Journal of Marital and Family Therapy

1986, Vol. 12, No. 4, 403-413.

Predicting Marital Success With PREPARE:

A Predictive Validity Study

Blaine J. Fowers & David H. Olson

In order to determine the predictive validity of the premarital inventory PREPARE, this study assessed the utility of PREPARE in predicting marital success. A 3-year follow-up study was conducted with 164 couples who took PREPARE during their engagement. As hypothesized, it was found that satisfied couples scored significantly higher on the inventory than dissatisfied couples, divorced couples, and couples who cancelled their marriage. It was also hypothesized and found that dissatisfied married couples did not differ significantly from couples who cancelled their marriage or those who divorced. Using discriminant analysis, it was found that the PREPARE scores from 3 months before marriage could predict with 80-90% accuracy which couples were separated and divorced from those that were happily married. These findings not only demonstrate the predictive validity of PREPARE, but its potential utility in identifying high-risk couples who could benefit from more intensive premarital counseling.

According the current estimates (National Center for Health Statistics, 1984), 40% of all first marriages end in divorce. Divorce has become an accepted cure for ailing marriages (Levinger and Moles, 1976; Weiss, 1975). In spite of the high divorce rate, marriage still continues to be popular. As Berscheid and Campbell (1981) have noted: “Ironically, at the same time that close relationships have become substantially more vulnerable to disruption and dissolution than they were just a generation or two ago, close relationships are seen by most people as being the prime source of personal happiness.” Marriage continues to be the most popular voluntary institution in our society, with over 90% of the population eventually marrying at least once (Glick, 1984).

While the average length of marriage is only 6 to 7 years (Glick, 1984), many of these relationships can be assumed to have contained the seeds of eventual breakup from the very beginning. Also, some sort of intervention might have been helpful if the potential trouble spots could have been identified. Successful intervention, therefore, would require the development of valid and reliable premarital instruments to identify these couples at risk.

This paper builds upon the pioneering work of Burgess and Wallin (1953) and Terman (1938), who attempted to predict marital success. Many of the same content areas are explored, but more rigorous methodology was used to measure these variables. Much of the more recent work on mate selection has failed to identify factors which relate to marital success (Murstein, 1980).

While much work has been done conceptually and empirically in the fields of marital and family therapy (Olson, Russell, & Sprenkle, 1980; Gurman, 1983), little work has been done in terms of preventing marital problems or helping couples prepare for marriage (Olson, 1983). In spite of the importance of having valid and reliable diagnostic tools for marital and family therapy (Filsinger, 1983), there has been a serious lack of inventories available for assessing premarital couples.

This paper describes a premarital inventory that can be used for preventative work with couples before marriage. Specifically, it focuses on the predictive validity of the PREPARE instrument for premarital couples. This study will also provide more empirical data on early marital adjustment. There have been only a few longitudinal follow-ups of couples in the early years of marriage (Rolfe, 1975; Schaefer, 1979; and Springer, 1983). The empirical evidence , to date, is primarily descriptive and based on small samples. This study may provide the beginning of a more predictive analysis of relationship variables that seem important in the early dissolution of marriages.

Adequate predictive information is the cornerstone of prevention efforts. The data ought to be of longitudinal nature, based on a reasonably large sample size, and come from measures that have strong validity and reliability for engaged couples (Baggarozzi & Rauen, 1981; Gurman & Kniskern, 1978; and Rozeboom, 1966). The current project is one step in that direction.

MARRIAGE PREPARATION AND PREPARE

Preparation for marriage has been suggested as one form of divorce prevention (Rutledge, 1968; Olson, 1983). While there has been increasing theoretical and empirical interest in marital preparation programs, two recent reviews of the literature (Baggarozzi & Rauen, 1981; Schumm & Denton, 1979) have pointed out several flaws in these efforts, namely, a lack of information on the needs of engaged couples and a lack of theoretical underpinning for the work that is being done. A further criticism has been directed at the inadequate methodology of the assessments of premarital programs.

