Marital Instability Over the Life Course [United States]: A Five-Wave Panel Study, 1980, 1983, 1988, 1992-1994, 1997

Data Archive > U.S. Surveys > General Population > National > Other > Summary


To examine the causes of marital instability throughout the life course, five waves of data were collected between 1980 and 1997 from married individuals who were between the ages of 18 and 55 in 1980. Information collected in 1980 (Wave I) focused on the effects of wives' participation in the labor force on marriage and marital instability. Measures predicting marital instability and divorce and assessing marital quality were developed. Variables include information on earnings, commitment to work, hours worked, and occupational status. The focus of Wave II, conducted in 1983, was to link changes in factors such as economic resources, wife's employment, presence of children, marital satisfaction, life goals, and health to actions intended to dissolve a marriage, such as divorce and permanent separation. Information on adjustment to marital dissolution, relationship with in-laws, size of home, parents' employment, use of free time, club membership, child-care arrangements, and responsibility for chores was gathered. Wave III, collected in 1988, further examined the impact of changes in employment, economics, and health on marital relationships. Questions were asked about divorce and remarriage, investment of energy and resource use in the care of aging parents and dependent offspring, asset value, awareness of aging, mental health issues, and history of disease. In 1992, a fourth wave of data was collected to look at changes in employment, economics, and health. Questions were asked about retirement issues, family structure, and the impact of caring for aging parents while at the same time caring for dependent offspring. Data were also collected in 1992 and 1994 from adult offspring who were living in the household in 1980 and had reached age 19 by 1992, thus providing parallel measures with their parents regarding the quality of parent-child relationships, attitudes, and support along with exploring the impact of childhood experiences on the transition to adult life. In 1997, the fifth wave was collected and interviews were conducted with a second sample of adult offspring (N=202) along with second interviews of offspring selected in 1992 (N=606). Wave 5 also examines the relationship between marital quality and stability and how it relates to changes in marital quality later in life. Among the variables included in all five waves are age, sex, educational attainment, marital status and history, attitude toward divorce, number of children, religious affiliation, and income level.

Data File
Cases: 2,303
Variables: 4,276
Weight Variable: CASWGT, WATE
Data Collection
Date Collected: 1980, 1983, 1988, 1992, 1994, 1997
Funded By
National Institute of Aging
Collection Procedures
Waves I, II, and III

Participants were first interviewed by telephone in the fall of 1980. Spouses were not interviewed. Random digit dialing was used to select the sample. Among eligible households the response rate was 65%. Three percent were partial interviews, 15% refused, and 17% could not be reached. In 1983, the response rate was 78% and in 1988, 84%. The re-survey in 1983 yielded 1,578 completed interviews and reliable information on marital status of another 150 individuals. In 1988 1,341 interviews plus reliable marital information on another 94 people were obtained. Because divorce was the central focus of the investigation, interviewers were instructed to obtain information on the individual's marital status even if they could not complete the interview or reach the individual. The marital status information could then be used when predicting marital instability from previously obtained interview data.

Wave IV

To maintain the integrity of the panel, respondents were tracked in the spring of 1989 and again in the spring of 1991. In April of 1989 and February of 1991 letters were sent to all respondents, giving them a progress report on the study. The letters were sent with "Address Correction Requested" so that study staff were informed about those individuals who had changed addresses or for whom the letter could not be delivered. Those individuals were tracked and lists of addresses and phone numbers updated.

Everyone who was reinterviewed in 1992 who had offspring 19 years of age or older who resided in the household in 1980 were asked to give us the name and phone number of a child. In cases where there was more than one eligible child, a random procedure was used to select the child to be included in the study. In households with two eligible children, the name, address and phone number of the second offspring was also obtained. In households with three or more eligible children, a second child was selected at random and the name, address, and phone number requested. We interviewed the second child when it was not possible to obtain an interview with the first. We felt it was better to obtain information on one child than none. We obtained phone numbers on 537 first selected and 299 second selected offspring. Six percent of the interviews are with offspring other than the first selected individual. Of the 1,341 people who were interviewed in 1988, we succeeded in reinterviewing 1,193 (89%) of the adults. Among those interviewed, 625 had children 19 years of age who resided in the household in 1980. Eighty-eight (14%) elected not to provide the name and phone number of the offspring. Of the 537 names of offspring provided by parents, interviews were
obtained from 471 (88%) of them.

