Indicators of Child, Family, and Community Connections, Companion Volume of Related Papers. V. What Do We Need to Know?


In light of the extremely limited knowledge we possess about the types of relationships that lead to, or substitute for, marriage, the most pressing need at the moment is to document the variety and nature of such relationships. Having done that, we need to understand the ways various types of relationships develop.

In any attempt to measure and analyze union formation, specific attention must be given to ethnic and cultural differences as these increasingly define our population. Gender differences are of obvious importance. We must also be attentive to pathways to union formation among those in mid or later life. Most divorced people remarry. Many widowed individuals remarry.(38) Each group may have its own patterns of mate selection, and its own pattern of relationships.

In the following section, I propose the information that would be needed and a strategy for obtaining it. Three topical areas of information could be gathered through national data collection efforts, including longitudinal surveys (e.g., NLSY97, SIPP.) First, is information to learn what types of relationships exist and what they are called. Second, is information to characterize those relationships with respect to traits associated with traditional marriage, such as sexual fidelity and the sharing of resources. Third is a sequence of questions measuring the degree of commitment to the relationship. These items follow the logic of Add Health, though they would be suitable for individuals of any age.

What types of relationships exist in America?

For classification purposes, we need to know the names used for contemporary relationships. This will require the use of an open-ended question. This is especially important in light of the ethnic and racial differences in America. It is also of growing interest to those concerned with gay and lesbian partnerships. The following information gathering efforts could identify the types of relationships that currently exist.(39)

  • Asking whether the person is currently involved in a romantic relationship with another person, a man or woman thought of as a steady, a lover, a partner or a relationship by some other name (YES/NO).
  • If yes, following with an open-ended inquiry into the nature of the relationship--what does the person call this relationship and how do they describe it.
  • If no, asking what type of relationship would the person be interested in having at this point in his or her life.

What are the characteristics of these types of relationships?

Once we have identified the different types of relationships and the various ways they are referenced, we turn to gathering information designed to characterize those relationships. I suggest that we do this by focusing on those relationship traits associated with a traditional monogamous marriage. I use traditional monogamous marriage as the reference because it is the most common, and most conventionally regarded end of courtship. As such, these characteristics serve as a useful benchmark, especially when considering how other types of relationships may foster or hinder progression to marriage. It is important to realize that this strategy would permit a researcher to understand any type of intimate relationship, heterosexual or homosexual, marital or otherwise. The choice of marriage as a comparison reference is not meant to imply that all relationships should be compared with marriage. Rather, it is simply an analytic strategy.

While there is no official rulebook for what marriage implies, there are clear consistencies in domestic relations laws, western religious customs, and social convention. In the United States, marriage is a free choice, based upon love. It involves co-residence and sexual fidelity. It is the traditional venue for childbearing. It involves some mixing and sharing of finances. It involves some division of labor in regards the management of tasks. And it involves the adoption of a conspicuous marital identity.(40) The sequence of questions that follows would assess the degree to which each type of relationship resembles this model of marriage.

Below are examples of the types of information that could be gathered on a national panel study. The information could be used to characterize all forms of relationships, marital and non-marital, heterosexual as well as homosexual, in sufficient detail to understand their broad outlines as relationships formed, dissolved, and resolved into new types. Combined with standard demographic information (on most surveys) about the respondent's age, ethnic/racial identity, marital history, fertility history, and sex, it would be possible to locate each relationship type in the life course, and characterize it by its similarity to a traditional model of marriage. For each person having a romantic relationship, one could enquire:

  • Whether person or partner has ever been married to someone else
  • Age, gender and race or ethnic identify of person and partner
  • Whether person's family and partner's family encouraged the relationship
  • Whether person and partner live together
  • Whether person and partner have sexual relations with one another
    If so, whether
    • person and partner have promised to be sexually faithful
    • person and partner are sexually faithful
    • whether person and partner had child together
    • if no then whether person and partner hope some day to have a child together
  • Whether person or partner spends some of the other's money
  • Whether person or partner does some of the other's chores every week
  • Whether person or partner earns enough to live comfortably without the other's income.
  • Whether person or partner have told friends or family about the relationship

Whether person or partner hopes to be together for life

How much commitment exists in each type of relationship?

Here the goal is to understand how contemporary relationships differ in the level of commitment involved. More generally, the concern is to understand how different types of relationships may foster or hinder the type of commitment found in stable marriages or enduring relationships.

In this paper, I focus on the 'exit costs' or the losses anticipated if the relationship were to end as one way of measuring commitment in contemporary relationships. The measures required to capture commitment were developed for the National Survey of Families and Households.(41) The sequence requires the respondent to imagine the consequences of ending his or her relationship for a number of areas of life. The individual who sees absolutely no negative consequence of ending her relationship may be thought to have no commitment to it. Alternatively, that person is independent of the relationship. In either case, the relationship is not likely to withstand the customary challenges faced by couples in the course of a long marriage.

Following efforts to gather information on the nature of the relationship, outlined above, people could be asked about the various areas of their lives that might be different if they separated from their partner, even if they think separation is very unlikely, and whether these areas are much worse, worse, same better, or much better. These areas might include, for example:(42)

  • standard of living
  • social life
  • job opportunities
  • overall happiness
  • sex life
  • being a parent
  • leisure time
  • friendships

View full report


"papers.pdf" (pdf, 639.68Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®