1. Research Context.
Research is being used widely in all of the fields of science nowadays. There is no way to make any conclusions or support any hypothesis without providing well-grounded research. It has to be systematic and efficient for maximum accuracy of the obtained results. “Research is a systematic search for answers to questions we might explore, investigate, examine, or uncover. The key is that it is systematic. It also is empirical in that it is derived from experiment, observation, or experience.” (Gail, 2002, p.4).
There is no general rule for the choice of research methods in order to test an hypothesis and enhance data interpretation and analysis: “In its rawest form, a great deal of sociological data is textual; for example, interviews, answers to open-ended questions in surveys, ethnographic accounts. How to analyze such data and draw inference from it is a largely open question. (Raftery, 2000, p.654). The paper examines research methodologies, which are not stochastic models, but rather purposive models, with structured designs and targeted sampling in order to balance cost, efficiency, feasibility, validity, interpretation, and precision of the process in obtaining results.
There are many research methodologies from which a scientist can choose, including: sampling of large accessible populations and application of statistical analysis to interpret the results; surveys of targeted populations with written questionnaires; polling of targeted populations; written survey questionnaires to smaller groups; in-depth follow-up interviewing to validate results of large sampling or surveys; standardized interviews with focus groups; individual interviews with experts or elite designated persons; and many others.
Before choosing the proper or most appropriate research methodology (or combination of them), the parameters of the research need to be established. The size of the sampling and the cost, feasibility and analysis of the data collection needs to be considered. The conduct and justification of the primary research focus, e.g., the use of structured, semi-structured, or unstructured questions and interview techniques, should be examined. The questions may be closed, with simple “yes/n” answers, or open, with the possibility of widely differing answers. The research may focus on a qualitative approach, which emphasizes interpretation of individual opinion and analysis, or a more quantitative approach, which uses statistical sampling and analysis of data.
For example, 5-point Likert scales (answers including strongly disagree, disagree, neutral, agree, strongly agree) are oftentimes used to give a quantitative assessment (sometimes in the context of an instrumentality theory with the dimension being rated and a numerical valence – or relative importance assigned to that topic, both of which are multiplied in order to give a quantitative value). Issues of validity, credibility of responses, and reliability of data arise. The overall review of the effectiveness of the research needs to be assessed. Finally, the recommendations for further study or step-up in the scope of the research methods, can be made through techniques of pre-testing and piloting of the research methodology.
There are different types of research. “Research can be divided into two types: scholarly and applied. Scholarly research seeks to understand why the world works as it does by exploring and testing theories… Applied research is intended to solve problems or provide information that can be used to do something.” (Gale, 2002, p.5). Applied research is used much more often because it actually makes great contributions into the realm of science. However, scholarly research is also very important because it lets experts to check theories on which all of the applications are based.
2. Research philosophy and approach.
Maximum validity of results can be achieved due to the wide use of both qualitative and quantitative methods in the survey. It is impossible to conclude that no quantitative research was used during the survey. Qualitative and quantitative methods always follow one another. “Anything that is qualitative can be assigned meaningful numerical values. These values can then be manipulated to help us achieve greater insight into the meaning of the data and to help us examine specific hypotheses.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>)
Qualitative methods are no less important than quantitative methods because they give a very full description of the phenomenon analyzed. However, it has not always been that way. “For a long time it was believed that good research had to use quantitative methods. Sometimes this led researchers to count what was accessible, even if what was counted was irrelevant.” (Gail, 2002, p.5). At present, both quantitative and qualitative methods exist side by side.
Some elements of quantitative research can be applied in qualitative research through the use of 5-Point Likert scales (strongly disagree/disagree/neutral/agree/strongly agree) during the interviews. Likert scales are particularly important for the questionnaire because they enable to measure the level of interviewees’ involvement in the subject. Likert scales are considered a very useful approach in hypothesis confirmation. “A Likert scale (named after its deviser, Rensis Likert, 1932) provides a range of responses to a given question or statement” (Cohen, 2000, p.253). Together with other scaling methods, it gives the researcher the estimation of interviewee’s attitude to the question: “like Thurstone or Guttman Scaling, Likert Scaling is a unidimensional scaling method.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>)
The choice of the optimal alignment among quantitative and qualitative methods of research is very efficient for the validity of research. Many authors argue that it is necessary to balance both types of research: “The “qualitative-quantitative debate” as it is sometimes called is one of those hot-button issues that almost invariably will trigger an intense debate in the hotel bar at any social research convention.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>)
