Quantitative research is classified as the systematic empirical investigation of properties that have quantifiable elements and or phenomena, as well as the relationships between these. In general it can be said that quantitative research is usually carried out using scientific methods, and these can include and encompass, generating models and theories, gathering empirical data, developing measurements and methods of doing so amongst other approaches. The roots of quantitative approaches in a formal framework go back to the philosophies of positivism developed by Auguste Comte in the 1830 and 40’s. Comte (1856) basically sets out positivism as meaning the uncovering of the laws by which both physical and human events occur through the use of scientific methods, although it is more commonly now thought of in sociology as studying society via the use of scientific methods. Comte was a French philosopher, who came up with the term of sociology and he envisaged one universal law that worked in all of the sciences, he called this the ‘law of three phases’. According to Comte this law is that society has gone through three phases: Theological, Metaphysical, and Scientific. It was to the last of the three phases that he gave the name “Positive” to because of the many meanings and connotations of the word. Furthermore, the 5 underlying principles set out of the positivism philosophy are that:
The scientific method as it is typically presented is that scientific research follows a linear path: This being that from an initial research question it then proceeds through observation, to the formation of a hypothesis, subsequent experimentation, before finally producing results and a conclusion. The scientific method is the approach by which researchers are able to make absolute or conclusive statements about their studies, without a significant influence from any bias.
Peter Achinstein (2004) covers the history and development of the Scientific method in thorough detail in his book on the subject, Science Rules. Briefly, the development of the scientific method has roots that go back to ancient times, and first evidence of the documented formal approach over philosophy goes back to Aristotle, who by many is regarded as the father of science. It was Aristotle who first proposed the idea of induction as a tool or method for the gaining knowledge and understanding. In addition he understood that the use of abstract thought and reasoning must be backed up and supported by real world findings to have merit. He was the advocate of the use of empirical data to back up his findings and made meticulous observations to record his investigations. The Muslim scholars of the 10th to 14th centuries are widely regarded as the next to exploit and define the scientific method, with scholars such as Al-Haytham using a scientific method very similar to that of today’s researchers. These scholars also understood the importance of repeatability of results, publication, and even peer review. However, it is through European scholars such as Roger Bacon in the thirteenth century, that the refinement of the scientific methods took place and that they began to become formalized to the form they are known in today. It was Bacon who established the idea of making observations and using these to develop a hypothesis before then experimenting in order to test and prove or disprove the hypothesis. Fundamental to all of this method was the meticulous documenting of his experiments so that his experiments could subsequently be repeated by other scientists to confirm the results as independent verification. Subsequently, around the time of the 1600’s scientists such as Francis Bacon put further emphasis on the importance of the process of induction as part of the scientific method. At the same time, in contrast to this Descartes put forward the belief that the universe was like a giant machine and therefore that if the basic laws of the universe were understood then you could from these deduce how anything will happen, act or occur. However, Galileo believed that it was impossible for researchers to conduct experiments which would be able to take into account every single variable that was possible in some circumstances, but that it could be possible to build up a body of evidence that would allow results to be extrapolated to the general theory if independent researchers could repeat experiments and reproduce results. So the elements of inductive and deductive thinking were both firmly established. The period of the sixteenth and seventeenth centuries is referred to as the Scientific Revolution, and it is during this time that many more elements required for the modern scientific method were postulated and introduced. The Royal Society, set up in 1660, declared that experimental evidence always supersedes theoretical evidence, which is one of the foundations of modern science. With The Royal Society came the installation of a group of experts, along with for the first time the founding and publication of scientific journals, then this led to peer review of advancements and theories. Finally it was Newton who really understood for the first time that both deduction and induction were needed and important in the scientific method together. Others have added other elements especially in the 20th century but essentially the method is unchanged.
The processes of the scientific method are sometimes also considered as an hourglass; starting from broad general questions, before subsequently focusing down to one specific area or subject, to then designing appropriate research where we can observe and analyse this detail as widely as possible before finally drawing conclusions and generalizations to the real world. However, it is important to be aware scientific research does not always proceed linearly. Science is not a straightforward linear process or progression, for example it does not necessarily have to start with an observation or a question, and it is not unusual that it does not even involve experiments. The experiments can be studies, comparisons, modelling, descriptive or other means to generate data to support the area of investigation. In the real application of the scientific method it is demonstrated to be a much more dynamic and robust process.
Fundamentally the practice through which questions can be addressed scientifically is the basis of the scientific research methods. Whatever the approach taken then it is though these methods that data is produced which can be analysed and from this interpretation be drawn out to give ideas in science such as hypotheses, theories, and laws. It is important that scientific ideas are developed and disseminated usually through literature in journals and conference proceedings where independent or even opposing views can discuss the interpretation and significance of the results. Over time, as more results and data evidence add weight to a theory or finding then it moves to become a part of the body of knowledge in science and also then feeds back into the research process again. Capri and Egger (2003) present an excellent graphical overview of the real processes and steps in the practice of the Scientific Method, as can be seen in Figure 1 below, which clearly illustrates that the scientific method is not simply a linear approach.
Qualitative research by comparison can be classified as the gaining of an detailed understanding of human behavior and the underlying explanation of the things that govern that behavior. Qualitative research refers not to ‘non-quantitative’ studies in general, but only to those data centric studies which are adopting systematic data collection and analysis techniques. Qualitative research was only used in the fields of anthropology or sociology until the 1970s. However after this time qualitative research began to be used in far more areas, becoming a prominent method of research in the fields of education, social work, gender studies, disability studies, psychology, and the like, as well as in many other fields. It began to be seen to be used in the consumer products industry as researchers investigated new consumer products and advertising studies. When qualitative methods are used to gain a deeper insight into the background of decision making, then smaller but more focused samples are often needed and used rather than large samples that may been seen in quantitative work. It is important to note that qualitative methods will only produce information the specific cases being studied, and any more general conclusions are only propositions based on the information. It was around the time of the late 1980s and early 90s that new methods of qualitative research began to emerge and develop to address the problems that were perceived to lie with the reliability and imprecise modes of data analysis that were often the source of criticisms from the advocates of quantitative research methods. There are many non-overlapping subcategories within qualitative research methods, and as Dittrich et al. (2007) make note, qualitative research ”may come in many different flavours aˆ¦ be used under different epistemological paradigms, and with different theoretical underpinnings”.
Qualitative data analysis can take a many forms but often it focuses more on words, language, signs, signals and meaning. The holistic and contextual analysis of the research is far greater significance in qualitative research as opposed to quantitative where the analysis is often seen to be isolationist and reductionist to get to the basic underlying law. However both methodologies require a systematic and transparent approach to analysis to ensure appropriate academic rigor. Qualitative researchers tend to follow varying approaches in collecting data for their studies, which can included techniques such as the grounded theory practice, storytelling, classical ethnography, text analysis, narrative accounts, verbal reports, or shadowing. The methods for collecting data for qualitative research can include any or combinations of the following, with the methods of participating and observing varying, including participant observation, non-participant observation, field notes, reflexive journals, structured interview, semi-structured interview, unstructured interview, and analysis of documents and materials.
In contrast to quantitative research then qualitative research test cases can be selected on purpose according to whether or not they typify certain characteristics. Also the researcher’s role in the research receives greater critical attention as the possibility of the researcher taking a neutral or passive role is more problematic in practical and/or philosophical terms. As a result of this the qualitative researcher should reflect on their role in the overall research and examine this as part of the analysis. Possibly the most important issue in qualitative research is validity which is described by Lincoln and Guba (1985). There are many possible different ways of establishing validity to the research and these can include:
Qualitative research is often used to develop a general sense of an event, situation or occurrence and to allow theories to be formulated which can be tested using yet further quantitative research studies. It is also often used to formulate ideas and generate a hypothesis, and also occasionally for explaining unusual quantitative results. Quantitative methods are used to then test hypotheses. However, both the quantitative and qualitative methods can sometimes be classified into empirical studies; these are characterized by systematic gathering and analysis of data. Research may be classified as qualitative if detailed analysis of long stretches of texts is relied upon, but quantitative if statistics are predominantly relied upon for the study. Furthermore, below the categories of quantitative and qualitative, the research may also be sub-categorized as being pure (e.g. only using quantitative or qualitative methods) or mixed (e.g. combinations of the two for example, quantitative supplemented with qualitative). Kuhn (1961) concludes that “large amounts of qualitative work have usually been prerequisite to fruitful quantification in the physical sciences”. A very good summary of the comparative strength and weaknesses of quantitative, qualitative and mixed methods is set out in the paper by Johnson and Onwuegbuzie AJ (2004), whilst setting out the case for and the benefits from application of a mixed method approach when applied to education research.
One of, if not, the most important elements in the success of carrying out the research is the correct choice of research methods that will be used to answer the specific research question in a particular field of study which is discussed by McNeill (2005). McNeil concludes that while significant attention can be paid to theory within the methodology that the researcher will employ, it is also vital to take into consideration the research tools actually used to develop the theoretical decisions that are made and used to provide the raw data for which the hypotheses is to be tested. Ruane (2005) describes how frequently the choice of research tool is not correctly assessed, which results in data being generated that is of little or no use to the research question and the researcher. There are many reasons that this can occur which may be in part due to the research method not satisfying the requirements of the research or even the incorrect use or application of the chosen research method itself. In addition, the capability, experience, bias or ability of the researcher themselves often has a bearing on the choice of research methodology and not by the objective requirements of the problem. This can in terms of time, cost or other related factors, which can be things such as the researchers familiarity with the chosen processes and techniques that are utilized within certain areas of research.
