Qualitative Research Methodology
This chapter – Research Methodology – outlines the details of the research strategy adopted to implement the empirical research. It also covers the means of collecting data, the sample selection and the framework for data analysis employed. In addition, the reader is exposed to the issue of potential limitations and problems associated with the chosen research strategy and its implementation (Biggam, 2008).
3.2 Research Strategy
The research strategy adopted for the implementation of this study is the qualitative case study approach. A qualitative case study methodology is normally associated with an in-depth investigative study (Cohen & Manion, 1995; Yin, 2003). It allows researchers to study complex phenomena within their contexts using data from variety of sources (Baxter & Jack, 2008).
Case study is one of the three traditional research strategies for real word research, inter alia, survey and experiment (Robson, 1993). Other research approaches exist; Historical, Ethnography (Davies, 1999), Action Research (Reason & Bradbury, 2000; McNiff & Whitehead, 2005) and Ground Theory Research (Glaser & Strauss, 1967; Dey, 1999). However, case study is adopted as a methodological approach for this study because it allows the researcher to probe deeply and to analyze intensely (Cohen & Manion, 1995, p. 106) the research phenomenon under observation – travel time reliability in urban transportation planning (using the road user as a base). Critics of case study approach assert that generalization is not always possible (Yin, 2003; Osuala, 2007). Bassey (1981, p. 85) however retorts that reliability of a case study is more important than its generalizability
This study is primarily qualitative in nature, using elite interviews to obtain primary data. Interview involves a conversation between interviewer and respondent with the purpose of eliciting certain information from the respondent (Moser & Kalton, 1971, p. 271). The use of interviews was harmonious with the overriding research objective of attaining an in-depth and qualitative insight into travel time reliability. This allowed the opportunity for exhaustive discussion with variety of germane transportation stakeholders.
Osuala (2007, p. 171) in his Introduction to Research Methodology, notes: The task of the qualitative methodologist is to capture what people say and do as a product of how they interpret the complexity of their world, to understand events from the viewpoints of the participants. It is the life world of the participant that constitutes the investigative field. Truth within this context is bound to humanistic caprice.
The qualitative researcher is therefore interested in validity of multiple meaning through intense description and explanation of phenomenon as opposed to behavioral statistics associated with quantitative research (Shank, 2002; Osuala, 2007, p. 170). Denzin & Lincoln (2000, p. 3) amplified; qualitative researchers study things in their natural settings, attempting to make sense of, or to interpret, phenomena in terms of the meanings people bring to them. Quantitative research, on the other hand, refers to research that is concerned with quantities and measurements aimed at making scientific generalization (Biggam, 2008, p. 86). However, this research work is interested in context-bound conclusions that could potentially point the way to new policies on auban transportation in Ghana, rather than mere scientific generalization that may be of diminutive significance (Osuala, 2007, p. 172).
It must be noted that a key weakness of the qualitative method is that, is it said to be time-consuming and subjective (Cohen & Manion, 1995; Theodoulou S. Z., 1999; Osuala, 2007). According to Osuala (2007, p. 173) when implementing qualitative research, there is a critical need for the researcher to spend a considerable amount of time in the research setting in order to examine, holistically and aggregately to make meaningful interpretation of research findings.
3.3 Data Collection
This study was interested in capturing qualitative data. However, the study employed both primary and secondary techniques to gather data. The primary source was mainly field interviews.
Because the study is qualitative and seeks to discover what occurs, how and why, the most apposite sampling strategy was nonprobability; purposive (Dey I., 1993). However, other non-probability sample techniques available include; quota sampling, snowball, and convenient sampling.
In the opinion of Merriam (1988, p. 48) purposive sampling is based on the assumption that one want to discover, understand, gain insight; therefore one needs to select a sample from which one can learn the most. The purposive methodologist, therefore, pursues the sampling problem with a predetermined plan in mind. Alternatively, probability sampling is that method of selecting a portion of a population so that each member of the population has an equal chance of being selected (Osuala, 2007). Examples of probability sampling methods include; simple random, stratified, cluster and systematic samplings. The basic distinction between non-probability and probability sample methods is that, the former does not involve random selection but probability sample does (Deming, 1950; Osuala, 2007). Nevertheless, purposive sampling was most appropriate for this study due to the need to reach target research subjects promptly, and given the fact that randomness is not a primary concern of this research project. The researcher thus employed purposive sampling technique to collect qualitative data for the empirical study. The researcher therefore sacrificed representativeness for in-depth analysis. The researcher purposively selected two (2) key respondents from the Ministry of Transport (MoT) and number of passengers to administer interviews. The MoT is the government’s principal transport policy adviser with oversight responsibility of public transportation in Ghana, whereas the passengers are the people the transport operators transport them across the length and breadth of the cities, the unit under investigation. With the aid of semi-structured open-ended questions, the researcher interviewed the Chief Director and the Director of Research Statistics and Information Management of MoT.Â From the passengers at various transport station and by the road site. Interviewing transport operators like GPRTU and MMT staff, aided cross-examination of the responses to ensure reliable and accurate data, and also to minimize possible human errors and biases.
The MoT personnel were questioned primarily on the roles of government in public transportation, and specifically on transportation planning within our city and how they measure congestion on our roads.
The passengers were, also questioned on their travel time and how reliable.
The use of semi-structured questions for qualitative interviews employed open-ended questions see Appendix A for samples of interview guide] to stimulate expressive responses which are relevant to the stated research objectives (Dey I., 1993). This presented the opportunity to discuss, with the various stakeholders, the subject of Ghana’s urban transportation planning comprehensively with flexibility. The interviews were audio-recorded, where possible, to ensure that the analysis of data is based upon exact records. This enhanced reliability and validity of the research findings. The researcher therefore aimed at generating knowledge which is not only valid but reliable as well. According to Bigaam (2007; p. 100), validity relates to how the researcher collects and analyses empirical data, whereas reliability on the other hand focuses on the need for a record of evidence that you did indeed do the research (in a fair and objective way).
The selections of the key respondents from the above named institutions – MoT passengers, MMT and GPRTU were based on their professional expertise and roles in the Ghanaian transportation sector. The sample selection, purposive sample allowed the researcher the discretion to tenaciously select key respondents who were of much use to this study (Merriam, 1988; Osuala, 2007).
Admittedly, the lack of randomisation associated with the chosen sample technique – purposive sampling, somewhat cast a slur on the representativeness of the study. However, in a study of this nature where expert view was required on the research objectives, the researcher controlled phenomena in order to obtain the desired relevant data.
As stated above the researcher was more interested in attaining deeper understanding of the research problem and relates it to other cases than generalizing the research findings. To enrich further the empirical data of the study, the researcher purposively interviewed transportation safety expert from the National Road Safety Commission (NRSC) was interviewed on Ghana’s urban transport system in general; and government’s role on transport safety in particular.
In all, nine (9) respondents participated in this study. The researcher holds the view that the sample techniques, frame and size employed for the purpose of primary data collection were adequate for a study of this nature. While the passengers represents the viewpoint of the case study, the MoT represents the stand of the Government of Ghana. The GPRTU, which is the dominant transport union, constitutes the voice of the private transport providers in Ghana (IBIS, 2005). As indicated above, secondary data were also collected to form part of the analysis. The secondary sources included books, journals article and working papers on transportation. Most of these works were found in the library, Electronic sources, Jstor, Sage, and Google Scholar, were also relied upon. The secondary data, coupled with the interview data, assisted in providing a rich picture of the politics of public transportation in Ghana
3.4 Framework for Data Analysis
A valuable aspect of this research project is the method by which the empirical data gathered was analyzed and interpreted. Data analysis in the view of Hatch (2002, p. 148) is a systematic search for meaning from a pad of information and data. This allows researchers to process raw data, in a scientific manner, so that what has been learned can be communicated to readers (Hatch, 2002, p. 148).
