Strategies for Estimating Causal Effects
In broad terms, there are three strategies for estimating causal effects that involve (1) conditioning on confounders, (2) identifying instrumental variables, or (3) specifying mechanisms. The first, and by far the most commonly used, involves conditioning on all confounders in a statistical model. For instance, it is likely that mother’s education predicts both whether a mother smokes and the birthweight of her child. In this case, if we do not adjust for mother’s education in our analysis of smoking and birthweight, then we would likely overestimate the causal effect of smoking on birthweight, as part of the effect is likely due to mother’s education as a common cause of both. As noted above, a failure to incorporate all confounders in an analysis leads to biased estimates of causal effects. Well-articulated causal graphs allow researchers to identify the confounders that need to be included in an analysis. In practice, we never have measures of all possible confounders available, and therefore we need to rely on sensitivity analyses and/or more modest statements about our findings with respect to estimating causal effects.
The second strategy involves identifying an exogenous source of variation in the focal independent variable and often using statistical models based on instrumental variables. An exogenous source of variation refers to some feature of the world that induces a change in the focal independent variable for at least some cases but is otherwise unrelated to the outcome. Quasi-experiments fit into this category. Natural disasters can certainly cause changes in people’s lives that may be otherwise unrelated to an outcome or policy-oriented changes can fulfill the same role. To give an example, some studies in medical sociology have leveraged changes in compulsory schooling laws to try to estimate the effect of education on health (Courtin et al. 2019). A source of exogenous variation can be treated as an instrumental variable, i.e. a variable that has an effect on the focal independent variable but not the outcome.
The third strategy, and least commonly used, involves (1) specifying all of the mechanisms that link a focal independent variable to an outcome, (2) estimating the effects of each mechanism, and then (3) adding up the effects to get a total causal effect. A mechanism refers to the process through which a variable has an effect on an outcome. If we return to the education and health example, there are a number of mechanisms thought to account for the positive relationship. To name a few, higher levels of education lead to better and higher paying jobs, which can be used to support health. Higher levels of education also come with greater knowledge of health promoting behaviors and how to navigate the health care system. In addition, higher levels of education tend to expose people to similarly more highly educated peers who might in turn reinforce health promoting behaviors. Each of these processes can be thought as one of the mechanisms linking education to health. As this example might suggest, however, such an approach is quite challenging. We rarely know all of the mechanisms linking a focal independent variable to an outcome, and even in cases where we might have a good grasp on all of the mechanisms, we often do not have measures for each.
With respect to causal research questions, the counterfactual framework provides a powerful approach to understanding the possibilities available for causal analysis with observational data and the assumptions needed to support a causal interpretation. The regression models described for descriptive outcomes are routinely used for causal analysis as well. The details of the research design (e.g. Are key confounders measured? Are there exogenous sources of variation?) dictate whether a causal interpretation is warranted.
Qualitative Methods
Although qualitative research strategies are diverse, qualitative scholarship is united by the use of non-numerical data. Qualitative data comes from observations of the world, interviews with people individually or in groups, examination of documents, or in-depth analysis of any other materials that help reveal the social world. Returning to our example of education and health, qualitative approaches can enrich our understanding of why education is associated with better health. Ethnographic studies may observe how patients with differing levels of education use their knowledge in interactions with health care professionals (Luftey and Freese 2005). Likewise, interviews with health care professionals could add to knowledge by revealing how doctors think about patients from differing educational levels (Thompson et al. 2015). Finally, someone interested in this question from a qualitative perspective may decide to look at training manuals for health care providers over several decades to understand the messages conveyed about patients with varying levels of education. In each of these, the focus is on linking the general finding about human capital and health to the larger social and medical context.
Ethnographic methods are primarily about observing and participating in the social world, which allows the researcher to bridge scholarly and folk understandings of the world. For example, classic medical sociologists observed health care organizations and medical training programs to understand how doctors are socialized to provide care in particular ways (Becker et al. 2002[1976]; Charmaz and Olesen 1997). There are many different approaches to ethnography that span from complete observer to complete participant. A complete observer of health care organizations may sit quietly in the back of a room, avoiding direct interaction with those in the setting. In contrast, a complete participant may be trained as a doctor or nurse and is actively involved in the interactions taking place. Most medical sociology ethnographies fall somewhere between these two and include deep levels of observation and some interactions. Ethnographic methods are especially useful for understanding the organizational context of healthcare, including how medical professionals work together and with patients (Cain 2019; Jenkins 2018). Ethnographers measure concepts through observing behaviors, the context surrounding the behaviors, and how actors talk about their behaviors.
Interviews are commonly used in medical sociology. Interviews range from unstructured, where the researcher may enter the conversation with a topic and a reason they selected a particular person, to structured, which includes a predetermined set of questions, probes to follow up, and the order in which the questions are asked. The level of structure to the conversation depends on many factors, including how much we already know about the phenomenon and how researchers want to speak to theory. Researchers using a “grounded theory” approach often use unstructured interviews, while those using other approaches may want more structure (Charmaz 2014; Strauss and Corbin 1997). Interviews with individuals are useful for gaining an in-depth understanding of their sense-making about the world (Barry et al. 2001). Interviews can also take place with a group of participants, sometimes called a focus group interview (Krueger 2014). Guiding the conversation when working with a group can be challenging and the researcher must keep in mind that small group dynamics (i.e. talking over one another, dominating the conversation, group think) can make analysis difficult. That said, group interviews can be an efficient way to learn about actors’ perceptions and attitudes and are often used in qualitative studies of health and medicine. One advantage of group interviews is that participants engage with one another to agree, disagree, build on, contradict, or enrich others’ perspectives. These group engagements are useful for understanding how people respond to changes in medical practices (Cain and McCleskey 2019). Interview researchers measure concepts through analyzing participants’ verbal responses to questions, often grouping similar types of responses into codes, and then linking codes to one another through the development of themes.
Medical sociologists can also learn about the world through qualitatively analyzing content produced for another reason. Examples include newspaper articles about health, laws that govern health policies, billboards for public health campaigns, patient educational materials, television shows, or any other cultural product that one may use to understand the world. Content analysis is technique where the researcher establishes a selection process for finding