The study, which appears in the February issue of the Journal of Marriage and Family, found that expertise in gauging how happy married couples are is greatest among couples who have been in long marriages.

The results surprised Rachel Ebling, a UC Berkeley psychology student, and UC Berkeley psychology professor Robert Levenson, who prepared the report.

"We thought marital researchers would do well," said Ebling. "But those with a vested personal interest in relationships did better. "

"We feel we have identified an important untapped resource for researchers and clinicians," added Ebling. "Clinicians-in-training may get some interesting and important insights by talking to couples who have been married for a long time."

The idea for this study came from, of all places, the wedding of one of Levenson's friends.

"At the table at the reception, there were all kinds of different people — coworkers, relatives of the couple and friends," Levenson said. "Everyone at the table had an opinion as to whether the marriage would last."

That prompted Levenson and Ebling to devise studies about two issues — how accurately people of varying backgrounds could predict whether a couple would stay married and how well people could evaluate whether a couple had a good marriage.

Ebling and Levenson studied nine different groups of people. Those in the professional groups were marital researchers, marital therapists, pastoral counselors and graduate students in clinical psychology. The nonprofessional groups included couples who were in long-term happy first marriages, couples in long-term unhappy first marriages, newly married individuals, newly divorced individuals and UC Berkeley undergraduates. The data were gathered from individuals in a very simple, straightforward way. People watched 10 three-minute videotaped segments (recorded six to 13 years earlier) of couples talking about a conflict specific particular to their relationship. The evaluators then filled out a questionnaire.

Some literature suggests that most people do not need much time — as little as 30 seconds — to make a social judgement, so Levenson and Ebling felt three minutes was enough.

On the issue of whether a couple would stay married, all groups were correct an average of 53.8 percent of the time, faring only slightly better than chance.

But when it came to the people who could best predict which couples had happy marriages, the nonprofessionals did better than the professionals. The average percent of correct ratings for the four groups of non-professionals was 61 percent, while the average percent of correct ratings for the four professional groups was only 52 percent.

Additionally, the group of participants in long-term satisfied marriages were the most accurate group of all, with a 64 percent accuracy rate.

"Those for whom marriage held high personal meaning, as rated by a panel of judges, were more accurate than those with professional training," the researchers wrote.

Other accurate judges were people who paid the most attention to expressions of disgust, criticism or contempt. "Being highly critical or feeling your spouse is falling below a standard is very indicative of an unhappy marriage," Levenson said. He added that the couples' expressions of anger did not necessarily indicate the happiness or unhappiness of the marriage.

The findings raised an important question: Why are certain kinds of professional training related to marriage associated with lower levels of accuracy in judging marital satisfaction, relative to certain kinds of personal experience?

"[A] possible explanation for the low levels of accuracy among marriage professionals in our study is that this results not from training but rather from the experience of working with primarily distressed couples," Levenson and Ebling wrote. "Another possibility is that marriage professionals have become accustomed to making judgements based on much more protracted observations of couples' behavior, and thus find it both difficult and unfamiliar to make these judgments based on the kinds of brief samples of marital behavior used in the present study."

The researchers also commented that the groups of marriage professionals likely included those for whom the personal significance of marriage varies.

So while Ebling and Levenson do not suggest couples abandon counseling, they do suggest that consulting with older friends and relatives may provide additional insights.