses and infrastructure are of poor quality, and are rapidly
deteriorating and require maintenance;
 3
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

(ii) New houses and Human Settlement development continue to place the poor
and low-income blacks in ‘‘ghettos’’ on urban peripheries, far from jobs and
services;
(iii)Occupants dislike the model of housing used, and would prefer larger houses
(main model was first changed in 1998 when Department of Housing, now the
Department of Human Settlement increased the minimum size of new houses
to 30m2, and was further increased in 2004 during the launching of the
Breaking New Ground Policy to 40m2);
(iv) The dominant model of free-hold tenure inadequately deals with the dynamics
of poverty, and several categories of the poor, such as temporary workers and
many women, which would be better served by rental accommodation as
against giving of houses;
(v) Because of these problems, people often sell or rent out their public houses
allocated through the subsidy scheme, and move back to squatter or other
informal settlements closer to their economic activities; and
(vi) Environmental concerns regarding the new developments include increases in
vehicular traffic caused by urban sprawl and land use changes.

From the above, it is thus obvious that both the design and the performance of these
buildings have become major concerns, and thus appraisal through the use of POE
should be of interest since it provides a mechanism for feedback/feed-forward
processes to be conducted among occupants, designers and policy implementers and
the Department of Human Settlement. The motivation of the research is borne out by
the fact that since the inception of government housing subsidy scheme in South
Africa, there has not been any research on the POE of the subsidy occupants. The
study hopes to fill the gap in this area. The significance of the study is to inform the
Department of Human Settlement on the satisfaction level of the occupants since the
government has been actively providing housing in different types of development
projects in various locations in South Africa. Hence, it is important for the
government to assess whether or not these development projects have met the social
and physical needs of the users, particularly on the subsidized housing which has
since been implemented after the drafting and implementation of the new Housing
Policy Framework in South Africa. This is because an understanding of the factors
that determine housing satisfaction levels is fundamental to the formulation of any
successful housing policy (Lu, 1999). In this direction, user’s satisfaction could be a
useful indicator to measure the performance of housing development by relevant
stakeholders in the housing development.

The objective of this study is to identify the factors which influence housing
residential satisfaction among beneficiaries of government housing subsidy schemes.
The paper commences with an overview of the literature on this topic, and then
presents the results of the analysis and findings of the research. Finally, the paper
draws some conclusions and recommendation.

BUILDING PERFORMANCE APPRAISAL

Liu (1999) informs that the most fundamental objective of a living organism is that of
survival, however, there are other aspects of consideration such as spiritual,
psychological, social and economic survival. All forms of survival entail the
maintenance of a balance between the individual and the environment. Markus et al.
4
Aigbavboa and Thwala

(1972; cited in Liu, 1999) suggests that in the field of design, any activity or object is
considered to function as part of a system and consequently, interdependencies of the
systems and the dynamism of the environment must be emphasized. The assessments
of building performance are of value only if they are considered as part of some other
processes such as the constant maintenance of balance between the beneficiaries of
the housing units and the environment. Liu further emphasizes that the ‘building
performance’ concept is based on the assumption that a building is designed and built
to support and enhance the activities and goals of its occupants. There are different
approaches to building performance appraisal:

(i) Overall approach to find out factors, on both physical and social levels, which
affect housing residents’ satisfaction;
(ii) Development of performance criteria and grading tools;
(iii)Relationship of residential satisfaction with children’s accident risk according
to Garling and Garling (1990) spatial density, crowding and neighbourhood
characteristics;
(iv) Quality appraisal of the building design in terms of both function and cost.
Depending on the approach taken to satisfy a particular research purpose, building
appraisal can be done during the design stage as in value engineering, or after
completion of the building as in POE. Figure 1 below shows that, while short-term
benefit is derived from the contribution of the POE process to immediate problem
solving in current projects, medium-term benefit is drawn from the next building
cycle in which a potential link between satisfaction and behavior will bring
improvements to unsatisfactory environments. This should result in changes in
beneficiaries’ satisfaction and in the social behavior of occupants (Bonnes et al, 1991).

Occupants’ residential appraisal is a measure of the degree to which a housing

(quality) performance has met the occupants’ expectation in terms of benefits and
needs. At the conception of housing occupation, a consumer builds some expectations
on the performance of the desired housing, the benefits it will provide and the needs it
should fulfill. The judgment of these begins immediately after occupation, which in
turn determines his level of satisfaction/ dissatisfaction. Based on the foregoing, the
work of Bruning, Langenhop and Green (2004) considered housing satisfaction as the
gap that exists between residential needs and aspirations and the current residential
context. These may include residents’ appraisal of neighbourhood safety, ease of
access to areas of interest, the quality of other homes in the immediate area, the
desirability of the community, and friendliness/pleasantness of the people in the
immediate neighbourhood.

Figure 1: Building Performance (Source: Liu, 1999)

 5
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

HOUSING SATISFACTION: THEORETICAL ISSUES

Various studies on housing satisfaction have appraised housing provisions by dealing
with problems of occupant satisfaction. The bulk of the scholarships have not adopted
a comprehensive methodology and most only assess housing provision subjectively.
As such, measures of satisfaction have been met with criticism. The criticism as
outlined by Francescato et al. (1987, p. 48) include: reporting that residential
satisfaction tends to be uniformly high and therefore cannot be assumed to indicate
the “true” state of affairs; subjective measures of satisfaction do not correlate with
objective measures of context and behaviour, therefore they cannot be considered
valid measures of the objective reality; satisfaction with an object varies, for the same
individual or social group, with time and with personal and social norms and
expectations, thus it is too trivial an indicator on which to base action; satisfaction
tends to be higher where there is a lower beneficiaries’ awareness of “better”
alternatives, which was argued tends to reflect unenlightened assessments on which
policy and decisions should not be based; and focusing on satisfaction rather than
attacking “real” problems may result in sub-optimal environments.

