When Louis Jordan asked this grammar-busting question in his popular song from 1944, he knew something that relationship scientists would not discover for another 65 years: relationships are either pretty good, or they are not. Being in an unhappy relationship does not mean you are just a bit less happy than your contented counterparts; being unhappy in a relationship seems to be a unique and rather distinct experience. Unhappy in your relationship? Either you is, or you ain’t. And for 3 in every 10 of us, the answer is: is.
Clinical psychologists Mark Whisman, Douglas Snyder, and Steven Beach recently reported a study in which they administered the revised version of Snyder’s Marital Satisfaction Inventory, or the MSI-R, to both spouses in more than 1000 married couples across the United States. Husbands averaged 40.7 years of age and 14 years of education. Wives were about two years younger (38.8 years of age) and had 13.8 years of education. Couples had been married an average of 14.8 years, 78% had children, and 76.3% of the respondents were white. The MSI-R itself consists of 150 true-false items which poke and prod spouses to report on virtually all aspects of their relationship, and it is what we call in the business a “psychometrically sound” instrument for research. The MSI-R is also among the very best questionnaires that practicing psychologists have for diagnosing the sources and varieties of relationship unhappiness.
For a while now, psychological scientists have been interested in using sophisticated statistical tools -- taxometric methods, as they are called -- that help determine whether some phenomenon is more like a bell-shaped distribution (with people differing only by degree) or whether the underlying distribution is lumpy, indicating that people differ in kind. Application of these tools has revealed, for example, that depression is more like a bell-shaped curve, whereas eating disorders are distributed into either/or categories; for the most part, either you have bulimia or anorexia, or you do not.
So which is it for relationship happiness? Whisman, Snyder, and Beach are among the first to apply these tools to spouses’ reports of their marital satisfaction. They have discovered that judgments of relationship satisfaction are “taxonic,” or best captured by a qualitative (either/or) distinction rather than a quantitative distinction. People within each of these groups are far from being identical – within the happy group there are still degrees of happiness, and within the unhappy group there are still degrees of unhappiness – but the new idea is that there is a real break between those couples who are happy and those who are not.
Using their nationally representative sample, Whisman and colleagues estimate that 31% of all marriages qualify for the “unhappy” classification. And by re-administering the same questions 6 weeks later, they were able to show that the happy/unhappy classification is pretty stable. This is arguably the best estimate available to date on how many of our friends and neighbors who have been married 12-18 years or so are truly struggling with their marriages – and personally I find this to be a staggering number.
So what kinds of questions on the MSI-R do the discriminating work? Turns out only 10 questions are needed. The chances of a person belonging to the unhappy group go up when he or she endorses some items by answering yes or true (“Does your partner often fail to understand your point of view on things?”) and when he or she answers no or false on other items (“Just when you need it most, does your partner make you feel important?”). Endorsing four or more of these items in a way that indicates discontent with these 10 key relationship domains is sufficient evidence that an individual has flipped from being in a relatively satisfying relationship to being in a dissatisfying relationship. Copyright issues prevent me from listing the entire set of items, but suffice it to say that none of them would surprise you: sex, disagreements, communication, and spending time together all matter in just the ways you would expect. The items are important, of course, but the real magic here comes from the careful application of the taxometric methods.
This study is limited to couples who had been married quite some time – nearly 15 years on average – and there is good reason to believe that relationships fluctuate some and then stabilize before that point. The peak period for divorce, for example, happens within the first 5 years of marriage. We might well expect less stability and perhaps even no evidence for qualitative differences among couples married for 5 years or less; the taxon might need time to emerge. Prior to that point, quantitative differences between couples – differences in degree – might win the day.
What Louis Jordan did not (to my knowledge) anticipate in 1944 was the development of effective therapeutic interventions for unhappy couples. Those 30% or so couples who are struggling should not give up hope, because careful experiments of specific, well-designed couple therapies yield pretty convincing evidence that many unhappy couples can join the ranks of the contented – and that they can sustain these changes over reasonably long spans of time.
The Take-Away Message
There appears to be a threshold in relationships which, once crossed, leaves spouses thinking and feeling about their relationship in ways that are quite different from the thinking and feeling of happy spouses. I would like to believe that this is a biologically-embedded call to action that arises when our crucial social relationships go awry – a constant reminder that all is not right in our emotional lives and that something must be done. Ignoring this call is very unlikely to make it go away, but heeding it just might put couples on the better side of this great divide.
Snyder, D.K. (1997). Manual for the Marital Satisfaction Inventory – Revised. Los Angeles: Western Psychological Services.
Whisman, M.A., Beach, S.R.H., & Snyder, D.K. (2008). Is marital discord taxonic and can taxonic status be assessed reliably? Results from a national, representative sample of married couples. Journal of Consulting and Clinical Psychology, 76, 745-755.
Whisman, M.A., Snyder, D.K., & Beach, D.K. (2009). Screening for marital and relationship discord. Journal of Family Psychology, 23, 247-254.
I am grateful to Dr. Douglas Snyder (creator of the MSI-R) and to Western Psychological Services (publishers of the MSI-R and the brief version of it, the MSI-B) for their generosity in giving me permission to use the two items presented here.
Sample items from the ‘MSI-Brief (MSI-B)’ research form copyright © 1997 by Western Psychological Services. Reprinted for reference by T. Bradbury, UCLA Department of Psychology, by permission of the publisher. Not to be reprinted for any additional purpose without the prior written authorization of WPS, 12031 Wilshire Boulevard, Los Angeles, CA 90025 USA (firstname.lastname@example.org). All rights reserved.