In any preventive approach, some assessment tool is needed to provide direction for the preventive efforts. First, factors predictive of unsuccessful marriages must be identified. Second, couples need to be assessed on those critical variables to assess their relationship strengths and work areas. Third, specific interventions need to be developed which will deal with couples’ problem areas. Unfortunately, premarital programs have attempted to bypass the first two steps and, therefore, most of the programs lack adequate theoretical and empirical grounding or clinical relevance for each couple. This study is an attempt to provide data for the first step by testing the predictive validity of the premarital inventory PREPARE (Olson, Fournier, & Druckman, 1986; Fournier, Olson, & Druckman, 1983). The development of PREPARE scales was based on the theoretical (Duvall, 1971; Rappaport, 1963; Rausch, Goodman, & Campbell, 1963) as well as empirical (Fournier, Springer, & Olson, 1979; Kitson & Sussman, 1977) indicators of the critical tasks related to early marital adjustment.

PREPARE is a 125-item inventory designed to identify relationship strengths and work areas in 11 relationship areas: (a) Realistic Expectations, (b) Personality Issues, (c) Communication, (d) Conflict Resolution, (e) Financial Management, (f) Leisure Activities, (g) Sexual Relationship, (h) Children and Marriage, (i) Family and Friends, (j) Equalitarian Roles, and (k) Religious Orientation, (Olson, Fournier & Druckman, 1986). Additionally, the instrument contains an Idealistic Distortion Scale.

For each scale, an Individual Score is provided for each spouse. An individual’s score on a category is revised, based on that person’s idealistic distortion score and the

correlation of that scale with idealistic distortion. In addition, a Positive Couple Agreement (PCA) score is provided for each category which measures the couple’s consensus on issues in that area.

A number of instruments were used to assess the concurrent validity of PREPARE (Fournier, 1979). Among them were: (a) Inventory of Premarital Conflict (Olson, Druckman, & Fournier, 1978), (b) Family Environment Scale (Moos & Moos, 1976), and (c) Marital Adjustment Scale (Locke-Wallace, 1959). There was a significant relationship between all the PREPARE scales and the Marital Adjustment Scale (p<.01 level). Directional predictions were made for 129 combinations of PREPARE subscales and criterion variables. Of these, 96 correlations were in the predicted direction with statistical significance (p<.01), 21 correlations were nonsignificant, and only 2 were significant and contrary to predictions. This indicates that PREPARE is measuring, in large part, what it purports to measure.

The reliability of PREPARE has been assessed for internal consistency (alpha) and test-retest on each scale. The internal consistency reliability (alpha) averaged .70 and test-retest reliability averaged .78 (Fournier, 1979; Olson et al., 1986). The alpha reliability for the scales were: (a) idealistic distortion (.88), (b) realistic expectations (.75), (c) personality issues (.74), (d) communication (.70), (e) conflict resolution (.72), (f) financial management (.67), (g) leisure activities (.61), (h) sexual relationship (.50), (i) children and marriage (.49), (j) family and friends (.70), (k) equalitarian roles (.77), and (l) religious orientation (.82).

In addition to the 12 assessment scales, PREPARE contains a variety of background information. This provides information on the following variables: (a) age, (b) education, (c) monthly income, (d) the number of months each person has known their partner, (e) the number of months prior to the marriage that the couple took the inventory, (f) the parents’ reaction to the marriage, (g) the friends’ reaction to the marriage, (h) parents’ marital status, (i) birth position, (j) number of siblings, and (k) population of the place of current residence and of residence during childhood.

This inventory was specifically designed to be used in the assessment of engaged couples. It was designed to identify relationship strengths and work areas for each couple. Its scales were also constructed in such a way as to promote couple dialogue and to help promote greater relationship enhancement (Olson, Fournier & Druckman, 1986). As such, it was designed primarily for use as a preparation tool.

If PREPARE could be shown to discriminate between couples who develop successful and unsuccessful relationships, then it could also be used in identifying high risk couples. These couples can then receive more intensive premarital counseling to help them build on their relationship strengths and deal with their work areas.

METHODS

Subjects

The subjects were 164 couples (328 individuals) who had been married 2-3 years and took PREPARE 3-4 months before marriage. The clergy that administered the inventory to the couple before marriage asked them to participate in this follow-up study. In addition to taking PREPARE before marriage, all these couples received on to two feedback session on their results.

This purposive sample was selected by clergy who had previously administered PREPARE to these engaged couples. The clergy were asked to identify 2-5 couples who were satisfied with their marriage and 2-5 couples who were divorced, separated or dissatisfied with their marriages.