Francescato et al. (1987) further provided answers to these criticisms and established
that while the criticisms point to limitations that should be taken into account when
interpreting results, they seem to warrant using the construct of satisfaction (Potter et
al., 2001). Additionally, Campbell et al. (1976) concluded that “exaggerated
skepticism of subjective responses is not warranted based on extensive consideration”
of (among others) the following: the reliability and validity of measures; the
comparison between objective and subjective indicators of well-being; the levels of
reality of domains being assessed, and the analytical intentions (Anderson &
Weidemann, 1997). It is important to be aware of these limitations; however, it is
clear that they do not prevent satisfaction from being a useful concept. There are
limitations to all research investigations; for example, there are always limitations to
the operationalization of abstract concepts. However, Kim (1997) states that the
criticisms in residential satisfaction point out the need for research that addresses
these criticisms, and illustrates the impact on theoretical models, and then proposes a
research direction with clear theoretical foundation. On the aspect of the methodology,
most of the previous studies used regression models, which is questionable because of
the ordinal nature of the dependent variables representing housing satisfaction.
However this present study used the simple descriptive statistics a method that has
also been criticized as not giving the true state of the occupant’s satisfaction. But it
should be noted that result of any descriptive statistics analysis gives an indication of
events in a typical evaluative situation, as will be found if regression models or an
ordered logit model was used.

The concept of housing satisfaction has been utilized in at least four different ways: as
a key predictor of an individual’s perception of general quality of life (Campbell et al,
1976); as an indicator of incipient residential mobility, and hence altered housing
demands and effected neighbourhood change (Speare, 1974; Varady, 1983); also as
an ad hoc evaluative measure for judging the success of housing developments
constructed by the private sector (Lansing et al., 1970); and also to assess residents’
perceptions of inadequacies in their current housing environment so as to direct
forthcoming private or public efforts to improve the status quo (Michelson, 1977;
Francescato et al., 1976). However, according to Amerigo and Argones (1990)
6
Aigbavboa and Thwala

quantitative studies in housing satisfaction can be divided into two distinct approaches.
Firstly, there are those studies in which residential satisfaction is considered as a
criterion of residential quality (Galster and Hesser, 1981; cited in Liu, 1999), the
objective of this kind of studies is to establish which factors determine the degree to
which the occupant is satisfied in the residential environment, which the present study
is based on; the other considers residential satisfaction not as a criterion but as a
predictor of behaviour. Using the latter approach, a low level of residential
satisfaction can predict behaviour as in moving house, or, in cases where this is not
possible, the adaptation of the housing to new needs as they arise, such as the carrying
out of home improvements (Liu, 1999). From the above, the studies which deal with
residential mobility and its consequences use residential satisfaction as a variable
predictor of behaviour. A study which combined the two approaches is the model
offered by Weidemann and Anderson (1982) in which residential satisfaction is
considered as an attitude, which was based on the conceptual framework developed
by the work of Fishbein and Ajzen (1975) in the theory of reasoned action. Bonnes et
al., (1991) informs that there has been a recent gradual shift of emphasis in research
away from the relationship between the individual and the physical characteristics of
the environment towards an approach defined as ‘contextual’ (Altman and Rogoff,
1987; Onibukun, 1974) which focuses on the relationship between the individual and
the socio-physical environment, in which the purpose is to study the problems arising
from this relationship in the contexts in which they occur.

Onibokun (1974) argues that the habitability of a house is influenced not only by the
engineering elements, but also by social, behavioural, cultural, and other elements in
the entire societal-environmental system. Thus, a dwelling that is adequate from the
engineering or from the design point of view may not necessarily be adequate or
satisfactory from the inhabitants' point of view. Onibokun (1974) concludes that the
house is only one link in a chain of factors which determine people's relative
satisfaction with their accommodation. Varady (1983) further argued that housing
satisfaction acted as an intermediary variable between background characteristics and
mobility behaviour. In the work of Lane and Kinsey (1980) they reported that housing
characteristics were more crucial determinants of housing satisfaction than
demographic characteristics of housing occupants.

A significant issue in most of the models of residential satisfaction is how the housing
attributes outlined in most of the past studies are measured. However, two types of
measurements are usually adopted, namely objective and subjective measures of
housing attributes, which are found in the literature (Francescato, 2002; Weidemann
and Anderson, 1982). Objective measures refer to the actual measurements, such as
the presence, the lack of, or quantities of attributes while subjective measures refer to
perceptions, emotions, attitudes and intentions towards the housing attributes. The
objective measures of the attributes of housing have been shown to be weaker
predictors than the subjective measures (Francescato et al., 1989; Weidemann and
Anderson, 1982). Finally, it has also been common, in measuring residential
satisfaction to use an index of highly correlated items rather than a single-item
variable of ‘how satisfied are you with your housing?’ In the model of satisfaction
conceptualized by Francescato et al. (1989), satisfaction was measured using an index
based on four questions which were:

(i) How satisfied are you with living here?

 7
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

(ii) How long do you want to live in this housing development?

(iii)If you move again would you like to live in another place like this?
(iv) Would you recommend this place to one of your friends if they were looking
for a place to live?

The reason for this was conceptual, because the authors had conceptualized
satisfaction as an attitude which has affective, cognitive and conative dimensions.
However, the reason given by other authors Carvalho et al., (1997; Weidemann and
Anderson, 1982) who also used such an index suggests that it increases the reliability
of the criterion since it would seem that an index is intrinsically better than a single
item. This study will be patterned according to the framework develop by Francescato
et al. (1989), and validated by Carvalho et al., (1997) and Wiedemann & Anderson
(1985). This paper reports on the factors which influence housing residential
satisfaction and factors of dissatisfaction among the beneficiaries of a housing subsidy
scheme in the Gauteng Province of South Africa. The approach adopted by the South
Africa government in delivery of housing and allocation of the subsidized house to its
citizens will be discussed in the next section.