3. Research Strategy and Methods.
Use of secondary research data
It is possible to conduct research on two types of data: primary and secondary data. “Marketing research routinely utilizes many types of data and information. The single most prevalent type is data and other pieces of information that were first collected for another purpose.” (Patzer, 1995, p.5). Despite the wide use of primary data, some types of research are usually based on secondary data, for example, “generally, marketing research relies on secondary data to complement primary data.” (Patzer, 1995, p.5).
Usage of only primary data in the research is in some ways challenging. On one hand, this approach has very important advantages because it enables us to obtain the information, which has not been misinterpreted by other sources. However, it is important to take into consideration the fact that secondary sources are subject to the deep impact of their author. Therefore, it is very hard to obtain an objective point of view from such a source. Primary information received as the result of research methods use can be much more reliable than in the case of secondary information analysis. The most well-known methods of primary research are interview, questionnaire, and focus group discussion. They let specialists obtain the most accurate data about the phenomenon which is being researched.
However, there are some disadvantages of the primary sources of information. Among them it is possible to mention additional time, expenses connected with interviewing and the need to analyze the data after the survey. Those disadvantages are not present in research based on secondary data. As Patzer states, “for marketing researchers and users of marketing research, an advantage of using secondary information is that numerous costly activities in the marketing research process are avoided. These activities are those normally associated with primary data collection and include the following: sample; data collection; data analysis; fieldwork.” (Patzer, 1995, p.17).
Secondary data is one of the most important sources of information in such fields and marketing, finance and others because there is no possibility for researchers to gather all of the information about the market themselves in the short period of time. They come to the results much faster with the help of secondary data which is always available in newspapers, magazines, editions of stock exchanges. All of the information can be analyzed and submitted in their own interpretation instead of gathering the data on their own.
Secondary data is not inferior in comparison with primary data. It is only another source of information for companies and individuals. “Secondary data is a potentially misleading term for people not experienced with marketing research. For example, it is misleading to think of secondary data as being of second importance, minor importance, inferior value, or in any way not necessary.” (Patzer, 1995, p.6). It can offer all of the necessary information to the researcher. “In addition to foundation and context, secondary data provide information about techniques and tools to utilize. The information in this regard can help to determine proper language and terminology with which to communicate with the sample units” (Patzer, 1995, p.9). For maximum efficiency, it is necessary to provide research with the help of both primary and secondary data because they let the researcher to grasp the essence of the problem.
Reliability and Validation of Results
Unfortunately, there is no single research method, which can lead researchers to completely accurate results. Every method includes a possibility of making some kind of mistakes. “The two major disadvantages pertain to relevance and accuracy.” (Patzer, 1995, p.19). However, accuracy is much higher in research methods based on primary data rather than on secondary data: “Lack of relevance and lack of accuracy in marketing research are associated with using secondary data rather than primary data. These disadvantages exist regardless of whether a marketing research project includes primary data. As a result, at times, not only may secondary data be of little or no assistance to a user of marketing research, they can actually be detrimental.” (Patzer, 1995, p.19).
However, it is necessary to note that the validity of the results is not going to be completely accurate due to the imperfections of the sampling process, which can be used in the research. As Trochim remarks, “…you should appreciate that sampling is a difficult multi-step process and that there are lots of places you can go wrong. In fact, as we move from each step to the next in identifying a sample, there is the possibility of introducing systematic error or bias.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>). The most accurate method used in the research may be the interview with an expert, because the opinions of the expert are for the most part based on his/her working experience (although experts may be wrong at times). The least accurate method is perhaps a questionnaire of random people in the street because they give answers only according to their impression of a subject, which is not supported with any knowledge or experience.
Despite the fact that it is impossible to find a completely accurate method of research, it is necessary to mark that the negative influence of survey mistakes can be eliminated with the help of extra research. “It is unwise to think that threats to validity and reliability can ever be erased completely; rather, the effects of these threats can be attenuated by attention to validity and reliability throughout a piece of research.” (Cohen, 2000, p.105).
