(1993, 1996) found marital disaffection is not normally distributed. Most of the

respondents had low levels of disaffection. Twenty percent of the sample indicated an

"above average" or "high" level of disaffection (scoring 43 or above on the MDS), with

women expressing feelings of disaffection more frequently than men (Kayser, 1993).

 A third study designed to test the validity of the MDS was conducted with a

random sample of female members of the American Counseling Association (Flowers,

Robinson, & Carroll, 2000). The sample consisted of 323 women. The participants' ages

ranged from 26-89 years-old with an average age of 47.9, and 77.6% of the sample were

currently married; for those who were married, the average time married was 18.2 years.

This investigation of the MDS found a correlation of-.94 with a measure of respondents'

positive feelings toward their partner, a correlation of-.63 with marital status, a

correlation of .36 with a measure of the partner's problem drinking, and a correlation of

.48 with a measure of "workaholic behavior of the spouse" (p. 1102).

 The MDS has not yet been validated for use with non-married participants,

although the author has expressed her belief that its use is appropriate with close

relationships other than marriage (Kayser, 1996). It also has not been tested with

culturally and socioeconomically diverse samples. Despite these limitations, the MDS

appears useful for several reasons. Because items are balanced in reflecting feelings of

both disaffection and caring, the MDS may be less susceptible to participants trying to

respond in socially desirable ways compared to a scale measuring only satisfaction.

Additionally, many measures of satisfaction or strain assess couples' level of conflict;

however as Kayser (1996) notes, conflict is usually present to some degree in all

relationships and does not necessarily indicate the relationship is deteriorating. The MDS