THE SOUTH AFRICA GOVERNMENT HOUSING SUBSIDY

SYSTEM
The South African housing policy introduced soon after the first democratic elections
in 1994 was developed by the National Housing Forum (NHF) from 1992 to 1994 and
was concluded in the White Paper on Housing. The NHF negotiators consisted of a
wide range of stakeholders, including civic and labour organizations. During and
after the negotiation criticisms emerged about the predominance of the private sector
and big business at the proceedings, and the influence this had on the nature of the
policy adopted. According to Charlton (2004), the outcome of the consultations has
been seen by some as a compromise between “popular demands to deliver complete
houses for all, and a concern to spread housing benefits widely”. This has been
labeled the ‘width versus depth debate’ – the notion of using inadequate resources
extensively to provide housing benefit for as many people as possible, versus the
notion of providing a more robust, comprehensive and complete unit for fewer people
(Charlton, 2004). The central point that was struck in the NHF negotiation targeted
mass delivery, with a strong emphasis on land, tenure and services, but also included
a basic 'starter' house or 'top-structure' as well. It went beyond the pure site-and-
service approach of the Urban Foundation and Independent Development Trust in the
late 1980s and early 1990s policy which provided land and basic services only with
no house. The new policy corrected this former approach which disadvantaged many
blacks who could not afford to build a house because of little or no economic
empowerment cause by the apartheid rule which had segregated the blacks. The South
African housing programme has demonstrated its ability to deliver at the scale as
reported previously. However, Rust claims that “it is widely acknowledged that South
Africa's housing programme has led to the delivery of more houses in a shorter time
period than any other country in the world” (Charlton, 2003). Likewise Charlton
(2003) further states that in contrast with the rates of delivery elsewhere in the world,
"… one must be impressed with what South Africa has achieved”. This has been a
major political boost, having been the most visible demonstration of the government's
8
Aigbavboa and Thwala

commitment to the delivery of housing and other service to the low-income and
disadvantaged group.

The Housing White Paper which set out the framework for the housing policy and
likewise defines the key elements of the National Housing Policy has seven main
strategies. A key strategy relates to providing subsidy assistance through the Housing
Subsidy Scheme. This encompasses financial assistance by the South Africa
Government to the poor to access housing. According to Charlton (2004), in the
aspect of building new housing stock, the subsidy provided by the government is
intended to cover the cost of purchasing the land, providing basic services
infrastructure such as water, sanitation, and roads; and constructing the house or ‘top-
structure’. Nevertheless, because the Housing White Paper had adopted a market-
centred approach, unfortunately, the State is not able to afford the costs of delivering a
complete formal house to every South African in need of housing because of an
inequitable allocation of funding between different low-income groups; a low rate of
delivery; the deconstruction of existing housing construction capacity; displacement
of communities; a reluctance on the part of the private sector developers to be
involved in conflict-ridden areas; and the reproduction of apartheid-style ghettos
(Landman, 2004). Therefore the state relied on the provision of housing credit (if the
beneficiary can afford to access it) or personal resources, such as savings, labour,
creativity amongst others to supplement the state's contribution (Charlton, 2004).

Consequently, it has never proved easy to help the poor and disadvantaged group
through housing subsidies, particularly in developing countries. Today, very few
governments are prepared to offer housing subsidies to the poor unless they are
delivered as up-front, targeted capital subsidies. However, the lack of resources has
forced each government into making difficult decisions about the size and the number
of subsidies to be offered. Dependent on those decisions, has come a series of
implementation problems relating to the quality of construction, the location of the
new housing solutions, the use of credit and how to allocate subsidies between so
many beneficiaries. Housing delivery for the low income group in South Africa is
reliant on the Housing Subsidy process. At the core of the National Housing
Strategy is the provision of housing subsidy assistance to eligible households. Capital
subsidy assistance is granted to low-income households in order to assist them in
accessing at least minimum standard accommodation. Subsidy assistance is provided
through three subsidy programmes, which are the Housing Subsidy Scheme, The
Discount Scheme and Hostel Redevelopment Programme. The Housing Subsidy
Scheme is the primary means of assistance in terms of the national housing policy. On
March 15, 1994, the housing subsidy scheme replaced all previous government
subsidy programmes for households with an income of R3,500 per month or less.
These households should not own property or have received a government housing
subsidy before and were expected to meet a range of criteria as contained in the
National Housing Act (1997).

The Policy makes provision for financial grants to assist homeless, low-income and
disadvantaged groups to become homeowners. The Housing Subsidy Scheme has
been the key to the delivery of housing since the advent of the government’s low-cost
housing programme mechanism which provides government-funded assistance
packages to households categorized as ‘poor’. Recent policy shifts have been
attempting to simplify the administration of housing subsidies and increasing the
 9
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

subsidy amount. In addition, government policy is placing an increasing emphasis on

the role that beneficiaries of government-funded subsidies should play in delivery,
partly in response to concerns of the culture of entitlement and ownership that
outright subsidies create. As a result, the government now requires that subsidy
beneficiaries contribute to the construction of their homes either through physical
participation or through the payment of a financial contribution.

For the past few years, the National Housing Subsidy has been increased annually to
account for inflation and rising building costs. In 2008, the increase was significant; it
went up by almost 12% for the mostly poor. Housing subsidies have reduced housing
problems in South Africa, giving the poor and the disadvantaged group
homeownership. The scale of South Africa’s government housing delivery is second
only to China, making the success of South Africa’s housing programme unparalleled
amongst other developing nations. Despite all the commendable efforts, the housing
backlog has grown in leaps and bounds from 1.5 million in 1994 and now stands at
approximately 2.1 million, which means that approximately 12 million South Africans
are still in need of better shelter (Tokyo, 2009).

