What is clear to me is that a mixed-method approach is best while researching. The quantitative approach is imperative to give weight and breadth to the research – utilizing a very small sample size of any demographic would minimize the worth of the data analyzed, indicating likely skewed results. Accordingly, such data used for proposals that would affect a large segment of the population would be, at best, problematic. The qualitative approach is more meaningful, in my opinion, in that it requires one to “dig deeper” within the heart of the matter. It allows the researcher to uncover the feelings and motivations of a subject, which can provide a richer understanding of an issue. In short, quantitative data is necessary to ensure a meaningful number of subjects and a large breadth of data, while the qualitative approach can uncover the deeper meaning(s) within that data.
The challenges and opportunities within the social sciences are numerous and diverse, which is natural as studies of humanity are bound to be as complicated and varied as the species they study. The issue of bias is a formidable challenge. As Mahzarin Banaji has shown with her testing of implicit bias, most of us are in thrall of cultural biases and prejudices, regardless of how much we would rather that not be the case. As Banaji stated when her own results were revealed, “’I was deeply embarrassed…[and] humbled in a way that few experiences in my life have humbled me.’” (Vedantam 2005). In addition, academic integrity is essential for the effectiveness of the social sciences in serving humanity, as well as for the reputations of their practitioners.
What I have appreciated during this course is learning how effective quantitative research is conducted. Firstly, I now have a grasp of the nomenclature used within research. For instance, instead of saying “studied subject(s),” I refer to “Unit(s) of Analysis.” Instead of “different types,” I now refer to “variables,” and so forth. And while I’d intuitively understood the differences, having the differentiation of discrete, ordinal, nominal, and continuous data “spelled out” was very helpful.
The explanation of different ways of conveying data results (and which are appropriate to use when) was of great help. For instance, a scatterplot is best when presenting a visual for the relationship between two variables, whereas a histogram or pie charts are more effective for a somewhat larger number of variables.
I must admit that I am still grappling with some of the more technical aspects of calculating and evaluating data. For instance, the concept of correlation fascinates me. The example of possible correlation between height and self-esteem given within the reading was easy to understand (“Correlation,” Trochim). However, the actual calculation that computes correlation was something that took a while to wrap my brain around (mathematics has always been my weakest area, unfortunately). I am determined to master this aspect of data collection and evaluation, though.
The roll of mixed methods in the social sciences is crucial. It is also clear that quantitative and qualitative research cannot be purely objective. As indicated by Onwuegbuzie and Leech, “… in the social sciences the vast majority of measures yield scores that are, to some degree, unreliable. This is because constructs of interest in the social science fields typically represent abstractions…” (Onwuegbuzie and Leech 377). Accordingly, to adopt a dogmatic attitude for or against either approach is inherently wrongheaded.
When researching independent sources in supporting the above statement, I came across a master’s thesis by Julie E. Penner, which dealt with Afghan immigrant women who suffered from chronic physical and/or emotional pain. Penner adroitly synthesized mixed methods in evaluating the experiences of these women. From cultural aspects of Afghan lives that determine their beliefs and coping mechanisms, such as spiritual beliefs in nazar (the evil eye) or jinn (spirits), to medically certifiable psychological conditions such as depression and PTSD, Penner lays the framework for blending the scientific and the cultural. (Penner 69).
Penner displayed sound ethics by issuing pseudonyms for her interviewees. (Penner 15). Displaying a qualitative approach, her interviews with Afghan women émigrés demonstrated how Afghan culture largely stigmatizes mental illness, as well as being a heavily traditional, patriarchal society. (Penner 3-4). Her interviewees also indicated various views of traditional Afghan society – some held strongly to those values, others rebelled. (Penner 39-50).
Penner used a quantitative approach by interviewing several Afghan émigrés, using standardized surveys to gather information for evaluation. In addition, she had the surveys translated into Persian, as needed – again displaying sound ethics and sensitivity in her research (although the Iranian Persian used in those surveys is slightly different than Afghan Persian, aka Dari). (Penner 28, 82). The results of the quantitative data are indicated in tables and histograms which show the respondents’ answers to how they evaluated their pain, general health, social functioning, mental health, and so on. (Penner 28-32). By mixing the qualitative with the quantitative, Penner demonstrated convincingly that there is far-reaching suffering for Afghan women émigrés.
Penner’s thesis is an excellent example of research that combines rigorous ethical and research standards, along with a compassionate approach that appreciates cultural and traditional beliefs with sensitivity. I hope that my efforts can mirror these attributes.
Bibliography:
Onwuegbuzie, Anthony J. & Leech, Nancy L. “On Becoming a Pragmatic Researcher: The Importance of Combining Quantitative and Qualitative Research Methodologies,” International Journal of Social Research Methodology, vol. 8 & no. 5, January 2005, pp 375-387. DOI:10.1080/13645570500402447
Penner, Julie E. “Chronic Pain in Afghan Immigrant Women: An Exploratory Mixed Methods Study.” 2013. University of Saskatchewan Saskatoon, Master’s Thesis. https://pdfs.semanticscholar.org/0821/0f80ecaf68d117fd1ad767a2ba1efaa96530.pdf
“Correlation.” Web Center for Social Research Methods, Trochim, William M.K., 2006, http://www.socialresearchmethods.net/kb/statcorr.php
“See No Bias.” The Washington Post, 2005, http://www.washingtonpost.com/wp-dyn/articles/A27067-2005Jan21.html