These couples were divided into several subgroups by their marital status (married, separated, divorced or cancelled marriage), based on their responses to a marital satisfaction questionnaire. These subgroups were combined to form four groups: (a) married satisfied (consisting of 59 married couples), (b) married dissatisfied (consisting of 22 married couples), cancelled (consisting of 52 couples who cancelled or delayed their marriages), and (d) divorced or separated (31 couples).

The average age of the husbands and wives was 25.2 and 23.2 years, respectively, and the couples were married and average of 23 months. The majority of couples had attended some college, and their combined median income before marriage was $1200/month ($14,400, yearly). Couples were primarily Caucasian and of a Christian religion.

Instruments

While PREPARE was taken by couples before marriage, a couple questionnaire was used in the follow-up study. The couple questionnaire consisted of two sections, used to assess marital satisfaction and relevant background information of the married couples. The first section contained demographic questions regarding their age, sex, number of children, educational level, the number of months the couple had been married, income, the population of their current living area, their parents’ marital status, and whether the couple had any relationship counseling.

The second section contained the Idealistic Distortion and Marital Satisfaction scales from the marital inventory ENRICH (Olson, Fournier, & Druckman, 1986). The Idealistic Distortion scale is a 5-item version of the 15-item Idealistic Distortion scale in PREPARE (alpha reliability = .92). The Marital Satisfaction scale is a 10-item scale, with 1 item assessing each of the major content categories in ENRICH (alpha reliability = .81).

Procedure

PREPARE users who had administered the inventory eight or more times before January, 1983, and were currently using it, were sent the Clergy Questionnaire, along with a set of Couple Questionnaires. They were asked to select 2-5 satisfied couples, 2-5 dissatisfied couples, and any couples who cancelled or delayed their marriage to whom they had administered PREPARE 1 to 3 years ago. The dissatisfied group could include married dissatisfied, divorced, and separated couples.

The clergy’s decision as to which group these couples fit into was to be made on the basis of their knowledge of the couples’ current marital satisfaction, not on their PREPARE scores. They were also instructed not to inform the couples of their group assignment.

Only the married couples, both satisfied and dissatisfied, completed the Couple Questionnaires. It was not appropriate for couples who were separated, divorced, or who had cancelled their marriage plans to complete the marital satisfaction scale or other items in that questionnaire.

To maintain confidentiality, the clergy forwarded a Couple Questionnaire to the married couples. It was accompanied by a letter explaining that the study was a follow-up on couples who had taken PREPARE to obtain data on early marital adjustment. The couples were instructed to complete the questionnaires separately. Questionnaires were given to 208 couples, and 103 couples completed the survey, a return rate of 49%.

The clergy’s assessment of the couples’ marital satisfaction was originally thought to be a useful criteria for assigning couples to the satisfied and dissatisfied groups. However, this was not possible, since the couple’s own assessment differed considerably from the clergy. It was decided to rely on the couples’ assessment of their relationship, based on the marital satisfaction score.

The median marital satisfaction score was 41. It was chosen as a cutoff score to separate satisfied from dissatisfied couples for the analyses. Since there is a marital satisfaction score for each partner, only those couples in which both partners’ scores fell at 41 or above, were included in the couple analyses of the satisfied married group. Also, only those couples in which both scores fell below 41 were included in the dissatisfied married subgroup.

RESULTS

The mean and standard deviations of couples’ positive agreement (CPA) on each of PREPARE’s 11 subscales and average CPA are presented in Table 1. These data are provided for each of the four groups to be compared: (a) married-satisfied couples, (b) married-dissatisfied couples, (c) couples who cancelled or delayed their marriage, and (d) couples who were separated or divorced.

Table 1 also presents the results of an analysis of variance to test overall differences between the four groups. As the results indicate, significant difference between groups were found in 8 of the 11 subscales. Specifically, the results indicate that there were differences in the PREPARE positive couple agreement score (percentage) between couples who were later classified as happily married, unhappily married, cancelled marriage and separated or divorced. These differences were found in: realistic expectations, personality issues, communication, conflict resolution, leisure activity, sexuality, family and friends, religion, and in the overall average couple positive agreement. On the other hand, the results showed that no differences between groups existed in the areas of financial management, children and marriage, and equalitarian roles.

