Making A Sociological Argument
Instructor: Greta Krippner
September 28, 2000
“Of course the method of presentation must differ in form from that of inquiry.”
(Karl Marx, 1867, p. 102)
Once you have developed a viable research question, your next task is to review the evidence in order to formulate an answer to your question. The answer to your question is your thesis, or your argument. Typically, researchers do original research at this point—they analyze statistical data, go to the field, administer surveys, conduct experiments, etc. We don’t have time for that in the course of one semester, so we will use existing research (also called secondary research) as evidence. Even though we are not collecting our own data, the logic is the same—you will use data (collected by others) to support your position. This does not mean simply parroting another researcher’s results: the unique (and creative!) part of your research project comes in assembling evidence from a variety of sources.
So, for example, you may want to argue that birth order does not provide a good explanation of (conservative) social attitudes. You are taking the same position that Freese et al. do, but while you will report their findings, you will not limit yourself to their research. Rather you will look for other researchers who have considered the relationship between birth order theory and social attitudes. How do their findings compare with the findings of Freese et al.? If they are also arguing against birth order theory, they support your argument, and you will include their findings as additional evidence in support of your position. If they contradict Freese et al.’s position, you will also include them in your discussion, but here your task is to explain why Freese et al.’s findings are more persuasive.
Perhaps you want to take another tack not by arguing for or against birth order theory with respect to a specific outcome per se, but rather by comparing how birth order theory “performs” as compared to the standard sociological variables (age, race, gender, etc.) across a variety of social outcomes. Perhaps Freese et al. convinced you that birth order is not a good predictor of social attitudes, but does birth order do a better job predicting other social outcomes, including education, achievement, personality, etc.? In this case, you would still present the findings of Freese et al. as evidence about the effect of birth order on social attitudes, but then you would go on to examine research on birth order and education, achievement, and personality.
Keep in mind the difference between summarizing and making an argument here. You are not merely summarizing Freese et al.’s paper, you are using their findings to make your own argument. The distinction is tricky, because making an argument requires you to summarize the research of others, but for your own purposes.
Two Strategies for Making a Sociological Argument
What you do in your argument depends a great deal on how your question is framed. Generally, there are two different tasks you can take on in making a sociological argument:
- Establish a relationship between two or more phenomena (variables)
This is the mode of sociological thinking/argumentation we have stressed most in class. We have already discussed several questions that involve this kind of argument:
Example 1: Does birth order affect social attitudes?
Example 2: How does co-habitation prior to marriage affect the probability of marital success/stability?
Example 3: Is low voter turnout explained by the educational levels of the population?
Each of these questions asks about a presumed relationship: does a relationship exist between cohabitation and marital success? Between birth order and social attitudes? Between voting and educational levels? Presuming that the variables are measurable, these sort of questions lend themselves to quantitative analysis: most of the relevant evidence will be of a statistical variety. Where variables aren’t measurable, though, qualitative research may be used to establish a relationship.
Example 4: Do families with only girl (or only boy) children exhibit more closeness?
This question is again asking about a relationship between variables: does the quality of family interaction (i.e., “closeness”) differ in families with all-girl (or all-boy) children as compared to families where the children are mixed-gender? Note that “closeness” is a subjective characteristic, and not easily measured. Very likely, then, research on this topic will be qualitative.
Regardless of whether the research you are using is quantitative, qualitative, or a mixture of both, if your question is about establishing a relationship then your argument will generally involve adjudicating contradictory findings. You will find research that both supports and contradicts the existence of the relationship you are assessing. You must first decide, based on all the evidence you have reviewed, where you come down on the issue: are you persuaded that the posited relationship exists? You will then systematically make a case in support of your position, citing the relevant findings as evidence. You will also discuss findings that contradict your position, explaining why you find them less credible. Eliminating alternative explanations is an important component of making a convincing sociological argument. More on this in a moment. . . .
2. Establish a Mechanism
We haven’t talked about this a lot in class, but there is another type of research question in sociology. These are “how” and “why” questions—rather than attempting to establish (and quantify) a relationship between two variables, this kind of research question is oriented towards explaining how something works or why a particular phenomenon is occurring. These are questions about process. Often (but not always!) qualitative research is better suited to addressing process questions than quantitative research.
Example 5: What explains the recent influx of Latino immigrants to the United States?
Example 6: Why aren’t third parties successful in the United States?
Note that this kind of question can’t be expressed as easily or naturally in the language of independent and dependent variables. This difficulty reflects the fact that while this type of question does specify an “outcome” (dependent) variable (e.g., Latino immigration, third party success), independent variables (causes) are left open.
The task here is to provide a plausible explanation for an event. The relevant evidence may be more institutional or structural than statistical in nature. For example, in order to explain the influx of Latino immigration, relative levels of socio-economic development in the United States and Latin America might be relevant to your argument. Perhaps political events in Latin countries in recent years, or changes to U.S. immigration law are important. Here the task of constructing a sociological argument consists of weighing these factors in order to determine which are most important. As before, you will want to consider and eliminate alternative explanations. If you believe, for example, that the most fundamental reason for third party failure in the United States is the structure of campaign finance laws, then you may want to argue against an alternative (contradicting) explanation for that failure, such as the position that the existing two-party system effectively meets the needs of a wide variety of Americans.
Finally, note that some arguments accomplish both of these tasks: they establish a relationship and posit a mechanism. For example, research on the cohabitation question could first establish that there is a relationship between cohabitating prior to marriage and marital success and then try to explain how that relationship works. Does cohabitating allow couples a “trial” period in which to determine if they are truly compatible prior to marriage? Does it enable couples to negotiate difficult issues before committing to a permanent relationship? Does cohabiting provide couples an opportunity to practice interpersonal skills that, once acquired, strengthen the marital relationship? Establishing a relationship and explaining how the relationship works will often involve combining quantitative and qualitative research.
Making your Argument Convincing
Your goal is to convince a skeptical reader of the correctness of your claim. Some things to keep in mind:
- Making a sociological argument involves selecting and prioritizing key factors or causes from a multitude of possible factors or causes. A paper in which you argue that everything under the sun is related to your problem is not particularly useful or informative. Instead, your task is to simplify a complex reality by telling the reader which factors or causes are most important for a given phenomenon you are trying to explain. It is not your task to be exhaustive; it is your task to convince readers as to what is most central. So, for example, “Residential Segregation is a key cause of urban poverty,” is a stronger, more interesting claim than, “Social, political, and economic factors contribute to urban poverty.” In general, strong (specific) claims are preferable to weak (non-specific) ones.
- However, if your claim is too strong for you to defend with believable evidence, you are better off backing down to a thesis you can squarely defend with the available evidence.
- Use the facts, figures, statistics, interview data, etc. of other researchers to support your points. Don’t just recite the claims that others make based on their data, show the evidence behind their claims.
- Depending on your question, you may want to introduce and refute counter-arguments or alternative explanations. This strengthens your claims, because instead of allowing the reader to come up with counter-arguments, you are saying, “you might be thinking my thesis isn’t true because of x, well let me tell you why it’s true despite x.” By eliminating alternative explanations, you are heading off your critics at the pass.
- The quote from Marx is intended to remind you that while the process of working out your argument is (necessarily) messy, the presentation of your argument in your paper shouldn’t be. In other words, avoid writing your paper as a blow-by-blow of your thought process while you were working out your argument. Rather, in writing, you begin where you ended in thought—with a clean, concise statement of your argument. You then use your argument to guide and structure the paper. We will deal more specifically with organizational issues in sociological writing in a few weeks.