Research Methods for Collecting Primary Research Data
A popular method of survey is a questionnaire. The reason of this method frequent choice in practice can be explained through the advantages of questionnaire use. “The questionnaire is a widely used and useful instrument for collecting survey information, providing structured, often numerical data, being able to be administered without the presence of the researcher, and often being comparatively straightforward to analyze.” (Cohen, Manion, 2000, p.245). Questionnaire is able to provide answers of many respondents on the hypothesis question. The analysis of this data will give a relatively adequate vision of the question by the respondents.
“The design of surveys must balance many competing goals. Due to the financial burden of recruiting or selecting individuals for studies, many survey questionnaires are formed by pooling questions from existing survey instruments in an attempt to obtain information that can be used for a variety of purposes.” (Grizzle, 1995, p.54). For maximum efficiency of survey, different methods ought to be applied: “To ensure an objective evaluation, the group conducting the survey should have expertise in a wide range of disciplines. Their varied backgrounds will allow them to see potential security vulnerabilities from many different perspectives.” (Roll, 1994, p.65).
However, there are some disadvantages of questionnaire, which need to be taken into consideration. First of all, it was necessary to realize that the results could be not accurate due to the sensitive of some questions to the respondents: “The questionnaire will always be an intrusion into the life of the respondent, be it in terms of time taken to complete the questionnaire, the level of threat or sensitivity of the questions, or the possible invasion of privacy.” (Cohen, Manion, 2000, p.245). It has also been noted by authors that the perception of the interviewer in many ways influences the results obtained: “the heart of the problem of questionnaires- that different respondents interpret the same words differently. Anchor statements’ can be provided to allow a degree of discrimination in response (e.g. ‘strongly agree’, ‘agree’ etc.) but there is no guarantee that respondents will always interpret them in the way that was intended.” (Cohen, 2000, p.252). The same disadvantage was noticed by Patzer: “Questions and words in a questionnaire are added, omitted, or reordered during a research project. The same researcher/interviewer interacts differently with either different subjects or the same subjects at different times. It occurs because researchers are human, their energy, boredom, personal problems, appearance, dress, and liking and comfort with different subjects all differ during a work shift.” (Patzer, 1996, p.25).
Questionnaires can be given on the street, in public, or in formal environment. In large-scale surveys, it is usually recommended to interview at least a thousand people but sometimes, even smaller numbers of people, such as thirty or fifty can be interviewed. The reason why such a relatively small sample can be used is because the standard deviation (or statistical error) of the results based on a random sampling would decrease only marginally with the use of a larger sampling – i.e., the accuracy of the results would not be enhanced by a large sample size. In addition, the sample size can be limited due to considerations of added cost and reduced feasibility of conducting the primary research, and more time could be spent on data analysis.
The questionnaire can be either unstructured or well-structured to offer interesting questions in order to obtain the best results: “The appearance of the questionnaire is vitally important. It must look easy, attractive and interesting rather than complicated, unclear, forbidding and boring.” (Cohen, 2000, p.257). The rule of questionnaire creation needs to be applied for maximum efficiency. For example, a small size of the questionnaire needs to be followed by open-ended questions which allow respondents give full and more extended answers: “Though there is a large range of types of questionnaire, there is a simple rule of thumb: the larger the size of the sample, the more structured, closed and numerical the questionnaire may have to be, and the smaller the size of the sample, the less structured, more open and word-based the questionnaire may be.” (Cohen, 2000, p.251). The advantages of short questionnaires have also been highlighted by other authors. “Pooling questions to make a long questionnaire, although cost-effective, increases the response burden on the individuals sampled and may result in a large no response rate. Several studies show empirical evidence that surveys with long questionnaires tend to have high non-response rates.” (Grizzle, 1995, p.54).
In the questionnaire, different types of questions need to be used in order to raise the respondents’ interest and obtain the fullest picture of the hypothesis question. For the most part, the questions can be closed and thus require only answers yes/no. Such types of questions have both advantages and disadvantages. They are efficient because they let to finish the interview fast and obtain all of the necessary information. However, they do not let the respondent to express his point of view on the question. “Closed questions … are quick to complete and straightforward to code, and do not discriminate unduly on the basis of how articulate the respondents are. On the other hand they do not enable respondents to add any remarks, qualifications and explanations to the categories, and there is a risk that the categories might not be exhaustive and that there might be bias in them.” (Cohen, 2000, p. 251).