Table 1

Mean Positive Couple Agreement (PCA) Scores for Four Groups

Positive Couple Agreement (PCA)

Group Means and Standard Deviations

Analysis of Variance and Linear Trend Between Groups

PREPARE

Scales

Married Satisfied

n = 59

Married

Dissatisfied

n = 22

Cancelled/

Delayed

n = 52

Separated/

Divorced

n = 31

Between

Groups

Linear

Term

Realistic

Expectations

48.0

(23.3)!

38.2

(19.2)

35.8

(21.7)

31.9

(23.2)

4.55**

10.15**

Personality

Issues

43.4

(23.1)

34.1

(26.1)

32.5

(20.4)

28.4

(26.5)

3.50*

7.83**

Communication

58.8

(22.1)

45.9

(24.8)

47.7

(22.6)

40.7

(27.3)

4.71**

9.75**

Conflict

Resolution

58.5

(20.7)

40.9

(30.4)

44.6

(25.6)

37.1

(26.9)

6.35**

11.57**

Financial

Management

42.9

(25.9)

35.9

(22.8)

36.5

(25.4)

33.2

(26.8)

1.19

2.43

Leisure

Activity

63.2

(22.1)

50.0

(25.4)

55.6

(21.0)

49.3

(26.8)

3.24*

4.76*

Sexuality

59.2

(18.4)

45.0

(20.9)

50.9

(21.0)

44.5

(19.6)

4.99**

7.23**

Children and

Marriage

49.7

(21.1)

47.7

(20.2)

47.1

(19.6)

45.2

(17.9)

0.37

0.99

Family and Friends

57.0

(23.0)

48.2

(30.5)

44.8

(23.6)

40.0

(30.4)

3.58*

8.70**

Equalitarian

Roles

59.5

(19.9)

53.6

(19.9)

51.9

(22.8)

51.0

(24.0)

1.57

3.13

Religion

48.1

(30.8)

37.3

(25.7)

39.6

(27.9)

25.8

(26.31

4.26**

10.21**

Average

PCA

53.5

(13.3)

43.4

(16.7)

44.3

(12.9)

38.8

(18.3)

7.95**

16.82**

! Standard deviations are listed in parentheses.

* p < .05 ** p < .01

FIGURE 1: Happily Married vs. Divorced/Separated:

Positive Couple Agreement (PCA) Scores on PREPARE Categories

The results of the linear trend analysis (Table 1) indicated, as predicted, that there was a descending linear trend in positive couple agreement in the order specified, (i.e., married satisfied, married dissatisfied, cancelled/delayed, separated/divorced). This linear trend was highly significant in all of the 8 premarital areas (subscales) for which differences between groups were found.

The second series of analyses was conducted to further test differences between pairs of groups. As was noted earlier, it was hypothesized that differences in PREPARE positive couple agreement would be found between couples who are happily married and those who are dissatisfactorily married, cancelled their marriage, and separated or divorced. No differences were expected among the latter three groups.

Of particular interest are the differences between satisfactorily-married couples and couples who separated or divorced. As the results in Table 2 indicate, significant difference between these two groups in premarital couple agreement existed in 10 of the 11 PREPARE categories and in the overall average couple positive agreement score. The difference between these two groups in all PREPARE subscales are graphically presented in Figure 1. Significant differences were found between married satisfied and married dissatisfied in 6 of the 11 categories and in the average CPA. Married-satisfied couples

Table 2

t-Test Differences Between Groups

Married Satisfied

vs.

Married Dissatisfied

vs.

Cancelled/

Delayed

vs.

PREPARE

Scales

Separated/

Divorced

Married

Dissatisfied

Cancelled/

Delayed

Separated/

Divorced

Cancelled/

Delayed

Separated/

Divorced

Realistic

Expectations

3.11**

1.76

2.84**

1.04

.45

.76

Personality

Issues

2.78**

1.56

2.62**

.78

.28

.79

Communication

3.41**

2.26*

2.61**

.72

-.30

1.27

Conflict Resolution

4.19**

2.50*!

3.15**

.48

-.54

1.27

Leisure

Activity

2.63**

2.30*

1.86

.09

-.98

1.18

Financial

Management

2.86**

1.76*

1.15

.43

.29

1.06

Sexuality

3.50**

2.97**

2.19*

.09

-1.12

1.39

Children and

Marriage

1.62

1.14

1.04

.68

.38

1.12

Family and

Friends

2.96**

1.39

2.74**

.96

.51

.80

Equalitarian

Roles

2.64**

1.84*

2.56**

.74

.38

.96

Religion

3.43**

1.47

1.52

1.58

-.34

2.23*

Average PCA

3.95**!