The built houses have encouraged homeownership among the disadvantaged group,
providing them an asset that can be used for further wealth creation thereby reducing
the effect of poverty and housing backlog in the country. But whether it is worth
tackling housing problems in this way, in conditions of high unemployment, huge
income inequality and widespread poverty, inclusive of its sustainability is another
question.

OBJECTIVE AND METHODOLOGY

Scope of Study
Although POE outcomes are useful to inform housing policy and planning
intervention to perform better than previously done in a number of ways (White,
1989), however, no significant POE studies have been carried out systematically in
the Gauteng Province of South Africa, to assess the success of the low-income
housing scheme. The scope of this study is based on the framework of POE (in stages
of indicative, investigative and diagnostic) of Preiser (1989). The occupants of four
government housing subsidy schemes were chosen as respondents to provide self-
reports of their satisfaction with their housing condition based on a list of elements in
the unit and beneficiary expectations before the housing units were allocated to them.

Objective
The objective of the study is to establish predictors, of both physical and social
characters, which influence the satisfaction of residents in subsidized housing
schemes in the Gauteng Province of South Africa.

Methodology
Amerigo and Aragones (1990) in a study on the residential satisfaction in public
housing in Spain accentuated the importance of obtaining distinct geographical
placement of residential satisfaction samples. In this study, the geographical area
chosen is Johannesburg in the Gauteng Province of South Africa. There are various
government subsidized public housing schemes in Johannesburg, Gauteng Province.
10
Aigbavboa and Thwala

Gauteng is a province of South Africa. It was created from part of the old Transvaal
province after South Africa's first all-race elections on 27 April 1994. It was initially
named Pretoria-Witwatersrand-Vereeniging [PWV] and was renamed Gauteng in
December 1994. Gauteng, (a Sesotho word for “place of gold”) continues to serve
as the economic engine room of the country and the subcontinent, responsible for over
34.8% of the country’s GDP, although it is geographically the smallest of the nine
provinces (Pocket Guide to South Africa, 2009). The main cities are Johannesburg,
the biggest city in southern Africa, and Pretoria, the administrative capital of the
country. Gauteng Province is currently home to 11.19 million people (Statistics South
Africa, 2010) as against 10.45 million people reported in the Community Survey
(2007) report. Gauteng Province is also the fastest growing province, with a 22.40%
share of the total population. This is mainly because of the high influx of people from
other provinces and neighboring countries. This is due to the fact that Gauteng is
considered the economic hub and power house of Southern Africa and contributes
heavily in the financial, manufacturing, transport, technology and telecommunications
sectors, amongst others.

Furthermore, housing provision in the Gauteng province has become a burden and a
nightmare to the Gauteng Provincial Government and the National Department of
Human Settlement, with a majority of the low-income housing construction being
given the almost consideration in Gauteng- Johannesburg to be specific. The study
concentrates on occupants of four different housing subsidy schemes in Johannesburg.
The four housing subsidy schemes chosen are Ivory Park Extension 2, Kanana Zone
12, Reiger Park, and Diepsloot.

The four chosen developments are all houses given to the low-income group through
the South Africa housing subsidy scheme. The average size of a housing unit is 40m2.
A structured questionnaire was used to conduct interviews with beneficiaries at the
four locations. This approach was followed to improve consistency in the responses
and ease of analysis. The method was also considered appropriate for a study amongst
the low-income group. This is because it has been suggested that when dealing with a
population likely to be of the low-income and disadvantaged group with low interest
and motivation, the structured interview for data collection is the preferable option.
The questionnaire was designed to seek the opinion of the respondents on their level
of satisfaction/dissatisfaction on the list criteria. The respondents were asked to
indicate the level of satisfaction/dissatisfaction on a scale of 1 – 4 Likert-type scale.

Beneficiaries were randomly selected in all four locations visited; these were
interviewed given the fact that they have been resident in the areas for more than a
month. Out of the 120 questionnaires sent out, 78 were received back; representing a
sixty five percentage (65%) of the total sampled frame. The data collected were
analysis using both descriptive and inferential statistics. The data presentation and
analysis made use of frequency distributions and percentages of all the respondents.
The questionnaire was administered to the heads of households or to the spouses of
the heads of households in the sampled household. One household head per house was
engaged in the questionnaire administration.
 11
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

BENEFICIARIES RELATIVE SATISFACTION INDICES

A 4-point Likert type scale was used to determine beneficiaries’ levels of satisfaction
with regard to the housing unit and the overall housing situation. The scale read as
follows, 1=Very dissatisfied, 2= Dissatisfied, 3=Satisfied, and 4=Very satisfied. The
4-point Likert scale was chosen rather than the 3- or 5-point scale because the study
was demanding more from the beneficiaries and in order to obtain definite answers
and to prevent faking. The neutral level (such as ‘just satisfied’) was omitted from the
list of options. Beneficiaries were thus forced to sincerely rate their level of
satisfaction based on the 4-point Likert scale provided.

The computation of the relative satisfaction indices (RSI) was calculated from the
total of all weighted responses and then relating it to the total responses on a
particular aspect. This was based on the principle that beneficiaries’ scores on all the
selected criteria, considered together, are the empirically determined indices of
relative satisfaction. The index of relative satisfaction (RSI) of a beneficiary is the
sum of the beneficiaries’ actual scores (on the four-point scale) given by all the
beneficiaries’ as a proportion of the sum of all maximum possible scores on the four-
point scale that all the beneficiaries could give to that criterion. Weightings were
assigned to each response ranging from one to four for the responses of ‘very
dissatisfied’ to ‘very satisfied’. This is expressed mathematically as:

N
∑aij
RSIj = i=1
N
∑Aij
I=1
Where,
RSIj = relative satisfaction index for criterion “j”
N = Number of respondents
aij = actual score on the four-point satisfaction scale by the “ï”th
respondent on the “j”th
criterion
Aij = The potential score (or the maximum score that respondent “ï” could
give to
criterion “j”on the satisfaction scale.

