A typology of Housing Search Behaviour in the Owner-Occupier Sector



Download 13.27 Mb.
Page21/60
Date30.04.2018
Size13.27 Mb.
1   ...   17   18   19   20   21   22   23   24   ...   60





















Chapter Five: Research Methods and Design



Key points from chapter five


  • This research responds to the limited evidence of housing search behaviour as described in chapter four

  • An interpretation of critical realism, brought together with old behavioural economics provide the methodological framework for the research

  • Quantitative methods and surveys are discussed in theoretical terms in relation to critical realism, behavioural economics and the variation in search behaviour

  • The research design process is described to show how the methods build on the conceptual model of housing search variation

  • The data analysis is described, including a detailed explanation of the combined use of principal components analysis and cluster analysis as means of grouping similar housing search behaviours to create a typology

  • Sheffield is introduced as the case study area

















5.1 Introduction


This research is a response to the lack of empirical evidence from an OBE perspective of households’ housing search behaviour. There is little evidence currently about the types of information source used, the length and extent of search, its intensity and the influences of different actors on search outcomes. In order to understand the norms and variations in behaviour, as well as to construct a typology of different housing search behaviours, evidence of the actual behaviours of households is needed. This chapter has two aims: to justify and relate the research methods used to obtain information about the housing search process, and to justify Sheffield as the case study housing market used in this thesis.

Chapters three and four explored how different ontologies of housing search are discernible in the different schools of economic thought. The conclusion of chapter four showed that from an OBE perspective there is a gap in knowledge of the variation in search behaviours across a range of variables across the different stages of search. Research is needed to answer the question, do households exhibit variations across these variables, and if so, can a typology of this variation extend our understanding of the housing market? These questions relate to the research objectives discussed in chapter one, and the methods required to answer these questions are outlined in this chapter.

The different schools have also led to different types of research method being operationalized. In NCE housing researchers have most frequently used quantitative techniques exploring secondary data about search outcomes (for example hedonic analyses using census and land registry information). Many of these research methods have been refined and expanded to increasing levels of technical complexity and have revealed many insights about housing markets. Yet many of these econometric approaches remain unsuitable for exploring the variation in housing search behaviour. There have been repeated calls from those working in housing and property markets to explore a wide range of methods and assemble pluralistic panoply of research methods that are suitable to exploring alternative questions (Kauko, 2001; Guy and Henneberry, 2002; Gibb, 2009), including qualitative and survey based insights (Watkins, 2008).

A survey approach is considered most suitable for exploring the variation in owner-occupiers’ housing search processes. Breadth of explanation is needed to explore the variability in housing search processes and depth is needed to explain the complex iterative process of determining household housing choice. To satisfy both the breadth and depth necessary a household survey across a housing market (with detailed interviews prior to the survey implementation, to test and refine the survey questions) is suitable. This fulfils the requirement for both the quantitative data to measure the extent of variation and qualitative insights into the rationales and details of search. Whilst it would be possible to gain some insight from local housing officers, planners, property professionals and developers about their perceptions of the household search process; there is a sturdier logic to asking the subjects themselves.


5.2 Philosophical underlabouring: the methodological framework


Chapters two to four revealed that housing economics contains many varied ontologies of human behaviour. The relationship, however, between philosophical perspectives and economic models and methods is rarely made explicit. As chapter four outlined BE is a broad school: NBE has a commitment to studying individuals as atomised, whilst OBE has a commitment to studying their behaviours in relation to others and wider trends of behaviour. The methods used in OBE relate to both process and outcomes of economic activities, as both are of interest in understanding the holistic economy. The focus here is on the individual household’s experience as well as the similarities and differences between households’ search processes and the norms within the housing market. Given the difference in emphases between NBE and OBE, it is not prudent to assume that methods can be substituted within BE variants, or that the insights from them are intrinsically transferable, despite the frequent calls for greater pluralism in housing and property economics research. A shared philosophical understanding is required before sharing methods, analysis or observations about the housing market.

