This section of the research describes the research methodology conducted in order to gather a sizeable and diverse collection of data. The data collection methods or techniques are stated and appraised and the justification for the adopted technique is given. This chapter follows a hierarchical principle and will move through the stages of selecting the appropriate method to answer the research question. In Dissertation Research and Writing for Construction Students, Naoum (2008) states that there are two main types of research methods, qualitative and quantitative. Naoum adds to this by explaining that depending on the aim of the study and the information available will ultimately affect the decision to what research method is chosen.
Quantitative Data
Quantitative research methods are objective in nature. In broad terms this method involves numerical analysis (Polit-Ohara 2004) for example price history within real estate market reports. According to Naoum (2008) quantitative research is used ´when you want to find facts about a concept, a question or an attribute… when you want to collect factual evidence and study the relationship between these facts in order to test a particular theory or hypothesis´. A limitation of quantitative research is the possibility of unreliable or biased data.
Qualitative data is non numerical and tends to steer towards emotions, beliefs and feels about a particular subject area and is therefore very subjective. Qualitative research collects views and opinions of experts who specialize or have an active role in the industry (Polit-Ohara 2004). With the research method qualitative data usually is obtained from questionnaires, interviews, case studies and focus groups.
Mixed Methods approach
The two traditions (quantitative and qualitative) appear quite different... most research does not fit clearly into one category. The best often combines features of each. (King, Keohane and Verba 1994 cited in ´Blending qualitative and quantitative research methods in theses and dissertations - Robert Murray Thomas. Tashakkori and Creswell (2007) define a mixed method approach as ´research in which the investigator collects and analyses data, integrates the findings, and draws inferences using both qualitative and quantitative methods in a single study or programme of inquiry.´ The benefit of a the mixed method aaproach is that the data produced by the different method can be complementary (Denscombe, 2007).
Method Chosen
A mixed methods approach will be used in this dissertation and will allow the researcher to obtain information from qualitative and quantitative methods within both primary and secondary sources of information thus hoping to achieve the benefits of both quantitative and qualitative data. Due to this ´young´ new market there is a relative shortage of quantitative and qualitative information and therefore by using the mixed methods approach will allow the researcher to acquire a wider range of information on which to base a conclusion on. It also feasible to use this method because of the small size of the market and the relatively small number of key actors and the structure of industry, the economy and government in Oman with strategic and policy decisions being taken by senior business and governmental elites of both Omani nationals and expatriate employees. The investigation will seek to find out what those in the industry think, feel and believe about the impact of the policy on the residential real estate market in Oman and this information will be collected through qualitative means. However there is a subjective conception of the impact of the policy and for this reason quantitative data (secondary source) will be used to help contribute to the analysis. Quantitative data will be obtained from real estate market reports more specifically from Cluttons, Hamptons and Savills. Although these reports will help allow the researcher to quantify the affect the policy has had on prices, the reports may be biased or inaccurate hence the use of three company reports. By using a combination of methods the researcher can exploit the strengths of a particular method without leaving him/herself vulnerable to the criticism in connection with that methods weakness (Denscombe, 2007). The next stage will be to appraise and decide what approach should be undertaken.
Primary and Secondary Sources
There are two sources of information primary and secondary. These are described below. The two are not mutually exclusive and can be used in conjunction with each other depending on the nature of the research.
Primary Sources
Primary data collection is necessary when the information cannot be found through secondary sources. Primary data creates original research. There are three different methods within fieldwork - case study approach, survey approach and the problem solving approach.
Case study approach
Case studies are used when the researcher uses in depth analysis of a person, organisation or a particular project to support his/her argument (Naoum 2008). The conclusion of such an approach will not be generalized as it focuses on one particular aspect however this does not mean to say that the method is somewhat flawed as it provides a very in depth analysis of a similar project or problem if such exists.