When the frequency is calculated to know the number of respondents on each score,
the mean item score (MIS) for each criterion is calculated to obtain the RSI as
follows:

RSI (on a four-point scale)

= 1n1 + 2n2 + 3n3 +4n4

∑N
Where;
n1 = number of respondents for very dissatisfied
n4 = number of respondents for very satisfied
N = Total number of respondents
12
  

The questionnaire for the analysis was recoded on a two-point scale of 1 and 2, where
1 through 2 on the four-point scale was coded as 1 for “not satisfied” and 3 through 4
was coded as 2 for ‘satisfied”. The formula then becomes;

RSI = �� � ��


The criteria are then ranked in descending order of their relative satisfaction index
(from the highest to the lowest).

  

Figure 2 below shows the length of stay of the beneficiaries in the housing units.
About 29.50% of them have been living in the subsidized housing unit for more than
five years. The precentage of those who have lived there between three and five years
is 21.80% while 25.60% is the percentage of those who have been living there for less
than one year. In essence beneficiaries who have lived in their housing units for many
years completed most of the questionnaires. It can therefore be inferred that the
respondents have adequate knowledge of their living apartments and out-door
environment.

Figure 2: Length of Stay in Housing Unit

Figure 3 below shows the beneficiaries’ intended duration of stay beyond what has
already be reported in figure 1.
 13
           

Figure 3: Intended Duration of Stay in Housing Unit

About 94.90% indicated that they intend to live in the housing units for more than five
years while 1.30% indicated they intend not to live there for more than one year. This
is a further confirmation that the occupants’ responses in the satisfaction survey are
based on a genuine motive, because they seek the good and betterment of the living
apartment and environment.

Table 1 below shows the weighted average from the relative satisfaction indices for
the four housing subsidy schemes. The major building aspect/element which the
beneficiaries were very dissatisfied with are ranked in descending order, include the
ventilation in the unit (2.81), numbers of rooms in the unit (2.79), exterior finish
(2.74) and interior finish (2.70). From the physical observation of the units, they were
neither painted nor plastered (figure 6 and 8) . Further observations revealed that the
walls of most housing units were cracked. Winston and Turner (2001) states that walls
act as a support system for the roof and should be constructed from good quality
material otherwise the walls will not be strong and will crack. Cracks in the walls
were part of the structural defects in the housing units which respondents did not
expect to find in the units as in picture 4 and 5. In terms of the weighted rank average
for the finishes, both exterior (2.74) and internal (2.70), there was a general trend in
the level of dissatisfaction as the residents in the different housing units were very
dissatisfied - RP= (2.65, 2.30), IVP= (2.85, 2.70), KE= (2.61, 2.89) and DSP= (2.88,
2.89).

Figure 4: Source- Site Visit Figure 5: Source- Site Visit

14
Aigbavboa and Thwala

The highest level of dissatisfaction was experienced by the residents living in

Diepsloot. Generally, lack of interior and exterior finishes in most of the housing units
influenced the satisfaction levels of the respondents. Also, the ventilation in the unit
ranked very dissatisfactory, because the units were not warm in winter and cool in
summer (figure 7). Respondents expected a housing unit that would protect them from
the elements and especially the harsh, cold winter. According to the World Health
Organization (2004), the quality of a house plays a vital role in the health status of
residents. The indoor air quality, humidity, low temperature and overcrowding in a
house usually poses threats to the health of the residents (WHO, 2004).

Figure 6: Source- Site Visit Figure 7: Source- Site Visit

Figure 8: Source- Site Visit

Other indicies are the size of the unit (2.63), noise level around the unit (2.51),
privacy in the unit (2.51) and safety in the unit (2.50), safety around the unit (2.31),
and position of the bedroom (2.07). Though the occupants were dissatisfied with the
size of the unit, they were at least satisfied with the social and physical elements in
the housing units. The position of the unit (1.97) and the position of the bedroom
(2.07) were very satisfactory as indicated by the weighted ranking averages. With
regard to the space in the unit, respondents indicated that the units were too small as
there was little space for movement after putting in their furniture and most were not
partitioned and could not take all their furniture. However, the weighted average
ranking of the elements shows that the beneficiaries were also not entirely satisfied
with the social and physical elements of the building.
 15
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

Table 1: Relative Satisfaction Indices (RSI) for the Four Housing Subsidy Schemes (in descending order of lesser satisfaction)
Building aspects Ivory Park Rank Diepsloot Rank Kanana Rank Reiger Rank Weighted Sub-group
(N=20) (N=20) Ext 12 Park average rank
(N=18) (N=20) (N=78)

Ventilation in the unit 2.80 1 3.22 1 2.83 2 2.37 11 2.81 1

Number of rooms 2.05 7 3.17 4 2.89 1 3.05 3 2.79 2
Exterior finishes 2.80 1 2.88 9 2.61 6 2.65 6 2.74 3
Space in unit 2.35 4 2.61 14 2.89 1 3.1 2 2.74 3
Interior finishes 2.70 2 2.89 8 2.89 1 2.3 13 2.70 4
Position of kitchen 2.20 6 3.20 2 2.72 4 2.55 8 2.67 5
Climate conditions of unit 2.40 3 2.71 11 2.17 10 3.25 1 2.63 6
Size of unit 2.30 5 2.68 12 2.78 3 2.75 5 2.63 6
Noise levels in the unit 2.00 9 2.59 15 2.67 5 2.9 4 2.54 7
Layout of the unit 2.15 7 3.00 5 2.28 9 2.65 6 2.52 8
Noise level around the unit 2.05 8 2.53 16 2.56 7 2.9 4 2.51 9
Privacy in the unit 1.90 10 2.61 14 2.89 1 2.63 7 2.51 9
Safety in the unit 2.80 1 2.31 17 2.33 8 2.55 8 2.50 10
Position of lounge 1.55 13 3.18 3 2.56 7 2.35 12 2.41 11
Kitchen bathroom/toilet 1.90 10 2.88 8 2.28 9 2.42 10 2.37 12
Safety around the unit 2.20 6 2.22 18 2.33 8 2.5 9 2.31 13
Position of bedrooms 1.70 11 2.94 7 2.00 11 1.65 17 2.07 14
Position of doors 1.50 14 2.63 13 1.83 13 2 14 1.99 15
Position of windows 1.40 14 2.47 16 2 11 2 14 1.97 16
Position of unit 1.40 15 2.78 10 1.94 12 1.75 16 1.97 16
Number of doors 1.60 12 2.95 6 1.39 14 1.9 15 1.96 17
Note: RP= Reiger Park; IVP= Ivory Park; KE= Kanana Extension 12; DSP= Diepsloot
16
Aigbavboa and Thwala