Watkins and McMaster (2011) caution against pursuing a pluralist agenda without being explicit about the philosophical foundations of the project. Marsh and Gibb (2011) likewise argue that clarity of philosophical commitments is needed, and suggest that a Critical Realist perspective may be fruitful in housing economics:

“We would add the observation that another potential avenue for exploration is work invoked by critical realism. There has been some work in housing that has explicitly invoked critical realism (e.g. Dickens, Duncan, Goodwin and Gray 1985, Fitzpatrick 2005), but this has largely passed by the housing economics community. This can be contrasted with other areas of economics; there has been a lively debate on critical realism as a mechanism for retaining aspirations relating to the nature of explanation in economics, without that necessitating a commitment to positivist account of closed systems, covering laws and constant conjunctions (e.g. Fleetwood 1999, Fullbrook 2009, Lawson 1997). We would suggest that an examination of these ideas in the housing market context is overdue.” (Marsh and Gibb, 2011, P.298)

Whilst further work is needed to identify possible overlap in methods in OBE and other schools of economics, CR offers several advantages as an ontological (and epistemological) approach to understanding the housing market from which it is possible to judge potential synthesis of methods.



Open World systems

Critical realism rejects the closed world system of some economic modelling (Bhasker, 1998). Closed world systems may be appropriate when considering non-agency based structures, such as astrophysics, yet the introduction of a conscious actor into the system leads to questioning of the clarity of a closed world in housing economics. In a closed world an event will always produce the same output, for example under specified conditions an interest rate rise will have a defined decrease upon the level of borrowing, although in the social sciences these conditions frequently only occur in laboratory settings (Sayer, 1992). Experimental work therefore needs to be verified in open world contexts before it can be applied (Benton and Craib, 2011). These laws are representations of both the inputs and outputs of a housing market, they will therefore explain event regularities. However, the failure of deductivist models to predict individual and market behaviour suggests that there is a failure in either the laws constructed or the antecedents of deductivist modelling where agents are involved. Lawson (1997, 2001) argues persuasively that the failure is due to the unrealistic assumptions involved in using a closed world deductivist model. Naess (2004) argues that under open world assumptions, it is still possible to study social phenomena in the urban realm for prediction purposes for planners.

Those working within critical realism have been criticised for their lack of work in bridging from their theoretical position to the methods employed in social science research (Pratt, 1995; Oliver, 2011). This is, in part, due to the success that critical realism has had in crossing paradigm boundaries. Indeed, critical realism spans research paradigms from theology to natural sciences and economics (e.g. Wright, 1992; McGrath, 2004; Lawson, 2001). The heterogeneity within critical realist accounts likewise lends itself to the lack of a clear relationship between theory and methods (Danermark et al, 2002). In this context it is unsurprising that criticism has been raised relating to methods, when in each paradigm critical realism must grapple with the prevailing epistemology and methodological considerations, yet maintain that the method should remain subservient to the ontology of the discipline. The diffuse range of methods utilised in critical realist accounts of social phenomena could be seen as a strength of the philosophical outlook. Danermark et al (2002) argue that there should not be a predefined set of methods, rather the method should be informed by the question under consideration.

Quantitative methods

Quantitative methods, de rigueur in economics, have at times reached an almost unassailable credibility. The quantitative hegemony in economics is in part due to the philosophical assumptions of NCE and its position within the discipline, and in part due to the numeric nature of economic subjects of interest, chiefly money (Sayer, 1992).

The hegemony of quantitative methods and mathematical abstraction may actually damage the credibility of economics if the subjects under consideration are not suitable for quantitative analysis, or if the assumptions underpinning a mathematical formula are inaccurate. Indeed, mathematical purity may detract from a greater understanding of economic activity if the collective energy of researchers is focussed on improving mathematical modelling and not critiquing the underlying assumptions that have been taken for granted. Mathematical tractability and accuracy of prediction has been one of the main arguments in favour of maintaining a NCE outlook, yet as has been argued above, when the ability to accurately predict fails, two options remain; readdress the maths or the assumptions underpinning the model. It is becoming increasingly common to address all quantitative modelling as an instrumentalist straw man, for example:

