Is it an interesting sociological fact that "urban people are better educated than rural people"?

Let’s suppose that this statement about urban-rural differences in educational levels summarizes census data by calculating the average number of years of formal schooling for people in cities and people in rural areas. Is this brute fact about two large populations a valid description of the two populations? Is it an important or illuminating sociological fact?

There are several types of questions we need to ask about this fact. First, there are questions to address about the statistical features of the data themselves. How are these two populations distributed around the mean? How wide is the variance around the mean? Do we get a different result if we measure the median number of years of schooling as opposed to the mean? Which of these measures is a more meaningful description of the population as a whole — median or mean?

Consider a hypothetical set of results. If the rural population is quite homogeneous around a mean of 10 years of schooling, while the urban population is widely distributed in a range from 5 years to 20 years, the fact that the urban mean is 11 years is somewhat misleading; the majority of urban people in this hypothetical case have less than 9 years (so urban is less well educated), while at the same time 20 percent of urban people have at least 14 years (so urban population is better educated). The point is this: the brute fact of a difference in the means is not particularly insightful in estimating the educational resources of the two populations.

Second, there are important sociological questions about the internal differentiation of the population into groups with very different educational patterns. Do women and men show different profiles in the two large populations? How about members of ethnic or racial groups? How about groups identified by income, wealth, or home ownership? What about groups defined by whether a parent had attended college? Arriving at this set of questions requires sociological imagination. The investigator needs to consider what internal differentiations within the large population might affect the sub-group’s educational characteristics.

Finally there is the question of finding possible causal explanations of the differences that are discovered across major populations (urban and rural) and within sub-populations (ethnic, gender, or class-defined groups), and tracing out some of the ways in which these patterns in turn cause other social outcomes (future inequalities of income, for example). Those causal mechanisms might be various: differences in the opportunities that are presented to members of different social groups (including the possibility of discrimination), differences in values and cultures, within families, differences in gender treatment, differences in religious traditions and practices, and differences in access to resources, to name a few.

Now suppose we have done quite a bit of empirical, theoretical, and causal analysis along these lines. Suppose we have found that the internal structures of eduational attainment statistics are quite different between urban and rural populations; that the taxonomy of sub-groups is different; and that the causes and social mechanisms of the differences in attainment across the rural-urban divide are substantially different as well. What salience does the original brute fact continue to have (that the means are different in the two populations)?

We might say that the brute fact is in fact a valid empirical observation; that it needs to be substantially further analyzed; and that the genuinely valuable and insightful sociological findings only emerge once we have further disaggregated the statistics across salient groups and have provided some hypotheses about the mechanisms that influence the educational attainment profiles of the various sub-groups. At that point we have some idea of the underlying sociology that produces the brute fact. But the brute fact itself is largely unilluminating.

Social knowledge: measurement of properties in diverse groups

When we gain knowledge about silver, DNA, or cholera, we can study virtually any samples of the item and arrive at a description of its properties and causal powers, and this description will correspond accurately to other instances as well. We learn about the type by learning about the individuals, and we don’t have to worry about substantial differences among individuals in the type. Cholera is cholera, whether it occurs in Mexico City or Bangalore. So knowledge we acquire about a few instances can be generalized to other instances. This feature of “type-uniformity” is found in many of the types of entities studied in the natural sciences.

There are exceptions in the natural sciences; there are classes of phenomena that embody substantial variance among individuals in the class. Hurricanes and volcanoes are examples of “type-heterogeneous” concepts in the natural sciences. In these examples, the phenomena are grouped together in terms of a set of crude observable characteristics; it is then a question for research to determine whether there are common structures and causal backgrounds that constitute one or more sub-groups of items within the classification. But more typical types of entities in the natural sciences fall into “natural kinds” with common structural and causal properties.

Consider now what is involved in arriving at knowledge about a complex social reality — the city of Chicago, for example. The social reality of Chicago is constituted by the social behaviors of the individuals who live in Chicago and the institutions that these individuals populate. Consider some of the topics concerning which we might want to gather knowledge:

  • What is the health status of Chicagoans?
  • What is the climate for race relations in Chicago?
  • What is the standard of living in Chicago?
  • What is the rate of economic growth in Chicago?
  • How do people in Chicago feel about higher education?
  • What is the climate for new-business startup in Chicago?
  • How well do the institutions of the mayor’s office and the city council work?
  • How are the Chicago public schools performing?

Notice that almost all these questions invite us to consider the heterogeneity of the population and its organizations. There is an average level of heart disease in the city. But the average is a poor indicator of any particular person’s health, because there are socially significant differences across groups with respect to almost all these questions. So a study of health in Chicago requires that we consider some of the ways in which different groups are affected by a variety of circumstances in ways that systematically affect their health status.

These facts suggest that a statement about a social characteristic of a large population needs to be nuanced, indicating the degree of variation of the characteristic across individuals and groups within the population as well as the most significant sub-populations showing the greatest variance from the group mean behavior. Race, ethnicity, gender, geography, age, income, employment, labor-union membership, and education might be variables that define groups with significantly different measures of the variable of interest. And once we have determined that certain social characteristics (race, income, and union membership, let us say) are associated with the outcome of interest (health status, say), then we are stimulated to ask the causal question: what are the social mechanisms at work that produce the associations that are discovered?

Generalizations about cluster items

What is the basis of classification of items into groups? And how do classifications fit into scientific inquiry and theory? First, what different types of classification are there?

Essential: The items may share a common defining characteristic (e.g. “liquid”, “metal”)
— Etiological: The items may share a common cause (e.g. “viral illness”)
— Structural: The items may share a common underlying structure (e.g. “protein”, “elm”)
— Functional: The items may share a common function (e.g. “weapon”, “school”)

Non-essential: The items may lack necessary and sufficient conditions but share some overlapping characteristics.
— Symptomatic: The items may share a set of observable characteristics or symtoms (e.g. “pneumonia”, “schizophrenia”)
— Cluster: the items may share some among a list of characteristics (e.g. “game”, “leader”)

(It is a bit curious to observe that this amounts to a classification of systems of classification.)

Now let’s consider some social science terms and see where they fall in this scheme: riot, civil war, bicameral legislature, ethnic group, democracy, charismatic leader, financial city, working class organization.

Several of these concepts are symtom or cluster concepts: riot, democracy. Several others are etiological or structural: civil war, bicameral legislature, financial city. One combines structural with functional criteria: bicameral legislature.

Now suppose we have identified a set of things as being “democracies”. They share some among a set of features that are associated with democracy, and there is no set of features shared by all instances. What kinds of social science inquiry can we do? First, we can do comparisons within the group of democracies we have identified; we can look for similarities and differences across the group. And second, we can ask whether membership in this group is associated with membership in some other group beyond what chance would predict. In other words, we can consider whether there are true statements like “all democracies are X” or “most democracies are Y.”

It is not the case that “all pneumonias respond to penicillin”–for the reason that there are two causal and structural kinds of pneumonia, and only one of these involves organisms treatable with penicillin. The causal heterogeneity of this group means that strong generalizations are difficult or impossible here.

The concepts of riot, revolution, and democracy are similarly heterogeneous, both causally and structurally. So we should expect only weak generalizations across this group and other social charactistics. On the other hand, the tools of social comparison are most valuable here. We can discover through additional comparative work within the category, whether there are similar structural and causal processes at work among instances of this concept.

%d bloggers like this: