Introduction
Saunders et al (2009) assert that an onion can be used to represent the research process (see Figure 1). The layers of the onion must be unwrapped before one can reach the innermost core of the onion (the data collection and data analysis). The outer layer comprises of philosophies and research approaches. The middle layers represent research choices, approaches, and time horizons. The inner core represents the research methodology (procedures and techniques). Thus, the research onion depicts the research strategies and approaches that will be employed in this study. They are discussed in more detail below.
Research Paradigm
The epistemological and ontological paradigms are the foundations on which research endeavours are based. They play an essential role in research since “they shape the approach to theory and method” (Marsh & Furlong 2002, p. 17). Epistemology is a theory of knowledge that mirrors the researcher’s view of what he/she can discern about the world (Marsh & Furlong 2002, p. 19). There are two distinctions that can be deduced with respected to epistemology. First, it is feasible to gain knowledge regarding the unmediated world devoid of interferences. Second, observation is not objective but constantly influenced by the social constructions of reality (Schmidt 1994).
On the other hand, ontology is the theory of being. It addresses the question of how the world is constructed (Marsh & Furlong 2002, p. 18). Two fundamental distinctions can be made here. First, there is a real world that is free from our knowledge, and that life is built upon these foundations (foundationalism). Second, there is no real world but the world is discursively and socially constructed and therefore reliant on a particular culture (Hay 2002, p. 61).
For instance, positivism assumes foundationalist ontology and a resultant epistemology. It emerged from the empiricist practice of natural science and views social science as competent of the same potential that is manifested in natural science. Thus, it is feasible to examine everything that occurs and comprehend it as devoid of any mediation thus denying any reality/appearance dichotomy. It can thus be used in natural science research to create hypothesis that can be tested via direct observation (objectivity). Positivity typically employs quantitative techniques since they are objective and the outcomes can be generalized and replicated (Marsh & Furlong 2002, p. 19). In essence they search for explanation of behaviour and not for the meaning. This study will be based on an objective reality since quantitative methods will be used to generalize the outcomes.
Quantitative techniques are typically used by positivists to generate outcomes in numbers which are later on analyzed for accurate results. Since this study used a closed-ended questionnaire to gather data from participants, utilizing a positivistic paradigm is deemed appropriate because it lends credence to objectivity thereby minimizing bias in research endeavours. Consequently, this research will employ a quantitative approach since the data collected is quantifiable and thus can be generalized. The rationale for this choice is to produce direct and precise causations that are indisputable. What’s more, the main advantage of using the quantitative approach is that the data is typically easy to replicate and easy to generalize (Hansen et al. 1998, p. 95).
The qualitative approach is not suitable for this study because it employs focus groups, interviews and other qualitative techniques to gain an in-depth knowledge into a topic. Qualitative approach is also limited by the fact that it is difficult to measure the data generated in terms of validity, reliability and generalizability (Gavin 1998, p.172; Lunt & Livingstone 1996, p. 90).
Deductive and inductive approaches are two methods of reasoning used in research. With respect to deductive technique, analysis starts from broad observations and moves towards precise facts. Conclusions are derived from existing facts. Under the inductive technique, analysis starts from precise facts and moves towards general theories. Conclusions are based on premises and subject to a degree of uncertainty. Arguments grounded on acknowledged rules, laws and principles are commonly employed in deductive reasoning while observations are typically employed in inductive arguments (Burney 2008, p. 7). Consequently, this study will employ inductive approach since data will be collected from specific participants then generalized using appropriate quantitative methods.
Research Design
Research Methodology
This study employed two types of instrumentations: faculty survey and police force assessment (Patton 2002). Faculty survey was done within the police department to determine the opinions of police officers on transformation in leadership style. The survey approach was selected because it is considered as one of the key areas of measurement in applied social science. According to Kuter and Yilmaz (2001), a survey is a technique used to collect data from study participants. A survey can be used to collect views from individuals or it may dwell on practical data about study participants. A non-experimental one-shot survey was thus employed to collect data from participants. The survey instrument was a closed-ended questionnaire consisting of 15 questions that sought a variety of opinions from police officers regarding the implementation of transformational leadership within the police department.
Concerns have emerged in the recent past regarding the validity of survey research. A survey research may be susceptible to causal interference (CI) and common method variance (CMV) (Rindfleisch et al. 2008, p. 261). Since cross-sectional survey is susceptible to CMV bias, this study will employ longitudinal survey whereby data will be collected from participants over a three-month period to reduce CMV bias.