In the context of modern economics, the central oppositions, dividing mainstream economists and their more heterodox opponents, are, if not always so expressed, between those who insist, or at least prioritise, closed-systems formalistic modelling methods and those who do not; between those who think that mimicking natural-scientific method is a sufficient basis for fashioning the methods of social science and those who do not; between those who neglect ontology and those who do not; and between reductionist and non-reductionist approaches to science. (Lawson, 2001, 178)

Lawson admits that this is a crude generalisation; however it serves to point out the necessity of considering ontology, methods and the relationship between them. Any credible quantitative analysis in economics must therefore be able to justify the abstractions made; research methods within a quantitative framework must therefore only be selected if the subject under consideration lends itself to quantitative abstraction and definitions. This is particularly the case when the research aims and objectives are not primarily focussed on prediction but description or modelling behaviour. An instrumentalist may be able to argue that, where prediction is paramount explanation of the structure of the world is unimportant. It is not the purpose of this section to provide a comprehensive discussion of the arguments for and against this ‘Friedman style’ affirmation (see Lawson, 1997 for more), merely to highlight the inherent contradiction in pursuing an instrumentalist argument where the research agenda is focussed on contemplating social structures. Given, therefore, the research aims and objectives of this research, instrumentalism is rejected and consideration is given to the structure of reality and therefore the methods used to explain that reality must be contingent upon it.

In rejecting deductivist logic some research has rejected the methods associated with it. Within critical realist, OBE and IE there have been calls to adopt a new set of methods (see below for a discussion on behavioural methods), however, whilst this would highlight the differences between those philosophical positions and NCE, the rejection of current methods is a logic step beyond the fundamentals of critical realism, instead the method should be related to the structure being studied. In short, the majority of methods are not precluded from critical realist research, however, the method must by dependent upon the theoretical understanding of reality in question, and based upon realistic assumptions about the makeup of that reality.

Relation to behavioural economics

Behavioural economics’ diversity is seen not only in its philosophical foundations, but also in the methods that it employs. The rise of BE has occurred at the same time as a rise in experimental economics. They have, in part, fed from each other, but have also been found separate, for example closed world experiments in game theory20. The clear distinction between NBE and OBE requires that the differences are respected when considering the methods employed.

The prevalence of experimental research methods in NBE is fitting given the proximity of the philosophy to a closed-world ontology. In this context experimental economics may reveal transfactual regularities, such as loss aversion, that will prevail in the real world. The logic steps between these philosophical assumptions and methods may be sound, however, as has already been argued, where there is a fundamental failure of theories, based upon these experiments, to predict actual behaviour in non-controlled environments they lose the tractability argument and hence concern over the underlying assumptions re-emerges. This is not to discount experimental research or its results per se, the work done on loss aversion, for example, may indeed have a great deal of relevance to research into the behaviour of households as they seek to sell a property. The concern is how experimental results are related to models of real world behaviour. Again, using loss aversion, controlled experimental results may provide a picture of a transfactual, however, the necessary closed world ontology may in fact prevent translation, as there is no guarantee in an open world system that an agent would change the course of their behaviour. This argument, at the very least, places an emphasis on humility and complete description of the contextual limitations of incorporating such evidence into a model of the housing market.

Old behavioural economics, with its emphasis on the agency of actors, builds an alternative grounded approach to understanding behaviour. The use of experiments therefore may be of some use, but will have its limitations. Simon (1994) argues that experimental conditions may be suitable for testing human economic behaviour, although he is perhaps overoptimistic about directly relating these findings to real world situations.

“Computer simulation of individual and organizational problem solving and decision-making has become an important way of stating theories in rigorous and testable form. Finally, there has been a recent vigorous development of experimental economics, especially the laboratory study of markets. All of these improved and new techniques facilitate progress in our understanding of human decision-making, and as a consequence, our understanding of human bounded rationality in the operation of business firms, markets and the economy.” (Simon, 1994, P4)

Elsewhere, Simon continues to support methods that gather both quantitative and qualitative data for interpretation by researchers, calling for a broader range of research skills, including importing research methods from non-economic disciplines (Simon, 2000). This approach needs to be applied across the housing search process and not limited to one aspect.



OBE ontology of housing search

Given the above discussion about studying behavioural analysis within markets, it is now necessary to turn to the precise objects of analysis within the market. In this open- multi dimensional world, what ontological objects should be considered to meet the research goals of this study?

In order to answer the research questions in accordance with the research framework information is needed about the whole search process. As figure 4.12 depicts, information should include: previous experiences of housing search; perceptions of the market; aspirations; motivations; search length; search intensity; information sources used; the timing of stages in the search process; properties viewed and offered and the extent to which aspirations were achieved. Whilst it is not necessary for the typology construction, data should also be collected about the household characteristics, as it is useful to explore whether some of the variation in search behaviour conforms to variation in household characteristics (e.g. household size or type). It is necessary therefore to gather a mixture of attitudinal and factual information about the search process and household characteristics. These questions are about two types of data: attitudinal and factual attribute (search event and household characteristics), which therefore require two different types of question in the research (see survey design section below).

Who needs to be represented in the data?

The term household is used throughout this thesis, and refers to the unit moving, whether an individual or a group of people (and indeed whether people living in more than one dwelling prior to the search or not). The decision making process is akin to a firm’s, or organizations approach and is hence considered as a singular entity for housing search here (Brown and Moore, 1970; Simon, 1972). In reality a multi-person household is likely to have competing notions of aspiration and resources, and decisions may be taken in the course of the housing search by different members that reveal these differences (Levy and Lee, 2004), and wider social networks may also play a role in shaping the decision (Levy et al., 2008). However, the search process is viewed as a response to a household’s collective aspirations, preferences and level of satisfaction (Ferrari et al., 2011). The household is therefore considered as a singular unit, this simplifies the models of search, but may given an overly simplistic representation of some households.



What data is possible to be collected? (epistemeology)

Questions of epistemology rarely enter NCE, but are more frequently addressed in heterodox economics. Whilst there is limited space here to expand on the details of the epistemological nature of economic research, or indeed of the epistemological assumptions underpinning each variable of the data required to answer comprehensively the research objectives, it is necessary to briefly address the issue of whether it is possible to learn about the search behaviour of households.

Critical realism has been criticised for its emphasis on ontology at the expense of engagement with questions of epistemology (Downward et al, 2002), although this critique does not take into account proponents of CR in economics who argue that the participant can know (at least approximately enough to constitute knowledge) some things and communicate these to the researcher (Lawson, 1997). The question of epistemology also relates to the issue: whether it is possible to combine a range of understandings of social objects. Critical realism recognises that meanings are frequently shared by social groups as part of an underlying structure (Sayer, 1992) and therefore it is possible to compare perceptions of events where shared meaning is likely. Given the relationship between OBE and OIE and conceptualisations of shared meaning through enduring habits, behaviours, laws and customs, it is reasonable to assume that there is some shared conceptualisation of housing search processes, which is known by the housing searcher and communicable to the researcher. In this research instance most of the objects of enquiry (the variables) relate to concepts that are intuitively shared between households in the housing market, and the clarity of these concepts can be tested through the pilot stages of the research.

The use of quantitative methods (particularly econometric) has received some criticism by CR theorists. However, Downward and Mearman (2002) argue that the tensions between CR and statistical methods in economics are no greater than the tensions with other methods, and therefore “the applied researcher wishing to face up to the philosophical criticisms made by critical realism should make sensible use of whatever empirical methods are at disposal.” (Downward and Mearman, 2002, P.412). Extending Downward and Mearman’s argument from Post-Keynesian economics to OBE, which likewise supports the non-duality between quantitative and qualitative methods, the use of statistics in empirical work is suitable.


5.3 Questionnaire and Quantitative methods


A range of methods is suitable for building up a description of social phenomena. Interviews, for example, could provide a rich source of data about the sources of housing market information obtained by households in their search, they could be used to describe in detail the emotional response and collective housing search process of households as they view properties and decide whether or not to place an offer. Within the behavioural tradition in housing, interviews are increasingly being used as a valid research tool (see for example, Levy and Lee, 2002; Smith et al., 2006 and Wallace, 2008). The insights from interviews have broadened our understanding of crucial elements of the decision-making process. Yet, the disadvantage of interviews lies in the pragmatic limitations of carrying out extensive research using interviews. Whilst this is of limited concern in highlighting the limitations of economic models, and in further clarifying the distinction between actors, it is of limited scope in trying to provide a more comprehensive model of the extent and variation in housing search behaviour.