2.83*

3.67*

.92

-.26

1.46!

!t-tests based on separate estimate. All other tests based on pooled variance estimates.

* p < .05 ** p < .01

had significantly higher couple agreement scores in the areas of communication, conflict resolution, leisure activity, financial management, sexuality, equalitarian roles, and overall average CPA. No differences were found in the areas of realistic expectation, personality issues, children and marriage, family and friends, and religion.

Comparing married-satisfied couples and couples who decided to cancel or delay their marriage, differences in couple positive agreement were found in 7 of the 11 categories and in the average CPA. The left portion of Table 2 shows the results of the t-tests for differences between married-satisfied couples and couples who are married dissatisfied, cancelled/delayed, and separated/divorced.

As hypothesized, no differences were found in premarital couple-agreement between married-dissatisfied, cancelled, or separated/divorced in any of the PREPARE subscales. Also, only one difference was found between the cancelled and separated/divorced groups, and that was in the religion category. The right-side portion of Table 2 contains the results of the t-tests for differences between (a) married-dissatisfied couples and those who cancelled their marriage, (b) marriage-dissatisfied couples and those who were separated or divorced, and (c) couples who cancelled their marriage and those who ended in separation or divorce.

The final series of analyses was geared toward testing the capability of PREPARE scores to differentiate between satisfactorily married couples and those who were dissatisfied with their marriage, who cancelled their marriage and those who were later separated or divorced.

Table 3 presents the results of discriminant analyses between the pairs of groups conducted to achieve this goal. The results indicate the percent of couples correctly classified into each group and total percent of correct discrimination when: (a) individual (males and females) PREPARE scores are used for classification; (b) couple (positive agreement) scores are used for classification; and (c) both individual and couple scores are used.

The results give ample support to the predictive power of PREPARE scores. Premarital individual scores of all the PREPARE’s subscales, for instance, can be used to correctly predict happily married from those who will end in divorce in 81% of the cases, happily from unhappily married in 79%, and happily married from those who cancel marriage in 78% of the cases.

When both individual and couple positive agreement scores are used for predicting successful marriages, they can correctly discriminate satisfied from dissatisfied marriages, and satisfied marriages from those who cancel, in over 80% of the cases. Furthermore, these combined scores can correctly discriminate successful marriages from those who end in divorce, in 91% of the cases.

Table 3:

Discriminant Analysis Between Pairs of Groups: Percent Correctly Classified

PREPARE

Scale

Happily Married vs.

Separated/Divorced

Happily Married vs.

Unhappily Married

Happily Married

vs. Cancelled

Total

Happy

Separated/ Divorced

Total

Happy

Unhappy

Total

Happy

Cancelled

Scores Used

Individual Scores

Couple Positive

Agreement

Both Individual

and Couple

PCA Scores

DISCUSSION

It was hypothesized that satisfied couples would have consistently higher scores on PREPARE couple agreement and individual scale scores than dissatisfied married and separated/divorced couples. The couple agreement measures showed that at the time of their engagement, satisfied couples had a great deal more relationship consensus that did dissatisfied couples. This lends considerable support to the notion that marriages that are distressed within the first 3 years contain the seeds of that distress from the very beginning.

It was predicted that there would be a linear decrease in couple agreement scores across the subgroups as follows: satisfied married, dissatisfied married, cancelled, and separated/divorced. The one-way ANOVA with a linear trend analysis generally confirmed this prediction. This data (Table 1), indicates that the level of couple agreement on important issues is significantly different across the subgroups and that this difference follows a discernible and predictable pattern.

As hypothesized, it was found that cancelled couples would have lower couple (PCA) scores than satisfied couples. These two groups were significantly different on 7 of 11 couple agreement scores. All of the comparisons were in the predicted direction. Clearly, couples who cancelled their marriage plans scored considerably different than those who went on to develop a satisfying relationship.

It was also predicted and found that cancelled couples would not differ from separated/divorced and from dissatisfied couples. This was supported, since only 1 of the 11 scales showed a statistical difference. Couples in cancelled couple had greater agreement on religious values than those who married and eventually became dissatisfied. It is clear from these results that those who cancel their marriages after taking PREPARE, and those who are dissatisfied, are not that different on PREPARE.