Survey approach
Surveys such as questionnaires and interviews are used to gather information from a large number of participants within a certain time. The data collected can then be analysed and generalized to give an accurate ´picture´ of the situation. According to Naoum (2008) there are two types of surveys, analytical and descriptive. Analytical surveys aim to establish a relationship between one attribute and another (cause and effect relationship.) for example in the case of this dissertation the relationship between policy change and real estate prices. On the other hand, the descriptive surveys aim to ´answer such questions as: How? Where? When? (Naoum 2008). In a descriptive survey, the number of participants that respond to a particular subject with certain attitudes/opinions are counted. This data is later analysed to illustrate trends.
Problem Solving Approach
In this approach also know as the action research approach the researcher reviews the current situation, identifies the problem, plays an active role in implementing changes and then evaluates the impact of these changes. This type of research tends to be done by practitioners or students such as part time students who are working within the industry and have identified a problem and wish to propose changes to improve the situation (Naoum)
Method Chosen
A main criticism related to the use of case studies is in relation to the credibility of generalisations (Denscombe, 2007) and using residential tourism to promote economic development can vary between different countries given their specific situations in terms of objectives or current economic situation therefore a case study approach would be difficult to generalise from one country to another. The research investigation is not seeking to determine and find a solution to a problem therefore the investigation and the researcher is not in a situation to play an active role in implementing changes there this research will adopt a survey approach as the appropriate research strategy. There are various methods within the survey approach and therefore a data collection technique(s) will have to be selected.
Primary Data Collection Techniques
Data or information can be collected by observation which involves watching people or by asking people through the use of interviews or questionnaires (Robson 1996). The method chosen ultimately depends on what the research aims to achieve and what the researcher thinks is the most suitable way to do it.
According to Robson (1996) simple rules of thumb are available for selecting data collection methods. The rules include:
To find out what people think. Use direct observation
To find out what they do in private use interviews or questionnaires
To find out what they think, feel, believe use interviews or questionnaires.
This research aims to find out what those involved in the real estate industry think, feel and believe about the impact of policy change in Oman. Consequently this research will use the following primary data collection techniques of interviews and questionnaires to establish the reasons behind and the impact of the policy change in Oman.
Questionnaires
A questionnaire is the most widely used data collection technique (Naoum 2008) when conducting surveys. Questionnaires have several advantages, they are cheap, reduce bias error and allow for greater anonymity (Robson 1996). Postal questionnaires have the advantage of covering a wide geographic area and can provide a mass of information. However questionnaires are normally closed ended questions and therefore do not provide flexibility. One fundamental disadvantage to questionnaires is the low response rate with statistics suggesting an average response rate of 26% for all surveys. Even within the responses there is no way to check the reliability or validity of the answers and will not necessarily gave an accurate cross section of the population. Due to technological advances a new sub type of questionnaires has been established; internet surveys. The internet surveys provide a faster and cheaper alternative to postal surveys (Dillman, 2007 cited in Denscombe, 2007). Using the internet the survey can be administered through;
An email - the questions are sent as part of the email
A questionnaire sent in the form of an attachment
A web based questionnaire
As with any questionnaire the response rate for internet questionnaires is influenced by the visual appeal and the ease with which it can be answered and returned (Denscombe, 2007). Internet surveys can be seen to be more difficult than postal surveys in terms of finding the target populations email addresses however this can be done by using ready made lists such as a businesses list of employees email addresses.
According to Sheatsley (questionnaire contraction and items writing 1983) well designed questionnaires should obtain complete and accurate information that helps meet the objectives and aims of the research within the limits of time and resources. A questionnaire needs to be crisp and concise which asks questions crucial to the research (Denscombe 2007). A key aspect to questionnaires is the associated instructions, wording and the response choices (Rea and Parker, 1997 Designing and conducting survey research). Robson (1996) outlines a checklist to help avoid problems in question wording (see below) and ultimately achieve more information to which to form conclusions on.
Interviews
Interviews are another method of obtaining primary data these will involve a series of in depth face to face interviews. An interview can be described as a ´face to face interpersonal situation in which an interviewer asks a respondent questions´ (Naoum 2008 ). There are three forms of interviews; structured, semi structured and unstructured. The wording of the questions will ultimately define the structure of the interview.