Research Method
Data Collection
There are two types of data used in this study. Primary data (collected for the first time) and secondary data (data that has already been gathered and analyzed by another person). The survey approach was used to collect primary data from the participants enlisted in this study. Close-ended questionnaires were sent to the selected respondents via online to capture their opinions regarding the implementation of transformational leadership within the police department. Permission was sought from the relevant authority within the Homicide Division in order to conduct the survey. Secondary data was also used to provide a wider scope regarding the topic under investigation. Secondary data for this study was obtained from books, journals and periodicals found online and from the University library.
The researcher conducted a pilot study to shed light on the potential bias that might emerge during the main survey. The pilot study entailed interviewing a sample of police officers within the Homicide Division regarding the issue at hand. The researcher employed a triangulation method whereby data from both the pilot study and main survey were mixed to provide diverse perspectives on the topic. The triangulation entails mixing different types of data (i.e. survey data and interviews) in order to validate the claims that might emerge from the pilot study.
Sample Size and Sampling Methods
The targeted population for this study was limited to police officers from the Homicide Division. A total of 350 police officers from the Homicide Division were randomly selected to participate in the study. The researcher use stratified random sampling method to minimize bias from the sampling technique (Lee & Wang 2003). The stratified random sampling method suggests that the researchers first collect unbiased sample before extending the number of the group. The sampling approach provided the required sample (n) for a population (N) up to infinity and was adapted for estimating the sample size. In this study, the entire population size was N=5000 and the sample size was n= 350.
The response rates from mailed questionnaires are usually low and a 30% return rate was deemed acceptable for this study (Cavana, Delahaye & Sekaran 2001). A conservative rule of thumb for testing R-square is n>50+8m, where: n=sample size, and m=number of predictors (Tabachnick & Fidell 2001). Consequently, the minimum number sample size in this study must be more than 186 for the results to be acceptable.
Data Analysis
The researcher used the quantitative approach in data analysis since the data collected was quantifiable. Descriptive data analysis was used to describe the data and to serve as the first phase in understanding the results gained. Later on, the data was reduced to descriptive summaries for the mean, standard deviation and correlation. Various graphical techniques (i.e. scatter plots, frequency distributions and histograms) were employed to visualize the summarized data. Finally, inferential statistical analysis was employed to test hypotheses regarding the opinion of police officers and transformation in police leadership style. In order gain a better understanding of the research population, participants were described with respect to gender, age, years of experiences, and income levels. In addition, the Pearson’s coefficient of correlation and the T-test were used to analyze the relationship between variables. Furthermore, SPSS was used in data processing. The reliability and validity of measurement scales were also scrutinized and reported.
Limitation to Methodology
There are several limitations associated with the methodology used in this study. With respect to the survey method used in data collection, some of the limitations encountered were: respondent’s reluctance to answer questions posed by unknown interviewers on issues they deem private; busy participants unwilling to make time for the survey; and some participants may answer to appear smart. Some of the limitations associated with online questionnaires are low response rate and researcher’s lacks of control over who answers. However, the study was based on some assumptions: All the respondents would participate willingly; answers given would be honest; and the response rate would meet the minimum threshold for the results to be accepted.
Ethics
Prior to conducting research, an informed consent was sought from the participants. The consent letter had details about the research procures, research instruments and potential benefits and risks involved in participating in the research. The participants signed and submitted the consent letter before taking part in the study. Participants were promised that their responses would be kept confidential. In order to ensure confidentiality, the data will be stored electronically on a password protected computer to prevent unauthorized access. In addition, the data will be destroyed after five years.
References
Burney, S 2008, Inductive & Deductive Research Approach, University of Karachi, Karachi.
Gavin, N 1998, Generalization, reliability and validity in media research, Leicester University Press, London.
Hansen, A, Cottle, S, Negrine, R & Newbold, C 1998, Mass Communication Research Methods, New York University Press, New York.
Hay, C 2002, Political analysis: a critical introduction, Palgrave, Basingstoke.
Kuter, U & Yilmaz, C 2001, Survey Methods: Questionnaires and Interviews, University of Maryland, Maryland, USA.
Lee, E & Wang, J 2003, Statistical methods for survival data analysis, 3rd edn, John Wiley & Sons, New Jersey.
Lunt, P & Livingstone, S1996, ‘Rethinking the Focus Group in Media and Communications Research’, Journal of Communication, vol. 46 no. 2, pp. 79-98.
Marsh, D & Furlong, E 2002, Ontology and Epistemology in Political Science, 2nd edn, Palgrave, Basingstoke.
Patton, M 2002, Qualitative research and evaluation methods, 3rd edn, Sage Publications, California.
Rindfleisch, A, Malter, A, Ganesan, S & Moorman, C, ‘Cross-Sectional versus Longitudinal Survey Research: Concept, Findings, and Guidelines’, Journal of Marketing Research, vol. 55, pp. 261-279.
Schmidt, J 1994, The reality of the observer, Westdeutscher Verlag, Opladen.