Recommended during latter phases of research projects.
Recommended during earlier phases of research projects.
Projects in different areas are not linear in approach.
All aspects of the study are carefully designed before data is collected.
The design emerges as the study unfolds.
The study of a drug interaction in x population is different than a study of language variation in y population.
Researcher uses tools, such as questionnaires or equipment to collect numerical data.
Researcher is the data-gathering instrument.
Sometimes subjectivity is at risk.
Data is in the form of numbers and statistics.
Data is in the form of words, pictures or objects.
Audience and style needs.
Objective -- seeks precise measurement & analysis of target concepts, e.g., uses surveys, questionnaires etc.
Subjective - individuals' interpretation of events is important, e.g., uses participant observation, in-depth interviews etc.
Both important to tell the entire story -- both sides of the picture.
Quantitative data is more efficient, able to test hypotheses, but may miss contextual detail.
Qualitative data is more 'rich', time consuming, and less able to be generalized.
Sometimes quantitative details are difficult to extrapolate into general populations; sometimes qualitative data lacks depth and robust proof.
Researcher tends to remain objectively separated from the subject matter.
Researcher tends to become subjectively immersed in the subject matter.
Again, subject matter and issues of subjectivity and potential bias.
(Source: Neill, 2007)
It is true that often the two sides seem antithetical to each other, however, rather than framing the methodological rubric as "Quantitative vs. Qualitative," it might be best to focus on the manner in which the techniques might be integrated and allowed to flush out the limitations of each (Booth).
1.6 Mixed Method Research -- as more and more social scientists began exploring the use of quantitative research to answer some basic question, it became apparent that a bit of a constructivist approach, or positivistic, tended to emerge. This often allowed the interpretation of pure data, but failed to convince in the extrapolation of that data into the population. Additionally, the manner in which the research was reported became an issue, with both sets of purists from either side vehement about their own methods.
The sensible approach, then, is to mix the methods and provide researchers with a clear alternative to siding with one camp or the other -- but to let the data and the particular subject matter and hypothesis rule the discussion. Indeed, what makes far more sense is to remove the philosophical issues surrounding the type of research and concentrate on the logic of justification -- the logic of the data set, and the combination, in varying degrees, of both types even within the same study (Onwuegbuzie, et.al., 2004)
Figure 1.4 Diagram of Mixed Method Research
(Johnson and Onwuegbuzie, 2004).
Chapter 2 -- Literature Review
2.1 - Quantitative research was originally used as an approach to studying the natural sciences. Quantitative research focuses on quantifiable data and conclusions. In other words, numerical data and measurements are used in order to draw numerical conclusions from the research. This approach lends itself particularly well to the natural sciences and numerical disciplines, as exact measurements are required for these (McBurney and White, 2009). Natural scientific language is used for the expression of facts in numerical terms. These digits should be universally acceptable in the scientific research environment, which means that its function should also be unchanged. Such language is then to provide research facts via an explanation of the statistical truth (Hara, 1995). Furthermore, the researcher's value judgments are not to be imposed upon the conclusions drawn from the data. The researcher is viewed as an objective observer. While most appropriate for natural scientific research, the quantitative research has also increasingly been used in social sciences in order to draw logical conclusions from existing facts (Haynes, 2007).
Being a social science, education has generally been studied by qualitative methods. The reason for this is that human beings are not an exact science, and facts have to be derived by an inductive rather than deductive approach. Best practice could therefore best be determined by making value judgments rather than drawing quantitative conclusions (McBurney and White). It has however been recognized that the social sciences can also be studied from an objective rather than subjective viewpoint. According to Hara, quantitative research can be used in education in order to discover existing facts by means of isolating human reality from the researcher's subjectivity. The data obtained in this way can then be presented in a numerical, objective way, in order to derive universally applicable conclusions. The value of such research is that it exists independently from values such as time, place and culture. Hence the conclusions can be applied to any educational setting (Drisco, 2008).
Hara cites two reasons for the use of quantitative research in education; the first is that the concepts behind quantitative research, as well as its methods and purposes, were appropriate to education research, and the second that it could provide logical standards for the study of education (Hara).
Education research can benefit from the quantitative paradigm precisely because it is so often a subjective field. Teachers and lecturers tend to take a subjective approach when working with their students. This is logical, as any field where human beings are involved will necessarily be subjective. However, the field of education should particularly be standardized, because it focuses on preparing the labor force of the future (Thomas, 1998). Future employers would therefore wish for a particular standard in their employees. This standard should be quantified. This quantification process can be achieved by applying quantitative research principles in education. The general subjectivity of the field can therefore be balanced by a more standardized logical approach. Logical facts discovered by these methods can then be used as a basis for personalized, subjective, and individual teaching approaches that each teacher considers appropriate for his or her classroom. The correlational approach involves the consideration of data to determine the degree to which two or more quantifiable variables relate to each other (Johnson, 2001).
This relationship is then used as a basis for further quantitative study. This is particularly useful in studies such as those focusing on education, as students' performance can be correlated with other influencing factors such as previous performance throughout the year. This can then be used in order to establish methodologies that would be more effective for future practice.
An advantage of this method is that it forms a good initial basis for further research. It provides the researcher with a sound springboard for further study. It can also produce useful data for retrospective analysis. A good correlation between variables can for