Where NCE has in the past relied on econometric methods using a range of aggregated data sources; OBE concern with the search process and not simply outcomes questions this approach. To understand the range of search behaviours, and in this instance a behavioural approach to housing search, social scientists turned to other methods, including surveys to provide the evidence base.

“While not ignoring such data, behavioural economists have sought more direct ways to observe actual decision-making in organizations, as well as the behaviour of consumers. Survey research aimed at collecting data about expectations, provided much of the early information about departures of actual behavior from perfect rationality." (Simon, 1994, P.4)

The empirically related research goals outlined in chapter one are numbers three and four. They are:



  1. Test the variation in search behaviour through empirical research

  2. Create a typology of housing search behaviour and relate this typology to the existing literature

These goals require data that can be used to exhibit the variation and similarity in search behaviours. The primary emphasis here is not to explain why behaviour may vary (i.e. there is no attempt to prove causality), but if and how it varies. The appropriate research method for answering these goals is different therefore to the research methods that would be appropriate if explanatory goals dominated the research. Whilst questions of explanation are relevant here, the weighting for the selection of a research method is lower than that given to the description of variation. With these goals in mind it necessary to select a research method that is capable of representing information from across the population, which is capable of reflecting the variation within that population.

Prior to the utilisation of primary research it is necessary to question the necessity of collecting new information. Secondary data analysis represents a frequently inexpensive and expedient research process for housing market research. Data sets such as the British Social Attitude Survey and English Housing Survey represent potential secondary data sources that could be utilised. However, neither of these data sets (nor any other which the author is aware of) represents information across the full range of stages of housing search behaviours. It is necessary therefore to undertake primary research to collect this information.

Given the breadth of information required to answer questions about variation across the whole search process (see Fig. 5.1) it is necessary to use a research method that is capable of reflecting this breadth across the population. Interview research methods would be capable of gathering detailed information across the whole range of stages of the search process, but would be expensive and time-consuming to undertake a representative study. Open-ended interviews would also be a useful research method for engaging with explanatory questions, but may prove difficult to analyse comparatively for the research goals of testing variation and constructing a typology of behaviour.

A survey method is therefore selected as the most suitable method to explore the variation in search behaviours across the population of owner-occupier dwelling purchasers. Sun and Manson (2010) argue that questionnaires have long been recognised as a suitable method for detecting variations in housing search behaviour, but have been under utilised because of their cost. Buckingham and Saunders (2004) set out nine questions to analyse whether a survey is the most suitable research method. The issues they raise cover the type of research questions, the level of detail and extent of coverage necessary, the timing of events, the complexity of behaviours, sensitive topics, units of analysis and the size of population. Under these criteria a survey is considered a relevant and robust research method for gathering the data necessary to complete the research goals21.



The survey method required several stages of work: survey design, piloting, sampling, distribution, data entry and data analysis. An overview of these stages follows.

5.3.1 Survey Design


The survey design was a balance between asking enough detailed questions to build up a picture of the variation in search behaviour across the stages of housing search and limiting the extent of the survey to encourage as many households as possible to complete it (longer surveys tend to receive much fewer respondents (Edwards et al., 2002; Bryman, 2008)). The survey was therefore limited to eight pages.

Three groups of questions were included in the survey: household, housing and search characteristics. Questions about the household covered the composition of the household (number of people, relationship between members of the household) and indicators of their stage in life and relation to wider social structures (e.g. employment, education). The housing characteristics collected covered the type of dwelling, size, age, garage and presence of a garden. Using the HMLR data, the location of the dwelling was known prior to distribution. The remainder of the survey was about the search process.