In order to ensure the mentioned disadvantages were overcome, open-ended questions can be introduced into the questionnaire. The use of open questions enables the respondents not only to choose the answers from the given list but also make their own contribution into the research. “Open questions, on the other hand, enable respondents to write a free response in their own terms, to explain and qualify their responses and avoid the limitations of pre-set categories of response. On the other hand the responses are difficult to code and to classify. The issue for researchers is one of “fitness for purpose” (Cohen, 2000, p.252). It is also possible to use multiple-answer questions in the questionnaire. Multiple-answer questions can be more efficient than closed questions because they would offer a large possibility of respondents’ creativity. “To try to gain some purchase on complexity, the researcher can move towards multiple choice questions, where the range of choices is designed to capture the likely range of responses to given statements.” (Cohen, 2000, p.252).
Focus Group Discussions
Another research methodology, which can be used in order to test the hypothesis of the study, is focus group discussion. It needs to consist of a number of professionals who have a good idea about the research question. The group can consist of 4-5 people who are involved into the field with which the question deals. After the answers are given by all of the members of the group, a synthesis of the answers can be obtained and used for the interpretation of the data.
A semi-structured interview with the focus group can enable to researcher to obtain excellent results. This type of interview is very suitable for this kind of discussion due to the fact that all the groups of respondents interviewed had different backgrounds. It was very important to ask them basic questions but also compose some other types of questions in the middle of the discussion.
Expert (or Elite) Interview
The interview with an expert in the research question is one of the most efficient research methods. The importance of this method use is dictated by its advantages of communication with an expert face-to-face. “An interview [is] an interchange of views between two or more people on a topic of mutual interest, sees the centrality of human interaction for knowledge production, and emphasizes the social situatedness of research data.” (Cohen, 2000, p.105). Another advantage of interview is that it is a unique and distinct source of information, which reflects the vision of one particular expert in the field. “Each interview is unique, like a small work of art (and sometimes the art may not be very good). Each interview has its own ebb and flow- its own pace.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>)
However, there have also been identified some disadvantages of interviews. “Cicourel (1964) lists five of the unavoidable features of the interview situation that would normally be regarded as problematic: there are many factors which inevitably differ from one interview to another, such as mutual trust, social distance and the interviewer’s control; the respondent may well feel uneasy and adopt avoidance tactics if the questioning is too deep; many of the meanings which are clear to one will be relatively opaque to the other, even when the intention is genuine communication; [and] it is impossible, just as in everyday life, to bring every aspect of the encounter within rational control”. (Cohen, 2000, p.105). If the results of the interview turn out completely invalid, re-interview survey should be applied by authors: “in the response-bias-type study, the re-interview survey is designed to be more accurate than the original survey. This is achieved through the use of “preferred” data collection procedures, such as a more qualified or experienced staff person, extensive probing techniques, or field reconciliation.” (Gastwirth et al, 1996, p.961).
The highly structured nature of the interview provides accurate and reliable data for the testing of the hypothesis. However, there could have been some bias on the part of both the interviewer and the interviewee. For example, the interviewee might have failed to demonstrate his/her attitude to the problem to the fullest due to the limited number of questions. “Unstructured interviewing … differs from traditional structured interviewing in several important ways. First, although the researcher may have some initial guiding questions or core concepts to ask about, there is no formal structured instrument or protocol. Second, the interviewer is free to move the conversation in any direction of interest that may come up.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>) To some extent, unstructured interviews are more creative than structured. They add scope to the general topic. “Consequently, unstructured interviewing is particularly useful for exploring a topic broadly.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>) However, in the case of the analysis of an exact question, it is always much efficient to turn to structured interviews. Besides, it is much easier to analyze the results obtained with the help of structured interviews. “Because each interview tends to be unique with no predetermined set of questions asked of all respondents, it is usually more difficult to analyze unstructured interview data, especially when synthesizing across respondents.” (Trochim, 2000, retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>).
