Written by Natasha Somji
For some time now, I have been debating whether or not to stay within the world of academia. While I love theories and conducting analysis, there is a sense of discomfort I feel whenever I consider pursuing a PhD that is not entirely captured by the monumental monetary costs or confusion about what I want to study. Much of this squeamishness has to do with questioning whether I buy into the mainstream version of academia: a field that uses theories and models to predict outcomes and then validates these predictions through quantitative studies. I find myself sometimes doubting the purpose of academia and how it is executed. If models were constructed with a particular audience in mind and studies are run on certain samples, how can these be generalisable to new contexts? Can studies that employ quantitative techniques – widely hailed as being more credible in substantiating results – really capture the subjectivity of the human experience?
Much of academia relies on using numbers to validate theories and models. The problem comes in when these same studies view people as mere numbers and do not consider more qualitative aspects of an individual or the culture in which models are operating. After all, is humanity not about emotions and subjective experiences and is not every individual unique? Yes, we may be able to measure how social benefits reduce employment rates, but how do we measure the loss to human dignity? Unfortunately, studies that are considered ‘credible’ require a quantitative component; however, it is just as important to recognise that the subjective experience of being human often cannot and should not be quantified. Of course numbers are important in substantiating results, but perhaps it is time to add humanity to our analysis by considering how qualitative factors impact or are impacted by results.
A recent article in The Christian Science Monitor titled Immigration: Assimilation and the measure of an American discussed how difficult it is to measure the nebulous concept of assimilation. One quantitative study attempted to compile civic, cultural, and economic indices by examining statistics such as marital status, number of children, citizenship, homeownership, etc. Looking at these variables only inform us of how the immigrant population compares to the native born in a narrowly defined way; it tells us absolutely nothing about how immigrants themselves view their assimilation into society. By attempting to be ‘objective’ we drown out the personal stories that are so important in understanding the holistic nature of the immigrant experience. In the name of research, we begin to look at people solely as numbers; we begin to disengage with humanity and lose the bigger context of how our research can contribute toward a better future.
In addition, many research studies attempt to identify causal relationships between two variables but rarely consider the impact of one variable on other outcomes. Many of these studies do not have the necessary funding to examine long-term trends or to consider the impact of a program or intervention on multiple facets of an individual’s experience. For instance, suppose that an intervention was found to have a positive effect on its desired outcome, and, as result, was scaled up. Now suppose in the long-term this same intervention had adverse health effects that were not measured both because of the short duration of the study, but also because they were not part of the researcher’s variable of interest. Imagine another such study that had adverse impacts on a concept as intangible as self-esteem, a concept that is incredibly difficult to quantify. While the intervention may be hailed as a success, by not considering these elements, we may be doing more harm than good.
Models, too, have their own set of problems. So much of research builds upon axioms, but rarely do we critically analyse this knowledge base. We must question the very foundations of our models and the assumptions that go into building them. A recent article by Ethan Watters in The Pacific Standard titled We Aren’t the World talks about Joe Henrich’s research on the problem of using models created in the West to generalise to populations in different cultures. When Henrich tried to apply game theory in diverse settings around the world, he found that the rules of the game were difficult to explain and results were markedly different from the same experiments conducted in the West. Indeed, the very foundations of the models, largely constructed in the developed world, failed to apply in the same manner worldwide. How many more of these kinds of models is research built upon and how problematic are they?
All of this is not to say that there is no value in numbers and models. Indeed, both are tools that can simplify an argument and leave the masses with a clear understanding of the main takeaways of a paper, in some ways making information more accessible. Numbers can help us with establishing clearly defined goals, and models can assist with framing ideas to generate greater debate and dialogue. But, in engaging with research, we must always be cognizant of the context within which we are operating and remember to never put ourselves before the grander goal, all in the name of ‘rigorous’ research.