Structured Interviews
The questions in structured interviews are predetermined, worded in the same format to all participants and presented in the same order. The questions within this interview technique tend to be close-ended however open-ended questions can be included. A structured format does not have the flexibility and would be harder to gather additional information however it does ensure consistency. Structured interviews make it easier to evaluate and compare the different answer objectively as answers obtained are standardized.
Semi Structured Interviews
Similarly to structured, in this method the interviewer has a number of predetermined questions which will be asked to all participants. However the interviewer does not have to follow the exact question structure as in structured interviews and can depart from these questions if other relevant issues or information. Semi structured interviews are flexible and adaptive as the interviewer and interviewee has the freedom to adapt and respond to the flow of the discussion as it goes on while maintaining and discussing the theme of the project in detail. This allows the interviewee to obtain as much information as possible about the specific issues related to the subject area (Naoum 2008). With semi structured interviews the researcher must be aware of both interviewer and interviewee bias and thus the reliability of the data. Another limitation of semi structured interviews is the possibility of the interviewee/interviewer going off topic during the interview.
Unstructured Interviews
This method tends to be used at the beginning of a research subject and act as exploratory research to find out more about the particular topic. As this method uses open ended questions and there is no structure in terms of format and wording of the questions then the interviewer will need to have a clear understanding of his/her research outline in order to develop relevant questions as the interview goes on. The data obtained from this method will be difficult to analyse and compare as each interview will be different (Naoum 2008).
Method Chosen
Based on Robson´s (1996) view that in order to investigate what those in the industry think, feel and believe about (in Omans case) the impact of the policy on the residential real estate market in Oman interviews or questionnaires. Although questionnaires are cheap and allow the researcher to gather lots of information, there are disadvantages which can relate more specifically to this research topic. As the real estate industry in Oman is relatively small and therefore the number of people involved in the industry is smaller, it may be more difficult to determine who to send questionnaires to. The employment culture tends to be based on a contract basis which means sometimes employees more specifically expatriate employees will be on 2, 3, 4 year contracts and then move somewhere so it is therefore difficult to send questionnaires to people who haven't been involved in the industry long enough or at the right time to talk about the effects of the policy change. Another limitation is the general low response rate to questionnaires and especially this may be amplified as the questionnaires will be sent overseas. Despite the aforementioned disadvantages, questionnaires will be used to help quantify some of qualitative information. Due to the limited postal service in Oman, a web based internet questionnaire will be designed on surveymonkey.com. A main advantage of questionnaires is targeting a large sample population however this research would benefit from more specific information and this will be obtained through interviews. The information gained from questionnaires will be used as a supplement to the information collected through interviews more specifically semi structured interviews. As mentioned above the real estate industry is small in Oman and by using the interview technique it allowed me to carefully select the participants who could provide me with the most amount of information and from different aspects of the industry such as the agency, public sector and developers. The main limitation of interviews relates to the research topic itself and comes in the form of cost, more specifically the cost of flights getting to Oman which ultimately reduces the chances of doing follow up interviews and inevitably puts pressure to get all the necessary information whilst visiting. To combat this, the researcher will use networks of contacts to set up interviews in advance. This will also help reduce the chances that participants were out of the country. All participants that took part in the interviews allowed me to take their email address for any further questions. Phone interviews were considered in order to reduce the limitations of face to face interviews however the business cultural in Arab world prefers face to face contact. There are also associated costs with doing phone interviews albeit they are lower. With face to face interviews and especially using semi structured interviews it is more advantageous to see the participant's reaction and makes it easier to develop a line of questioning. In order to generate valuable information and data, the source(s) or the sampling method will have to be selected.
Sampling Methods
It is important to both researcher and the dissertation to clearly define and clarify the target population. There are many types of sampling methods; the most prominent are detailed below. There are advantages and disadvantages to all however the choice will ultimately depend on the objectives and aims of the research. There are two kinds of sampling; probability (where the probability of the selection of each respondent is known) and non probability (where it isn't known) (Robson 1996).