Table 2 below shows the distribution of the residents’ relative satisfaction indices of
the housing units. The numbers of respondents who are satisfied with each of the
building elements are indicated starting with the highest. This implies that the
criterion having the least frequency of relative satisfaction index will have the highest
frequency of relative dissatisfaction index and vice-versa.

Considering all the listed building elements, the residents were more satisfied with the
physical factors in their houses. The respondents who are satisfied with the position of
windows and doors in their houses have the highest frequency (80.77%). This is
followed by the position of the bedroom (76.64%) and the position of the unit which
are all physical factors in the house. The social factors the residents were more
satisfied with were the safety around the unit (67.95%) and safety in the unit (61.54%).

Findings also showed that there was a correlation between the elements residents were
dissatisfied with as shown in table 1 for the weighted average of the ranked items in
the different housing locations and the MIS in table 2. The elements of dissatisfaction
in table 2 are ventilation (32.05%), internal finishes (23.08%), exterior finishes
(43.62%), number of rooms (33.33%), space in the unit (44.87) and size of the units
(38.46%).

Though the residents were satisfied with the physical factors of the unit and not
satisfied with the social factors, when their expectation before the housing units were
given to them and after were examined findings show that their expectation for bigger
housing units (84.62%), houses with quality finishes (98.72%), and more consultation
with the Gauteng Department of Human Settlement (92.31) were not met. This was
not in line with the Department of Human Settlement housing policy goal which
mandated the provincial and local spheres of government to consult meaningfully
with individuals and community affected by housing development, thus facilitating
the active participation of all relevant stakeholders in the housing development.
Nevertheless, residents indicated that their expectation for a house that will improve
their living condition from shacks (slums housing) was met (87.18). Also they
informed that they now have more comfort than their previous living environment
(83.33). Other benefits were better sanitary system (56.41) and clean environment
(53.33), which were all expectations they had before the houses were allocated to
them, as shown in table 4. Only four elements out of ten were met as against the
original intended expectations before allocation of the houses.
 17
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

Table 2: Occupants’ Relative Satisfaction

Building elements Satisfied Not Satisfied
Position of windows 63 (80.77) 15 (19.23)
Position of doors 63 (80.77) 15 (19.23)
Position of bedrooms 59 (76.64) 19 (24.36)
Position of unit 59 (75.64) 19 (24.36)
Number of doors 58 (74.36) 20 (25.64)
Safety around the unit 53 (67.95) 25 (32.05)
Safety in the unit 48 (61.54) 30 (38.46)
Kitchen bathroom/toilet 48 (61.54) 30 (38.46)
Position of lounge 44 (56.41) 43 (43.59)
Privacy in the unit 44 (56.41) 34 (43.59)
Noise level around the unit 44 (56.41) 34 (43.59)
Layout of the unit 39 (50.00) 39 (50.00)
Noise levels in the unit 38 (48.72) 40 (51.28)
Climate conditions of unit 38 (48.72) 40 (51.28)
Size of unit 35 (44.87) 43 (55.13)
Space in unit 30 (38.46) 48 (61.54)
Position of kitchen 29 (37.18) 49 (62.82)
Exterior finishes 27 (34.62) 51 (65.38)
Number of rooms 26 (33.33) 52 (66.67)
Ventilation in the unit 25 (32.05) 53 (67.95)
Interior finishes 18 (23.08) 60 (76.92)
Note: Figures in parentheses are row percentages

Literature (Darkwa, 2006) informs that when the gap between what is expected and
what is received decreases; residential satisfaction increases. Occupant’s satisfaction
with the housing units was affected with lesser of their expectations being met. Also,
residential satisfaction being a subjective evaluation and relies heavily on the
beneficiaries’ views, perceptions, previous experiences, behaviour, norms, values and
emotions, and a complex construct, affected by a variety of environmental and socio-
demographic variables. It can therefore be concluded that the satisfaction of the
occupants living in the subsidized housing units was not met, but from the basic
expectation of improved living conditions compared to a shack and more comfort that
previous living, it can be said that beneficiaries are thus satisfied with the overall
housing condition even though most of their expectations were not met.