These data support the notion that PREPARE not only distinguishes between couples who will eventually become dissatisfied and those who will develop satisfying relationships, but that it was also useful to a large number of couples who decided not to get married after taking it. Although PREPARE users have been cautioned against making specific predictions of marital success (Olson, Fournier, & Druckman, 1986), the inventory can identify high-risk couples and serve as a preventive tool, as suggested by Baggarozzi and Rauen (1981).

If PREPARE is to have a preventive function, then it must be able to correctly identify those couples who are statistically likely to have marital difficulties in the first years of their marriage. The ability of PREPARE to identify the couples at high risk of marital dissatisfaction and divorce indicates both its predictive validity and its potential for preventive work with those couples.

The best predictive combination was the individual scores together with the positive couple agreement score, which correctly classified 91% of the separated/divorced couples and 93% of the highly satisfied couples.

PREPARE can, therefore, be useful in identifying those couples who are at risk in terms of dissatisfaction and divorce. Options that these high risk couples may want to consider are: (a) lengthen their engagement for the purpose of further preparation; (b) obtain professional help in areas such as communication, conflict resolution, role relationship, and issues around the partner’s personality or habits; and/or (c) re-evaluate their marriage plans. As Guerney and his colleagues have amply demonstrated (Avery, Ridley, Leslie, & Milholland, 1980), communication and conflict resolution skills can be taught and maintained by premarital couples.

This study does indicate some relevant topics and issues that premarital programs should address in order to assist couples in developing more satisfying relationships. The 10 PREPARE categories which most clearly discriminate between the satisfied and dissatisfied couples would be useful areas for premarital programming. The topic areas that should be considered are: (a) realistic expectations, (b) personality issues, (c) communication, (d) conflict resolution, (e) leisure activities, (f) financial management, (g) sexuality, (h) family and friends, (i) equalitarian roles, and (j) religion.

The question raised by Baggarozzi and Rauen (1981) regarding whether bad marriages are prevented by premarital programs, can be partially answered by this study. The data indicate that a substantial number of couples (10% of the available population), chose to cancel their marriage plans after taking PREPARE. To what extent these decisions were due to the inventory is unclear, but the fact is that these couples progressed in their marriage plans far enough to contact a clergy member to make plans for marriage. The data clearly indicated that the cancelled couples’ scores on PREPARE were significantly lower than those who went on to develop a satisfying relationship, and were very similar to those who developed dissatisfied marriages. These findings suggest that these couples probably made a wise decision in choosing no to marry that person at that time.

There are several important cautions related to the use of these data. While identifying some of the factors that differentiate successful from unsuccessful relationships, the direct application of these data must be used with caution. While this study provides strong support for the predictive validity of PREPARE, the inventory should not be used for predicting success rates for individual couples.

Also, the accuracy rates for these predictions are probably higher because extreme groups were compared. The couples studied were those who chose to be married by a clergy. This may explain, in part, the importance that religious orientation played in some of the analyses. Participation in the study was also voluntary and was limited to couples who had remained in the parish where they were married. Also, all these couples not only took PREPARE before marriage but they all received an average of two feedback sessions based on their scores. Further work should be done with a broader sample in order to cross-validate these findings.

This study provides some useful data on what differences were found to exist during engagement among couples that eventually developed satisfying relationships versus dissatisfying relationships. It clearly demonstrates the predictive validity of PREPARE and indicates the importance of the premarital period in developing a satisfying marriage.

REFERENCES

Avery, A. W., Ridley, C. A., Leslie, L. A. & Milholland, T. (1980). Relationship enhancement with premarital dyads: A six-month follow-up. American Journal of Family Therapy, 8, 23-30.

Baggarozzi, D. & Rauen, P. (1981). Premarital counseling: Appraisal and status. American Journal of Family Therapy, 9(3), 13-30.

Berscheid, E. & Campbell, B. (1981). The changing longevity of heterosexual close relationships: A commentary and forecast. In M. J. Lerner & S. E. Lerner (Eds.), The justice motive in social behavior. New York: Plenum Press.

Burgess, E. W. & Wallin, P. (1953). Engagement and marriage. New York: Lippincott.

Duvall, E. M. (1971). Family development, 4^th ed. Philadelphia: Lippincott, Co.