Probability Sampling Methods
The target population chosen in probability sampling is based on the idea that the researcher believes that these will be a representative cross section of the whole population being studied (Denscombe, 2007). For this reason it is sometimes referred to as representative sampling. Whilst probability samples allow the researcher to ´generalise from sample to population such generalisations are themselves probabilistic´ (Robson, 1996).
Simple Random Sampling
In this method of sampling individuals are randomly selected from a list of the ´population´. This gives every single individual a chance of being selected. The use of random sampling assumes that if there is a large number of the target population selected genuinely at random then the resulting sample will be a representative cross section (Denscombe 2007). Simple random sampling guarantees a sample free from bias, but is often impractical.
Systematic Sampling
Systematic sampling is a variant of random sampling. In this method, the researcher randomly picks the first subject from the population list. Then the researcher will select each ´n´th´ subject from the list (this is referred to as the sample interval. This is more precise than simple random sampling as evenly spreads over the population list however if the population has a specific arrangement then it may not give an accurate representation
Stratified Sampling
The population list is divided into groups and then participants are selected by using a simple random sampling method from each particular group. Stratification will almost certainly give a representative sample of the population however this method can be very time consuming as the categories (population groups) have to be identified.
Non Probability Sampling
Non probability sampling is conducted without knowing whether those included in the sample are representative of the overall population. With this method it may be ´possible to say something sensible about the population from non probability samples but not on the same kind of statistical grounds´ (Robson, 1996). The basis for selecting non probability sampling is varied but Denscombe outlines 3 main reasons:
The researcher feels it is not feasible to include a sufficiently large number of examples in the study
The researcher does not have sufficient information about the population to undertake probability sampling
It may be difficult to contact a sample selected through conventional probability sampling techniques
Within non probability sampling, how and why people get selected becomes an important factor and a defining characteristic of this method in comparison to probability sampling is that the choice of people to be included in the sample is not a random selection.
Purposive Sampling
In this method, the target population is ´hand picked´ for the research as the researcher knows something about the people and deliberately selects people on the basis they are more likely to produce the most valuable data. Denscombe illustrates this by asking the rhetorical question; ´Given what I already know about the research topic, who or what is likely to provide the best information? ´
Snowball Sampling
In snowball sampling (also known as chain referral sampling) participants or informants with who contact has already been made use their social networks to refer the researcher to other people who could potentially contribute further information on the subject. This process continues until a target sample size has been reached or until no new information can be obtained. Although this method does not give representative samples, it is particularly useful when searching for specific information or more importantly specific people that due to their expertise or role within subject can contribute valuable information. This method is compatible with the purposive sampling method and allows the advantages of both to be undertaken.
Convenience Sampling
This method involves choosing the nearest and most convenient people to act as respondents (Robson, 1996). Denscombe makes an interesting point by suggesting an element of convenience is likely to be in most research due to limited time and money. According to Robson this is one of the most widely used and least satisfactory methods of sampling.
Method Chosen
The aforementioned probability methods involve random sampling which due to the specific nature of this research topic these methods would not be relevant to obtain the information needed to analyse and form a strong conclusion. Therefore a non probability method will be used. Using convenience sampling may result in a lack of information, more specifically valuable information. The snowball sampling method will be used in this dissertation with the purposive sampling method being used to select the first participants of the research. Although there may be a greater risk of bias due to subjectivity the data collected through the interviews and questionnaires will be observed in conjunction with the quantitative data.
Secondary Sources of Data
Secondary data is described as ´data collected using the desk study approach… and are obtained from other (already published) sources´ (Naoum 2008). In terms of time constraints and cost secondary sources are a more preferred option. Descriptive documents such as the previous chapters literature review (compiled by the use of books and various publications) are the main type of secondary sources of data. The literature review allowed the researcher to analyse the theory behind and the role of government intervention and identify various aspects/problems which could be investigated further during the interviews. Statistical quantitative data through real estate market reports will also be collected this can provide historic comparative data and thus helping to create a complete picture of the real estate market from the time before the introduction of the Royal Decree through to the present day values of real estate in Oman. It is important to remember that some aspects of the market reports may be biased or prove inaccurate. The weaknesses however can include that some information may not be available or sufficient.