Table 3: Level of Housing Satisfaction According to Beneficiaries Expectations

Areas of expectations Expectation Expectation
After allocation Before allocation
Yes No Yes No
Improved living conditions from
68 (87.18) 10 (12.82) 77 (98.72) 1 (1.28)
shacks
More comfort than previous living
65 (83.33) 13 (16.67) 77 (98.72) 1 (1.28)
environment
Good sanitary systems 44 (56.41) 34 (43.59) 76 (97.44) 2 (2.56)
Clean environment 40 (53.33) 38 (50.67) 75 (96.15) 3 (3.85)
Bigger plots 36 (46.15) 42 (53.85) 64 (82.05) 14 (17.95)
Adequate hot and cold water 34 (43.59) 44 (56.41) 70 (89.74) 8 (10.26)
More consultation with the
27 (36.99) 51 (69.86) 72 (92.31) 6 (7.69)
municipality
Free services 18 (23.08) 60 (76.92) 58 (74.36) 20 (25.64)
Structure with quality finishes 12 (15.38) 66 (84.62) 73 (98.72) 5 (1.28)
Bigger units 10 (12.82) 68 (87.18) 66 (84.62) 12 (15.38)
Note: Figures in parentheses are in percentages
18
Aigbavboa and Thwala

CONCLUSION
This paper set out to consider residential satisfaction in South Africa low income
housing schemes; using Johannesburg Subsidized housing schemes in the Gauteng
province as a case study. Literature review showed that the South Africa government
has vigorously ensured that houses were provided to advance the lives of its citizens
through the introduction and implemented of the Housing Subsidy Scheme, hence
eliminating the incidence of slum housing associated with poverty. The empirical
study, although based on a relatively small sample of four locations of low-income
housing in Gauteng, provides an insight into the post occupancy experience of the
beneficiaries of government subsidized housing.

The findings revealed that residents were satisfied with the physical attributes in the
houses, but not satisfied with the social attributes, except in the case of the safety in
and round the unit. Also, despite the majority of the respondents’ expectation were
not met, beneficiaries were satisfied with the privacy and improved living conditions
in the housing units compared to where they were previously living. Additional
findings from the research revealed that the progressive realization of the right to
adequate housing as contained in the South Africa constitution is being met by the
government, as a majority of the beneficiaries that were allocated houses informed
that their quality of life has increase because the provided houses has given them an
improved living condition and they now live in a clean environment. Thus the
Department of Human Settlement objective of the broader housing vision in
promoting social cohesion and improving quality of life for the poor is being achieved
as findings showed.

RECOMMENDATIONS
However, the following are therefore recommended in order to increase the
satisfaction level of beneficiaries: Meaningfully consultation should be held with
individuals and the community affected to facilitate the active participation of all
relevant stakeholders in a housing development and to improve the overall housing
delivery and the satisfaction of the housing subsidy beneficiaries. Also, it is
recommended that the Department of Human Settlement and the administrator of the
subsidized housing policy in the Gauteng Province should conduct a complete and
thorough needs assessment of the beneficiaries of a proposed housing subsidy
development.

The results of the needs assessment should be explained and limitations of the
housing development need to be identified. For example, the beneficiaries may have
indicated a need for a two-bedroom housing unit, but the subsidy amount and
beneficiary contribution might only be sufficient to supply a unit with one bedroom. It
is also recommended that government should provide as wide a choice of housing and
tenure options as is reasonably possible. This can be achieved through the rental
housing option. Finally, it is further recommended that in accordance with the
findings of this study, the Department of Human Settlement should formulate a better
quality control mechanism so that the houses that will be delivered through the
Housing Subsidy Programme will be of good physical quality and also satisfy the
social attributes of a typical housing unit.
 19
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

REFERENCES
Akintoye, A. (2006). Public Partnerships for Sustainable Development of
Infrastructure in Developing Countries, in Okewole I. A et al (Editors). The
Built Environment: Innovation Policy and Sustainable Development. Covenant
University, Ota, Nigeria. Pp433 – 442.
Altman, B. Rogoff. (1987). World views in psychology: trait, interactional, orgasmic,
and transactional perspectives, in: Stokols, Altman _Eds.., Handbook of
Environmental Psychology, Vol. 1, Academic Press, New York (19)., pp. 7–40.
Amerigo M., Aragones J.I. (1990). Residential satisfaction in council housing,
Journal of Environmental Psychology 10, 313–325.
Anderson, J. and S. Weidemann (1997). Developing and utilizing models of resident
satisfaction. Advances in Environment Behavior Desing. G. Moore and R.
Marans, New York: Plenum. 4: 287- 314.
Aribigbola A. (2006). Housing affordability as a factor in the creation of sustainable
environment in developing world: the example of Akure, Nigeria. Dept of
Geography and Planning Sciences Adekunle Ajasin University, Nigeria.
Blakstad, Olsson, Hansen, and Knudsen. (2010). Usability Mapping Tool, CIB World
Congress, pp. 216.
Bonnes M., M. Bonaiuto, A.P. Ercolani. (1991). Crowding and residential satisfaction
in the urban environment, contextual approach, Environment And Behavior 23
(5), 531–552.
Bruning, S. D., Langenhop, A. and Green, K. A. (2004). Examining city-resident
relationships: linking community relations, relationship building activities, and
satisfaction evaluations, Public Relations Review, 30, 335-345.
Campbell, A., P. E. Converse, et al. (1976). The quality of the America life:
Perceptions, evaluations, and satisfaction, New York: Russell Sage Foundation.
Carvalho, M., Goerge, V. R., & Anthony, K. H. (1997). Residential satisfaction in
conominos exclusivos (gate-guarded neighborhoods) in Brazil. Environment
and Behavior, 29, 734–768.
Charlton, S (2003) The Integrated Delivery of Housing: A Local Government
Perspective from Durban. In Harrison, P Huchzermeyer, M and Mayekiso, M
(eds) Confronting Fragmentation: Housing and Urban Development in a
Democratising Society. Cape Town: UCT Press.
Charlton, S. (2004). An Overview of the Housing Policy and Debates, Particularly in
Relation to Women (or Vulnerable Groupings). Research report written for the
Centre for the Study of Violence and Reconciliation. South Africa.
Community Survey. (2007). Basic results. Statistics South Africa. p.2. Available at:
http://www.statssa.gov.za/publications/CS2007Basic/CS2007Basic.pdf.
Darkwa, I. (2006). Post-occupancy evaluation of State-subsidized housing units in
Kayamandi, Stellenbosch. Masters in Consumer Science (Housing), University
of Stellenbosch, 2006.
Fishbein, M & Ajzen, I. (1975). Belief, attitude, intention and Behaviour: An
introduction to theory and research. Addison-wesley publishing company.
Francescato, G et al. (1976). A systematic method of evaluating multifamily housing.
Ekistics, 41 (242), 60-3.
Francescato, G. (2002). Residential satisfaction research: the case for and against. In
Residential environments: Choice, satisfaction and behavior, Bergin and
Garvey, London.
Francescato, G., et al (1987). Residential satisfaction: its uses and limitations in
20
Aigbavboa and Thwala