In order to carry out the projects they work upon then engineers are required to have knowledge and understanding of the appropriate sciences behind these areas. Therefore as an engineer will work on many areas and projects throughout their career they are continually learning new information and material throughout this time. It is not unusual in engineering for multiple different solutions or options to exist to the solution of a particular project, and therefore as a matter of course an engineer will assess and weigh up the advantages and disadvantages of different design decisions, before choosing the solution that best meets the particular requirements. This is the crucial and important, and some would say unique, task of the engineer to identify, assess, and interpret these design constraints in order that they can develop a successful end product. Usually it is not simply adequate to develop a technically proficient product, but also one that must meet many other constraints imposed upon it. These can be anything from the technical aspects, physical restrictions, available resources or materials, but also many such consideration as cost, ability to service the product, the manufacturing and assembly constraints, safety, and also environmental, as well as a host of other possible considerations. These constraints can then be used by the engineer to develop specifications for which the product or system has to be designed, manufactured, or operated within. In doing so an engineer will draw upon their knowledge of mathematics, economics, experience, logic, as well as science, to find suitable solution, or solutions, for the problem. Along the way they will often create mathematical models of the problem or part of the problem in order to help them analyse the problem and to test the solutions or aspects of the overall solution. The multiple solution possibilities is a very common occurrence in the disciplines of engineering. In order that the engineer can predict how their solution may perform prior to final production they will often use methods such as scale models, simulations, prototypes (both physical and virtual), stress analysis, and testing which can be both destructive and non-destructive, with the testing ultimately ensuring that the chosen solution performs to expectations and collates to prediction.
From a pragmatic viewpoint engineering as an applied science exists firmly in the quantitative arena, as has been described in the preceding sections. Therefore the research methods debate becomes a philosophical view of qualitative versus quantitative for the social scientists.
As an engineer my work is embedded in the in the quantitative camp, and as such engineers are “positivists” not “hermeneutics”. This is not to say qualitative research methods have no place in Engineering, indeed they have been developed and are commonly used to handle the complexity of issues involving human behaviour, therefore where the engineering is more associated with human behaviour or how this interacts with the product being engineered then it clearly is appropriate.
However, for the purposes of this research project, with an underlying theme of welding and distortion engineering, it is not a matter of how we feel about these issues and phenomena but how we address or develop them. The results and phenomena occurring are quantifiable, factual, and not the subject of opinion, so the appropriate research methods will be dominated by quantitative methods, and primarily explored using a scientific methodology approach. It is envisaged that in order to pull the multidisciplinary and wide ranging projects looking into the phenomena associated with welding and distortion together into an holistic framework then a Case study approach will be applied to develop the individual subjects of investigation into the cohesive thesis body of work.
As discussed the scientific method has continued to evolve over the ages to ensure that scientists make, new, relevant and meaningful discoveries that have foundations based upon reason, science and fact rather than upon pure emotion. Between the differing scientific disciplines the exact scientific process may differ, but generally all scientific research methods follow the principles discussed earlier of observe – predict – test – generalize. Given the quantitative nature of the work then following this broad guideline is appropriate to the areas of investigation.
In drawing multiple projects together relating to a common underlying theme then it will be appropriate to use case study research. Where case study benefits strongly is at allowing researchers to develop an understanding of a complex issue, which can extend experience or add strength and body of knowledge that is already known through previous research. Specifically case studies make emphasis on the detailed analysis with a specific context of a limited number of events or conditions and the underlying relationships involved. The case study research method has been used for many years by researchers working across a wide variety of multiple disciplines. One area in particular where it has been used is in social sciences, with researchers making use of this qualitative research method to examine current real-life situations. This has then been used to provide the basis for the application of ideas and extension of methods. Yin (1984) defines the case study research method as an empirical inquiry that investigates a contemporary phenomenon within its real-life context; when the boundaries between phenomenon and context are not clearly evident; and in which multiple sources of evidence are used. Furthermore, Yin states that it is appropriate to use the use case study method when:
The essay has set out the basic types of research methodologies, looked at the basics of how, where or when these are applied, along with some of the principles. It has been shown that quantitative methods predominate and are employed within my area of practice in my professional career in engineering.
Within engineering specifically relating to subject areas where the results are quantifiable and measured empirically then the scientific method is the research methodology underpinning research into those areas. Therefore it is appropriate to use the scientific method to individual investigations to determine the impact of certain elements and processes on the distortion process that occurs in welding. Furthermore there are many areas and methods to address distortion, and during my professional practice many disparate approaches and techniques have and will be used to control or measure distortion. Therefore these individual projects will be drawn together into a holistic model using a case study framework.
It has been shown in the previous section that the impact of distortion upon the processes employed in the engineering industry is considerable, simply by looking at one straightforward example highlights the high cost of failure associated with this phenomenon. This all impacts on the bottom line profitability for manufacturing companies and reduces efficiency and competitiveness, therefore it is a strategically important area to address if companies involved in fabrication are to be more efficient and continue to optimise their processes.