Wave V

To maintain the integrity of the panel, respondents were tracked in the fall of 1993 and again in the fall of 1994. Letters were sent to all respondents giving them a progress report on the study (see Appendix C). The letters were sent with “Address Correction Requested” so that study staff were informed about those individuals who had changed addresses or for whom the letter could not be delivered. Those individuals were tracked and lists of addresses and phone numbers updated. Offspring of the panel respondents who were interviewed in 1992 were also tracked in the fall of 1993 and 1994. Offspring respondents were sent a letter in 1995 updating them on the study as well as a short survey asking about changes in their work/family life since 1992. A follow-up postcard was sent to those respondents who had not yet returned the short survey.

Respondents who were re-interviewed in 1997 who had offspring 19 years of age or older who resided in the household in 1980, and who did not have offspring who were interviewed in 1992, were asked to give us the name and phone number of a child. In cases where there was more than one eligible child, a random procedure was used to select the child to be included in the study. In households with two eligible children, the name, address and phone number of the second offspring was also obtained. In households with three or more eligible children, a second child was selected at random and the name, address, and phone number requested. We interviewed the second child when it was not possible to obtain an interview with the first. We felt it was better to obtain information on one child than none. We obtained phone numbers on 250 first selected and 119 second selected offspring. Three percent of the interviews are with offspring other than the first selected individual. Of the 1,193 people who were interviewed in 1992, we succeeded in re-interviewing 1,077 (90.3%) of the adults.
Sampling Procedures
The population sampled was all husbands and wives in households in the contiguous United States in which both spouses were present and under the age of 55 and who had access to a telephone. A clustering technique was used to reduce selection costs. It was estimated that the respondents were sufficiently dispersed geographically so as not to create a design effect requiring attention in the analysis of the data. An additional random procedure was used to select a couple if more than one lived in the household and to select the husband or wife.
Principal Investigators
Alan Booth, Paul Amato, David R. Johnson, John N. Edwards
Related Publications
Booth, Alan and David R. Johnson. 1985. "Tracking respondents in a telephone interview panel selected by random digit dialing." Sociological Methods and Research. 14:53-64.

White, Lynn K., and Bruce Keith. "The Effect of Shift Work on the Quality and Stability of Marital Relations." Journal of Family Issues. 55 (1990), 453-464.

Amato, Paul R., and Alan Booth. "Consequences of Parental Divorce and Marital Unhappiness for Adult Well-Being." Social Forces. 69 (1991), 895-914.

Amato, Paul R., and Alan Booth. 1997. A Generation at Risk: Growing Up in an Era of Family Upheaval. Harvard University Press.
Sample Representativeness and Weighting
Using Census data, the investigators evaluated the sample on the following background characteristics: age, race, sex, household size, presence of children, home ownership, husband's education, region, and metropolitan residence. The sample is comparable to the national distribution of married individuals, husbands under 55 on the following characteristics: husband's age, wife's age, race, household size, tenure, region, and presence of children. The sample does have the biases in sex and education expected in survey research; the sample is more likely to be female (59% versus 52% for the population), and better educated (30% with a bachelors degree versus 24% for the population). Because of the higher refusal rates in metropolitan areas only 63% of the sample, compared with 74% of the nation, comes from standard metropolitan statistical areas. This disparity occurs in all areas but the south. To increase the generalizability of the survey, the sample was weighted for metropolitan residence by region.