Four events were identified as key in the search process, which would be used to distinguish between stages of search: the first time you considered moving from your previous home; first physical viewing of a property; first offer on a home; and moved into this home. These events were used to identify the time for each stage (between events) as well as changes in the household’s aspirations across the search process and household’s perspectives on the pressure their search was under. The survey covered households’ experiences of the search process at various stages: pre-search attitude towards searching and the dwelling (e.g. satisfied and not regularly looking for a dwelling etc), the motivations for moving dwelling, the experience of moving dwelling and reflections on the outcome and process of searching. The survey also covered the information sources used and their importance, frequency of use and timing of use, on the search process22.

Table 5.1 List of survey variables and data types: across stages, household and dwelling characteristics


Aspect

Pre-Search

Search

Visit

Offer

Move

Variables

Data type description

Previous experience

X













Tenure

Categorical

X













Time

Date

X













Location

Postcode

Perceptions

X













Size

Likert (Agree)

X













Type

Likert (Agree)

X













Location

Likert (Agree)

Attitude to search

X













Search consideration

Likert (Agree) & text

X













Opportunity availability

Likert (Agree) & text

X













Dwelling & finance satisfaction

Likert (Agree) & text

X













Event changed perception

Likert (Agree) & text

Motivations

X













Economic

Likert (Agree) & text

X













Family

Likert (Agree) & text

X













Size/Design dissatisfaction

Likert (Agree) & text

X













Location stress

Likert (Agree) & text

X













Finance stress

Likert (Agree) & text

Aspirations


X













Wealth

Likert (Importance) & text

X













Social Status

Likert (Importance) & text

X













Comfort

Likert (Importance) & text

X













Stimulation

Likert (Importance) & text

X













Enable personality

Likert (Importance) & text

X













Proximity to friends

Likert (Importance) & text

X













Good social influence

Likert (Importance) & text




X

X

X




Change in aspirations

Binary & text

Sources of information




X

X

X

X

Importance

Likert (Importance) & text




X

X

X

X

Frequency

Likert (Often) & text

Time




X

X

X

X

Time

Date

X

Time pressures

Likert (Agree)

Properties







X

X

X

Properties viewed

Number










X




Properties offered

Number

Outcomes













X

Aspirations met

Likert (Agree)













X

Dwelling - Type

Categorical













X

Dwelling - Bedrooms

Number













X

Dwelling - Bathrooms

Number













X

Dwelling - Garage

Categorical













X

Dwelling - Garden

Categorical













X

Dwelling - Age

Categorical (ordinal)

Experience

X

Search experience

Likert (Agree) & text

Household













X

Tenure

Categorical













X

Income

Ordinal













X

Ethnicity

Categorical













X

Relationship types

Categorical













X

Age

Interval













X

Sex

Categorical













X

Working status

Categorical













X

Employment group

Categorical













X

Educational qualifications

Categorical













X

Employment/education location

Postcode

A key survey design question related to the use of measurements for variables. As can be seen from Table 5.1, the variables, which are necessary to answer the research objectives, were coded according to data type (e.g. ordinal, Likert). The decision about the appropriate data type for each variable is based on two questions: what data types can represent the variable (epistemology), and what data types are needed to reveal the variation in behaviour and to construct a typology. In this case the response to every variable is not comprised of the same data type, as for some variables the data would be difficult to accurately obtain (or may not be possible to represent in this way). The full survey can be found in Appendix C.

There are nine different data types in table 5.123. Some of these data type descriptions relate to intuitive categories (e.g. date [month and year] of an event), some are less intuitive (e.g. Likert). The following section describes the survey design considerations relating to the data types: Likert; text; binary; and categorical.



Likert

The use of Likert items, based on semantic differentials, was determined for the survey in order to provide a flexible scale, but with regular intervals that would be comprehensible for respondents. The use of Likert items is very common in social science research, particularly for the measurement of attitudes (Brill, 2008) across a range of declarative statements (Peterson, 2000) and has been used in housing market studies in identifying buyer preferences and behaviours (Bible and Brown, 1981; Greaves, 1985; Munro and Lamont, 1985). The use of Likert items also includes the possibility of extensive use of quantitative methods in the analytical stages of the research, as the items are readily quantifiable through the use of representative consecutive integers. Five point scales were used; although adding points to this scale allows additional granularity there is a trade off in terms of the precision with which people can accurately identify their attitudes with extensive scales24. Even number point scales (forced choice scales) were rejected, as it was considered important that households could indicate that they neither agreed nor disagreed. Two potential biases occur with Likert items that may be of issue here; acquiescence bias and central tendency bias (Welkenhuysen-Gybels et al., 2003; Bertram, 2003). These biases are normally overcome by asking multiple questions on the same topic, but framed differently, for example some positive and some negatively (Brill, 2008). However, given the extensive nature of the research topic this was not possible as the survey would have been overly lengthy.