Filsinger, E. E. (Ed.). (1983). Marriage and family assessment. Beverly Hills, CA: Sage Publishing.

Fournier, D. G. (1979). Validation of PREPARE: A premarital counseling inventory. Unpublished doctoral dissertation, University of Minnesota, St. Paul.

Fournier, D. G., Olson, D. H. & Druckman, J. M. (1983). Assessing marital and Premarital relationships: The PREPARE-ENRICH Inventories. In E. E. Filsinger (Ed.), Marriage and family assessment (pp. 229-250). Beverly Hills, CA: Sage Publishing.

Fournier, D. G., Springer, J. & Olson, D. H. (1979). Conflict and commitment in seven stages of premarital and marital relations. Unpublished manuscript, Dept. of Family Social Science, University of Minnesota, St. Paul.

Glick, P. C. (1984). Marriage, divorce, and living arrangements. Journal of Family Issues, 5, (1), 7/26.

Gurman, A. S. (1983). Family therapy research and the “new epistemology.” Journal of Marital and Family Therapy, 9 (3), 227-234.

Gurman, A. S. & Kniskern, D. P. (1978). Research on marital and family therapy. In S. Garfield & A. Bergin (Eds.), Handbook of psychotherapy and behavior change: An empirical analysis, 2^nd Ed., New York: Wiley.

Kitson, G. C. & Sussman, M. B. (1977, April). Marital complaints, demographic characteristics, and symptoms of mental distress among the divorcing. Paper presented at the meeting of the Midwest Sociological Society, Chicago.

Levinger, G. & Moles, D. C. (1976). In conclusion: Threads in the fabric. Journal of Social Issues, 32, 21-47.

Locke, H. J. & Wallace, K. M. (1959). Short marital adjustment and prediction tests: Their reliability and validity. Marriage and Family Living, 21, 251-255.

Moos, R. H. & Moos, B. A. (1976). A typology of family social environments. Family Process, 15, 357-371.

Murstein, B. I. (1980). Mate selection in the 1970’s. Journal of Marriage and the Family, 42, 51-66.

National Center for Health Statistics. (1984). Births, marriages, divorces and deaths for November, 1983. Monthly Vital Statistics Report, 32, 11, Washington, DC: Author.

Olson, D. H. (1983). How effective is marriage preparation? In David Mace (Ed.), Prevention in family services (pp. 65-75). Beverly Hills, CA: Sage Publishing.

Olson, D. H., Druckman, J. M. & Fournier, D. G. (1978). Inventory of premarital conflict. Unpublished inventory, Dept. of Family Social Science, University of Minnesota, St. Paul.

Olson, D. H., Fournier, D. G. and Druckman, J. M. (1986). Counselor’s Manual for PREPARE-ENRICH. (rev. ed.), Minneapolis, MN: PREPARE-ENRICH, Inc.

Olson, D. H., Russell, C. S. & Sprenkle, D. H. (1980). Marital and family therapy: A decade review. Journal of Marriage and the Family, 42, 973-993.

Rappaport, R. (1963). Normal crises, family structure and mental health. Family Process, 2, 68-80.

Rausch, H., Goodrich, W. & Campbell, J. (1963). Adaption to the first year of marriage. Psychiatry, 26, 368-380.

Rolfe, D. J. (1975). Marriage adjustment of couples: A pre-marital assessment and follow-up in marriage. Unpublished doctoral dissertation, Michigan State University, East Lansing.

Rozeboom, W. (1966). Foundations of the theory of prediction. Homewood, IL: The Porsey Press.

Rutledge, A. L. (1968). An illustrative look at the history of premarital counseling. In J. A. Peterson (Ed.), Marriage and family counseling: A perspective and prospect. New York: Association Press.

Schaefer, M. (1979). Linking premarital and marital interaction types and dynamics: A longitudinal study of transition to marriage. Unpublished doctoral dissertation, University of Minnesota, St. Paul.

Schumm, W. R. & Denton, W. (1979). Trends in premarital counseling. Journal of Marital and Family Therapy, 5, 23-32.

Springer, J. (1983). Transition to marriage: A longitudinal study of cognitive and behavioral change. Unpublished doctoral dissertation, University of Minnesota, St. Paul.

Terman, L. M. (1938). Psychological factors in marital happiness. New York: Dryden Press.

Weiss, R. S. (1975). Marital Separation. New York: Basic Books