housing research, in: Housing and Neighbourhoods : Theoretical and

Empirical Contributions, Greenwood Press: New York.
Francescato, G., Weidemann, S., & Anderson, J. R. (1989). Evaluating the built
environment from the users point of view: An attitudinal model of residential
satisfaction. In W. F. E. Preiser (Ed.), Building NY: Plenum Press evaluation,
pp. 181–198.
Galster G.C., Hesser G.W. (1981). Residential satisfaction. Compositional and
contextual correlates, Environment And Behavior (13) 735–758.
Garling A., Garling T. (1990). Parents’ residential satisfaction and perceptions of
children’s accident risk, Journal of Environmental Psychology 10, 27–36.
Issac A. Mier, Yaakov Garb, Dixin Jiao and Alex Cicelsky. (2009). Post Occupancy
Evaluation: An Inevitable Step Toward Sustainability, Advances in Building
Energy Research, Vol 3, pp 189-220.
Jenkins, P. (1999). Difficulties encountered in community involvement in delivery
under the new South African housing policy. Habitat International, 23(4),
431–446.
Kim, S. (1997). Outdoor environment satisfaction: contributions of landscape design
to multi- family housing residents' satisfaction. Environmental Design and
Research Association.
Lackney, J. A. (2001). In The 32nd Conference of Environmental Design Research
Association, EDRA., Edinburgh, Scotland.
Landman, K. (2004). Gated communities in South Africa: A Review of the relevant
policies and their implications. CSIR Building and Construction Technology,
South Africa.
Lane, S and Kinsey, J. (1980). Housing Tenure status and housing satisfaction.
Journal of consumer affairs 14(winter) 341-65.
Lansing, J. et al. (1970). Planned residential Environments. Institute for Social
Research, Michigan.
Liu A.M.M., Residential satisfaction in housing estates: a Hong Kong Perspective
Automation in Construction, 8, 511–524, 1999.
Lu Aye, B. Charters, M. Chiazor and Robinson, J. (2004). In association of
Researchers in Construction Management 20th Annual Conference, Vol. 1 (Ed,
Khosrowshahi, F.) ARCOM, Heriot Watt University, Edinburgh, Scotland, pp.
277-286.
Lu, M. (1999). "Determinants of Residential Satisfaction: Ordgered Logit vs. R
egression Models." Growth and Change 30(Spring): 264-287.
Markus T.A., P. Whyman, J. Morgan, D. Whitton, T. Maver, D. Canter, J., Fleming.
(1972). Building Performance, Applied Science Publishers, Barking.
Mastor, Shah and Ibrahim. (2010). Post Occupancy Evaluation Practices: A
Procedural Model for a Successful Feedback, CIB World Congress, pp 277.
Michelson, W. (1977). Environmental Choice, Human Behaviour, and Residential
Satisfaction. Oxford University Press, UK.
National Department of Housing Urban Development Framework. (1997), Available
at: www.info.gov.za/view/DownloadFileAction?id=70505, accessed on
19/02/2009.
Onibokun, A.G. (1974). Evaluating consumers satisfaction with housing: An
application of a systems approach, AIP Journal, 189-200.
Pocket Guide to South Africa PROVINCES (2009)
http://www.gcis.gov.za/resource_centre/sa_info/pocketguide/2009/006_proven
cies.pdf.
 21
An Appraisal of Housing Satisfaction in South Africa Low Income Housing Scheme

Preiser W.F.E. (1989). Towards a performance-based conceptual framework for

systematic POEs, in: W.F.E. Preiser _Ed.., Building Evaluation, Plenum, New
York.
Preiser, W. F. E. and Vischer, J. (2005). Assessing Building Performance, Butterworth-
Heinemann, Oxford.
Speare, AA. (1974). Residential satisfaction and intervening in residential mobility,
Demography, 11, 173-88.
Statistics South Africa. (2010). Mid-year population estimates
http://www.statssa.gov.za/publications/P0302/P03022010.pdf.
The World Health Organization. (2004). Review of evidence on housing and health,
Fourth Ministerial Conference on Environment and Health. Geneva. The
World Health Organization. Background document No. 5046267.
Tokyo Sexwale. (2009). Speech at the occasion of the Budget Vote, Number 26 for the
Department of Human Settlements, National Assembly, Cape Town. Available
from: http://www.housing.gpg.gov.za/Speech09.htm.
Varady, D. (1983). Determinants of residential mobility decisions. Journal of the
American Planning Association, 49, 184-99.
Weidemann S., J.R. Anderson, D.J. Butterfield, P.M. O’Donell. (1982). Residents
perception of satisfaction and safety. Basis for change in multifamily housing,
Environment And Behavior (14) 95–724.
Weidemann, S., & Anderson, J. R. A. (1982). A conceptual framework for residential
satisfaction. In I. Atman, & R. Werner (Eds.), Home environments London:
Plenum Press, pp. 154–182.
White E.T. (1989). Post-occupancy evaluation from the client’s perspective, in: W.F.E.
Preiser _Ed.., Building Evaluation, Plenum, New York.
World Commission on Environment and Development (1987); Our Common Future.
Oxford University Press, Oxford.
Zimring, C. (2002). Post-occupancy evaluations: Issues and Implementation, John
Wiley and Sons Inc, New York.