The study employed qualitative content analysis to generate meaning from the raw data. Qualitative content analysis is a sense-making effort (Patton, 2002, p. 453) aimed at interpreting volume of a text data by identifying important themes and patterns through the researcher’s careful examination and constant comparison. This technique was chosen as an analytical tool due to its ability to make faithful inferences (Zhang & Wildemuth, 2009).
The researcher implemented the qualitative content analysis technique through three intellectual approaches; data description, data analysis, and data interpretation. The first step was to develop thorough descriptions of the phenomenon under study. The description provides the context of action, the intentions of the actor, and the process in which action is embedded (Dey I., 1993, p. 32). However, the ultimate goal of analysis was not just to describe the data but to interpret, explain, understand, and perhaps even to predict (Dey I., 1993).
In the view of Patton (2002, p.503-504), a comprehensive and fascinating report provides sufficient description to allow the reader to understand the basis for an interpretation, and sufficient interpretation to allow the reader to understand the description. Data description, therefore, sets the basis for analysis, but analysis also lays the basis for further description (Dey I. , 1993).
The researcher further constantly compared and contrasted the responses (empirical data) with the findings from the Literature Review (chapter two). This process of iterative comparison constitutes the qualitative analysis of the described data (Zhang & Wildemuth, 2009). This was followed by an interpretation of the synthesized research findings – data interpretation. These three interrelated intellectual activities – data description, data analysis, and data interpretation – were employed to aid the researcher in attaining the overall research aim and objectives.
3.5 Limitations and Problems
This research project, like any other study, is not without limitation and potential challenges. In the view of Osuala (2007, p. 37), limitation are those aspects of the research process that, may limit the researcher in making generalizations.
A key limitation that impeded the smooth implementation of the empirical study was the difficulty of getting access to the research subjects (interviewee) to participate in the study. The issue was that the respondents were mostly top administrative officials, and were preoccupied with busy work schedules. However, the researcher instituted measures to establish a close rapport with the respondents to ensure their cooperation. It must be acknowledged that, depending on interview as a main source of primary data relies on personal opinions which is susceptible to biases and inaccuracies (Biggam, 2008). This is due to human factors such as poor memory, interviewee prejudice, and amongst others. Nevertheless, this was dealt with by giving respondents ample time, ahead of the interview schedule, to prepare adequately for the questioning. Again, interview responses were audio-taped, where possible, and later transcribed to avoid distortion of memory on the part of the interviewer. It also aided the interviewer (researcher) to focus on the questioning to avoid any divided attention. The researcher was also aware of the tendency of the interviewees withholding vital information or even withdrawing from the whole research process because of fear of exposure. To address this possible challenge, ethical practices – assurance of anonymity, inform consent were duly observed to repose some confidence in the researcher. This was taken care of by the introductory section of the interview guide. Refer to the introductory section of the interview guide in Appendix A. Nevertheless, most of the interviewees did not mind to be identified with their responses. Again, the use of a non-probability sampling technique purposive sampling to select respondents appears to cast a slur on the representativeness of the study. But the researcher upholds that the views of the respondents were strategic in capturing the relevant data for the research objectives. This is due to their professional and civic expertise in Ghana’s public transportation matters.
The study was also limited by time constraints. The use of interviews, as a main tool, to collect primary data was time consuming. It was also difficult to explain phenomenon by relying on individual subjective opinions (Theodoulou S. Z., 1999). The adopted strategy case study usually makes it difficult and almost impossible to draw any scientific generalization. However, the case study approach provides us with in-depth understanding of the research objectives rather than mere generalization. The in-depth understanding of the research objectives thus compensates the inability to make any generalization. This chapter has provided the details of the research methods (and their justifications) employed in this study. It has also addressed the potential limitations of this research and clarified the steps used to minimize such potential limitations.