Fig. 5.1: An example of the Likert item questions in the survey

A horizontal scale for the Likert items was selected (over a vertical scale). A horizontal scale allows respondents to move quickly through the range of questions that are grouped together (e.g. factors influencing the decision to search) and there is less ambiguity about where the tick should be placed (Bryman, 2008). Horizontal scale can be clearly related to numeric values (e.g. Strongly Agree (1), Agree (2) etc.) in order to save space in the survey, and hence avoid length based non-response or missing data (e.g. Fig 5.1).



Text based data

The survey included some free text boxes for respondents to reply in an open setting (see Fig 5.2 for an example). Whilst free text boxes allow greater detail into respondents’ perceptions of their behaviour their use may increase non-response rates (Bryman, 2008), and hence were used for only some of the questions, which were priorities for open text (see Table 5.1 for variables containing free text boxes). The text from these boxes could be transformed to quantitative data (for example counts of the numbers of specific words), but in this research project the quantitative variables required for analysing the variation and creation of the typology are all included in the survey in other forms (e.g. Likert and binary). The inclusion of text based questions enables qualitative checks on the survey responses and also provides examples to illustrate the narrative of housing search variation and classification.

Fig. 5.2: An example of a text based question in the survey





Binary data

Some variables required only a binary response, these variables relate to the presence or absence of the variable. For example, for variables relating to the characteristics of the dwelling purchased this might relate to the presence or absence of a garden (a binary distinction). For these forms of question a tick box for ‘Yes’ and a tick box for ‘No’ were selected (see Fig 5.3 for an example). Other forms of binary response are possible, for example writing ‘Yes’ or ‘No’, but tick boxes were selected to require as little work as possible from the respondent (and therefore minimise fatigue) and to facilitate the coding of surveys upon completion. Horizontal options were selected in order to limit the space taken up by the questions and to follow a similar format to the more frequently used, Likert item, questions.



Fig. 5.3: An example of a binary question in the survey



Categorical data

Categorical data is required for some variables relating to household and housing characteristics. Variables that use categorical data must be capable of being represented by discrete (non overlapping) options, for example property types are normally known by discrete categories (e.g. semi-detached). For the same reasons as those for binary data, categorical data options were horizontally embedded in the survey.



A note on sensitive questions

Questions on household income typically receive the highest non-response rate in surveys (Ross and Reynolds, 1996). This may be because of concerns about the personal nature of the topic and confidentiality (Peterson, 2000). In this instance three measures were undertaken to mitigate non-response for this question. Respondents may also be concerned about their income being low on the list of options; the inclusion of low-income brackets may therefore encourage responses from low-income households (Peterson, 2000). First, the introductory letter highlighted the confidentiality and anonymity of respondents. This was repeated in the instructions for the income question, including the phrase “This information will be treated in the strictest confidence.” Second, sixteen categories of income ranges were selected for the survey so that households will not be required to reveal their precise income.25 Third, the question categories began with low-income levels (Below £5,000) in order to encourage owner-occupiers to respond.



Introductory letter

The survey included an introductory letter setting out the research agenda, funding for the survey and that the research was conducted by the University of Sheffield, as there is some evidence to suggest that response rates are higher for university based research than other research bodies (Edwards et al., 2002). This letter was designed to fulfil these three aims in a clear and brief manner and perspicuity and brevity have been shown to increase response rates (Couper, 1997).



Download 13.27 Mb.

Share with your friends:
1   ...   17   18   19   20   21   22   23   24   ...   60




The database is protected by copyright ©sckool.org 2020
send message

    Main page