According to the investigators, attrition in the second and third waves influences the representativeness of the sample only slightly. Table 5 in Appendix 1 shows the 1980 characteristics of those reinterviewed in 1983 and 1988. Attrition occurred among those who were 14-24 years old, black, those with Spanish origin, males, individuals from three person households, renters, those without any college, people from the south, those with children under 6, and those living in metropolitan areas. Out of 54 comparisons between 1980 and 1988 characteristics, 6 were more than by 3 percent and 41 of the comparisons were below 2.5 percent. Probit analysis was used to test whether some demographic variables (age, sex, home ownership, education, presence of children) and five indicators of marital quality (happiness, disagreement, interaction, problems, divorce proneness) affect the probability of not being included in the second and third wave. Home ownership and educational attainment were the only statistically significant predictors of a non-interview. Renters and those with lower educational attainment are more likely to drop from the panel. The panel remains fairly representative of the target population.
Full Reponses for MRELNA80 and MRELNB80
001 Neglects house/home
002 Husband and/or children neglected
003 Husband and wife grow apart - conflicting schedules = problems
004 Contributes to family understanding and growth - appreciation
005 Wife would resent it - have bad attitude if she had to work 8-5 Marriage would suffer (she doesn't want to work)
006 If woman has young, pre-school children/family she should stay home with them
007 Depends entirely on each individual couple and circumstances
008 Husband becomes jealous and threatened - insecure
009 Couple need to consider each other - understand and care for
010 Man feels interview should ask about whether wife helps him with chores, i.e., mowing, snow shoveling
011 Wife working has little or no effect on marriage - other factors more important
012 Wife gets "too independent and career minded" - self-centered - threatens marriage
013 Children in families where mom works are more independent - well adjusted
014 Wife needs to be more organized at home when she works (organization is critical)
015 Wife with children can work part-time with few problems
016 Increased respect for wife who works and contributes to family (both should work)
017 Family, especially husband, has to share chores and responsibilities
018 Wife gets too involved with social aspects outside family - especially men - affairs
019 Puts stress on marriage/family - causes problems
020 People shouldn't care/they get too involved with material things
021 Children more apt to go astray and be insecure
022 Wife is less dependent on husband/likes independence
023 If wife doesn't work, she could become an alcoholic - lose self-esteem
024 Competition over who's in charge of the money - his/ her money - who's paying the bills
025 Woman feels better about herself - happy
026 Women's lib is destructive to the home
027 Wife becomes resentful because she has job and housework (husband has job only)
028 A strong marriage can stand wife working
029 Place/calling for woman is in home/wife and mother is demanding role
030 The reason women work is to pay bills
031 Loss of communications (need to keep open)
032 Good for women to work if they want to/no problem
033 Woman has to set goals/priorities for herself/can handle both home and job
034 Wife is bored/unhappy staying home
035 People you work with pressure you into their lifestyles
036 Working mothers (may) cause divorce/break up
037 Woman more in touch with world - better outlook
038 More money - standard of living up
039 Makes for conflict of interests
040 Just don't think woman should work
041 A woman who works carries a double load
042 Put physical and mental stress on woman
043 Couple needs to work everything out beforehand - housework finances - all the ramifications
044 One parent must be home to nurture the children 9
045 Children are ones most affected when wife works 25
046 Okay if husband is agreeable (mutual agreement); if not, problems
047 Man looks upon wife's education as investment - wants a return
048 Girls nowadays should be trained for some work
049 Couple aren't going to have children so wife has plenty of free time even though she works
050 Family that has faith in God can be happy
051 There is great lack of adequate child care (hard to find someone)
052 Now there is more flexibility for women to do what they want
053 Its okay/good for wife to work if just husband and wife at home
054 A wife who is really happy doesn't go to work
055 Women should do volunteer work, especially in schools
056 Couple/family get less time together if wife works
057 Couple shouldn't bring home problems
058 Women who work seem to be divorced or single
059 Women shouldn't take a man's job
060 Costs too much for clothes, car, child care, if wife works
061 Doesn't want wife gone overnight
062 Woman has hard time adjusting to being a housewife after working in her profession - misses career
063 The Lord/Bible wants women to be helpers/mothers
064 Causes changes in roles and priorities of wife and husband
065 It is not the quantity of time the mother spends with kids, it's the quality
066 Problems arise cause people are inflexible
067 Wife only working to help husband
068 Respondent had good experience growing up with both parents working
069 Wife works for security in case of husband's death
070 Wife who works has no time for activities outside home (school volunteer)
071 Meeting other men/women
072 Depends on kind of job/hours, i.e., stripper or nurse
073 Makes for misunderstanding/poor communications
074 Okay as long as both work same hours
075 Gives woman a chance to get out of a bad marriage
076 Enhances marriage - better communication - sharing - closeness
077 Children want their mother home - doing more for them
078 Wife might not be home when needed (late dinner)
079 Hard to plan week end trips and vacations
080 Two people are working for common goal
081 Okay to work but put family first before job
082 Woman shouldn't work unless absolutely necessary
083 Woman can work if kids are small; not when 12+
084 "Bad role model" for boys if woman stays home - they think women shouldn't work
085 Its hard to start work after you've been home entire married life
086 Spouse start spend money unwisely
087 More responsibility placed on the man
088 Okay if mom is home when kids are home
089 U.S. economy causes most family/marital problems
090 Women are discriminated against in business world
091 Man resents/refuses to do domestic chores
092 If woman works - go right after baby is born rather than 3 years (it feels rejection)
093 Problems because "head of household" idea is disappearing
094 Need more hours in the day/hectic
095 Family was helped by Family Service Counseling
096 We eat out more often
097 Big difference between "have to" - "want to" work
098 Don't worry about money problems
099 Working helps wife keep her sanity
100 More divorces when wife makes more money
101 Husband/wife works different shifts - stressful
102 Goals should be defined adequately/goals/be aware of roles
103 If wife doesn't like job, can cause problems
104 I do my duty, what's expected for my family
105 Woman who works misses out on a lot of family things
106 If people want to stay married don't live in Dallas, Texas
107 This is my own opinion
108 Misunderstandings of husbands ego
109 Good and bad effects (wouldn't clarify)
110 It's a way of life for young couples
111 Wife felt guilty about leaving baby to go to work;
husband felt she thought more of baby than him
112 Wife worked some/stayed home some - no problem - worked out okay
113 It all depends on whether wife is happy with job
114 Each should maintain individuality and be able to give to the other
115 Bad enough one person coming home irritable, not two people
116 Man should be boss/breadwinner
117 Couple relies too much on wife's income
118 Lack of cooperation from family
119 If it's week ends and nights, its difficult for family
120 Couples would have to "make time for each other"
121 Women have more to contribute than just staying home
122 Husband makes more money for household
123 Wife expects too much from man/gets bossy
124 Exposure to people and things good
125 Children tend not to be as close to the parents/mother when she works
126 I think it enlightens and broadens marriage and the whole family
127 Everything is great if wife keeps her cool
128 I was more conscientious of children; it could have been guilt motivation
129 I feel environment is important for the security of the children; they can sense if she is working for her pleasure
130 My working allows him to be lazier
131 He has guilt feelings about income difficulties
132 Gives one less private, personal time
133 Father gains more understanding for children by watching children when wife works
134 If both parents work, kids must have good child care
135 Problems when man is student and wife working
136 Both salaries should go in common pool
137 Not fair for wife of high income husband to work
138 Husband don't like her to work, but they need the money
139 Husband says when women get money, they think different
140 In many cases they want things to be the same
Marital Instability Scale
These indices of marital instability are based on 27 items asked of people who are married. The initial development of the scale is described in detail in the following publications:

Booth, Alan, David Johnson, and John N. Edwards. 1983. "Measuring Marital Instability." Journal of Marriage and the Family 45: 387-393.

Edwards, John N., David R. Johnson, and Alan Booth. "Coming Apart: A Prognostic Instrument of Marital Breakup." Family Relations 36: 165-170.

Johnson, David R., Lynn K. White, John N. Edwards, and Alan Booth. "Dimensions of Marital Quality: Toward Methodological and Conceptual Refinement." Journal of Family Issues 7: 31-49.

The scale was revised slightly in 1984 and it is that revision that is available for all three years. Higher scores indicate greater marital instability. The variable was logged to bring the distributional quality more into line with the assumptions needed for multiple regression.
Marital Problems Scale
A summated scale using 13 items indicating presence of marital problems because either or both spouses: 1) gets angry easily; 2) gets easily hurt; 3) is jealous; 4) is domineering; 5) is critical; 6) is moody; 7) won't talk to the other; 8) has sexual relationship with others; 9) has irritating habit; 10) is not home enough; 11) spends money foolishly; 12) drinks or uses drugs; and 13) has been in trouble with the law. Higher scores indicate greater marital problems.
Marital Happiness Scale
A summated scale using eleven items reflecting amount of happiness with: 1) extent of understanding received from spouse; 2) amount of love received; 3) extent of agreement about things; 4) sexual relationship; 5) spouse as someone who takes care of things around the house; 6) spouse as someone to do things with; 7) spouse's faithfulness; 8) evaluation of marriage as very happy, pretty happy, or not too happy; 9) compared to other marriages, respondent's is better, same, or not as good; 10) comparing marriage to 3 years ago, it is getting better, staying the same, or getting worse; and 11) strength of feelings of love respondent has for spouse.
Marital Interaction Scale
A summated scale using five items: 1) eat main meal together; 2) go shopping together; 3) visit friends together; 4) work around home together; and 5) go out together. Scale has possible values from 5 to 20. Higher scores indicate greater interaction.
Sex Roles Scale
A summated scale using seven items: 1) wife's most important task: caring for children; 2) husband should earn higher pay than wife; 3) husband shouldn't worry if wife gone overnight in connection with job; 4) if wife works full-time, husband should help with housework; 5) if jobs scarce, wife shouldn't work; 6) working mothers can have just as good of a relationship with kids; and 7) even if wife works, husband should be main breadwinner. Response categories were: 1) strongly agree; 2) agree; 3) disagree; and 4) strongly disagree. Higher scores indicate more traditional values.
Marital Disagreement Scale
A summated scale using the following items: 1) disagreements over share of housework done by respondent; 2) frequency of disagreements with spouse; 3) number of serious quarrels with spouse within last two months; and 4) arguments involving physical abuse. Scale values range from 0 to 12. Higher scores indicate greater disagreement. The variables were z-scored and added together.
Wife's Job Commitment Scale
A summated scale using four items:

1. Wife works because she wants a career
2. Wife works for a feeling of accomplishment
3. Wife works because she wants financial independence
4. Wife's work preference: full-time, part-time, none