All of the research methodologies discussed in the paper greatly contribute to the confirmation of the hypothesis which the researcher needs to evaluate. The most efficient methods used were focus group discussion and interview with expert, but the questionnaire with random respondents also contributed to this research.
This reliability and validity of the study results benefited greatly from “triangulation”, because it was chosen to examine the issue from a number of points of view: questionnaires – the user, focus group – test some key areas, and interview – the expert/academic/industry. A triangulation approach can be used as an efficient way to reduce the possibilities of bias (systemic error), misinformation and inaccuracy: “the basic premise of triangulation is that the weaknesses of a particular method are compensated for by the strengths of other methods. . . . If data collected through different methods converge, greater confidence in the results is generated” (Statham, Richardson, Cook, 1991, p. 24).
With the help of research methods, it is possible to confirm the hypothesis which has been made by the researcher in the beginning of the study. The results of the questionnaire, interview and focus group discussion research produce useful, accurate, validated, reliable and relevant data which support the conclusions of the study.
Bishop, Y. M. M., Fienberg, S. E., and Holland, P. W. Discrete Multivariate Analysis: Theory and Practice, Cambridge. MA: MIT Press. 1975.
Cohen Louis, Manion Lawrence, Morrison Keith. Research Methods in Education. Routledge Falmer. London. 2002.
Cook, T. D., & Campbell, D. T. Quasi-experimentation: Design and analysis issues for field settings. Boston: Houghton-Mifflin. 1979.
Denzin, N. K. Triangulation. In J. P. Keeves (Ed.), Educational research, methodology, and measurement: An international handbook. New York: Pergamon Press. 1998.
Dillon, J. T. Using discussion in classrooms. Buckingham: England: Open University Press. 1994.
Dillon, J. T. The practice of questioning. New York: Routledge. 1990.
Johnson Gail. Research Methods for Public Administrators. Quorum Books, 2002
Luck Mike, Pocock Rob, Tricker Mike. Market Research in Health and Social Care. Contributors: Publisher: Routledge. Place of Publication: London. 2000.
Patzer Gordon L. Using Secondary Data in Marketing Research: United States and Worldwide, Quorum Books. 1995.
Patzer Gordon L. Experiment-Research Methodology in Marketing: Types and Applications. Quorum Books. 1996.
Suen, H. K., & Ary, D. Analyzing quantitative behavioral observation data. Hillsdale, NJ: Lawrence Erlbaum Associates. 1989.
Statham, A., Richardson, L., & Cook, J. Gender and university teaching: A negotiated difference. Albany, NY: State University of New York Press. 1991.
Adams, L. M., and Darwin, G. “Solving the Quandary Between Questionnaire Length and Response Rate in Educational Research,” Research in Higher Education, 17, 231-240. 1982.
Gastwirth Joseph L., Sinclair Michael D. On Procedures for Evaluating the Effectiveness of Reinterview Survey Methods: Application to Labor Force Data. Contributors: Journal Title: Journal of the American Statistical Association. Volume: 91. Issue: 435. 1996.
Dillman, D. A., Sinclair, M.D., and Clark, J. R. “Effects of Questionnaire Length, Respondent-Friendly Design, and a Difficult Question on Response Rates for Occupant-Addressed Census Mail Surveys,” Public Opinion Quarterly, 57, 289-304. 1993.
Grizzle James E., Raghunathan Trivellore E. A Split Questionnaire Survey Design. Journal of the American Statistical Association. Volume: 90. Issue: 429. Publication Year: 1995.
Long Claudia, Stansbury Kendyll. Journal Title: Phi Delta Kappan. Volume: 76. Issue: 4. 1994.
Nunn Claudia E. Discussion in the College Classroom: Triangulating Observational and Survey Results. Journal of Higher Education. Volume: 67. Issue: 3. 1996.
Raftery Adrian E. Statistics in Sociology, 1950-2000. Journal Title: Journal of the American Statistical Association. Volume: 95. Issue: 450. 2000.
Roll Frederick G. Simple Survey Saves the Day. Security Management. Volume: 38. Issue: 3. Publication Date: March 1994.
Trochim, William M. The Research Methods Knowledge Base, 2nd Edition. Retrieved on May 3rd from source: <http://trochim.human.cornell.edu/kb/index.htm>. 2000.