University of Southern California Dissertations and Theses

The Development And Evaluation Of Three Therapeutic Group Interventions For Widows

(USC Thesis Other)

The Development And Evaluation Of Three Therapeutic Group Interventions For Widows

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1.The sign or "target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity. 2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete. 4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received. Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 75-1044 ■ BARRETT, Carol Jeanne Clark, 1948- THE DEVELOPMENT AND EVALUATION OF THREE j THERAPEUTIC GROUP INTERVENTIONS FOR WIDOWS. University of Southern California, Ph.D., 1974 Psychology, clinical t Xerox University Microfilms, Ann Arbor, Michigan 48106 © 1974 CAROL JEANNE CLARK BARRETT ALL RIGHTS RESERVED THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED. THE DEVELOPMENT AND EVALUATION OF THREE THERAPEUTIC GROUP INTERVENTIONS FOR WIDOWS by Carol Jeanne Barrett A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) August 1974 U N IV ER S ITY O F S O U T H E R N C A LIF O R N IA T H E G R A D U A T E S C H O O L U N IV E R S IT Y P A RK LO S A N G E L E S . C A L IF O R N IA 9 0 0 0 7 This dissertation, w ritten by ......... under the direction of h$F.... Dissertation Com mittee, and approved by a ll its members, has been presented to and accepted by The Graduate School, in p a rtial fu lfillm en t of requirements of the degree of D O C T O R O F P H IL O S O P H Y f 'T r i Dean D a te. DISSERTATION COMMITTEE ACKNOWLEDGMENTS I would like to warmly thank the faculty members who encouraged my research--Jim Birren, Steve Frankel, Judy Friedman, Jim Peterson, Dick Brown, Pauline Ragan--and especially A1 Marston. I would also like to thank Lisa Pomeroy and Linda Thistle who worked with me as group leaders and gave so much, and Cruz Diaz and Robin Kunert who helped to code the data. I am grateful to the Institute of Lifetime Learning in Long Beach for permission to do the pilot study for my dissertation research. And thank you Al, for letting me grow Steve, for your hugs Linda, for always hearing me Ruth and Marion, for never doubting Gert, for your consistent interest Elwin, for daring to be with me in so many ways. And to the women I love, whose lives are the real substance of the pages that follow, I would like to say that God is a widow. TABLE OF CONTENTS Page ACKNOWLEDGMENTS..................................... ii LIST OF TABLES..................................... v Chapter I. INTRODUCTION ............................... 1 II. REVIEW OF THE LITERATURE.................... 3 The Cultural Context Demographic Features of Widows The Stresses of Widowhood Variation in Response to Widowhood Services for the Widowed III. THE RESEARCH METHODOLOGY.................. 61 The Design of the Study The Subjects The Group Interventions Subject and Dependent Variables Method of Data Analysis IV. PRESENTATION OF RESULTS.................... 83 Sample Description Pretest Analysis Manipulation Check Analysis of Change in Psychological Functioning, Reactions to Widowhood and Life Style Analysis of Change in Attitudes toward Women Analysis of Behavioral Measures Analysis of Group Evaluation Measures iii Chapter Page V. DISCUSSION OF THE RESULTS................... 130 The Sample Assignment of Subjects Response to the Program: Treatment Versus Control Group Response to the Program: Comparison of Three Treatments The Nature of Change Over Time The Importance of Subject Variables VI. CONCLUSION................................. 151 REFERENCES......................................... 155 APPENDICES A. News Release for Obtaining Subjects .......... 164 B. Initial Letter to Participants .............. 167 C. General Consent Form ......................... 169 D. Letter to Waiting List Controls .............. 171 E. Suggested Widows' Consciousness-Raising Topics 173 F. General Information Form ..................... 175 G. Reactions to Widowhood ....................... 187 H. Widowhood Project Evaluation Form ............ 196 I. Roles of Women ............................... 202 J. Widowhood Research Project Final Evaluation F o r m ....................................... 206 K. Special Evaluation Form for Confidant Groups 210 iv LIST OF TABLES Table Page 1. Age-Standardized Percentiles of Women Who Are Widowed, by Ethnic G r o u p............ 1 2. Frequency of Depression Among 280 Aged, Widowed and Married Persons With and Without a Confidant ....................... 39 3. Percentage of Widowed Callers Requesting Various Types of Help, by Employment Status 46 4. Percentage of Widowed Callers Requesting Various Types of Help, by Duration of Widowhood................................. 48 5. Age Distribution of the Sample............. 84 6. Duration of Widowhood in the Sample......... 84 7. Religious Affiliation of the Sample......... 85 8. Education of the Sample..................... 87 9. Employment Among the Subjects............... 88 10. Subjects' Total Household Income Prior to Widowhood and at Present................. 89 11. Analyses of Variance on Age, Duration of Widowhood, Cause of Death, and Preparation for Widowhood............................. 92 12. Chi-Square Analysis of Perceived Role of Group Leader by Treatment................. 94 13. Change in Attitude Toward Remarriage at Posttest as a Function of Group Leader . . 96 v Table Page 14. Mean Change in Self-Esteem, Grief, Health Prediction, and Attitude Toward Remarriage at Posttest............................... 98 15. Analysis of Variance on Change in Psychologi cal Functioning, Reactions to Widowhood and Life Style at Posttest.............. 99 16. Mean Change in Self-Esteem, Grief, Health Prediction, Attitude Toward Remarriage and Social Roles at Follow-up................. 104 17. Analysis of Variance: Test for Non-Zero Change (Error Effects) in Psychological Functioning, Reactions to Widowhood and Life Style at Follow-up................... 105 18. Analysis of Covariance at Follow-up: Regression Effects Influencing Change in Self-Esteem and Attitude Toward Widowhood 107 19. Posttest Analysis of Change in Attitudes Toward W o m e n ............................. Ill 20. Posttest Analyses of Covariance on Attitudes Toward Women: Regression Effects for Duration and Preparation................. 113 21. Analysis of Change in Attitudes Toward Women at Follow-up............................. 115 22. Follow-up Analyses of Covariance on Attitudes Toward Women: Regression Effects ......... 117 23. Extent of Contact with Group Members During Treatment Period......................... 119 24. Group Means and Standard Deviations for Contact with Members and Life Change at Follow-up................................. 120 vi Table Page 25. Multivariate Analysis of Variance on Contact with Group Members and Life Change at Follow-up................................. 121 26. Subjects' Evaluation of Treatment Groups at Posttest................................. 124 27. Subjects' Evaluation of Treatment Groups at Follow-up................................. 126 28. Confidant Relationships at Posttest and Follow-up................................. 127 29. Group Composition Preferences at Posttest Among Experimental Subjects.............. 129 vii CHAPTER I INTRODUCTION America's widows comprise one of its most oppressed minority groups. They face prejudicial reactions as women, as old persons, as poor persons, often as members of ethnic minorities, and always as reminders of death. The poten tial resources in this group of ten million strong are vast, but largely untapped. Psychological and social services for the widowed are extremely rare. The purpose of this research was to develop a program of therapeutic discussion groups for women dealing with the multiple stresses of widowhood, and to assess thoroughly the relative efficacy of three different group interventions: (1) a self-help group, (2) a "confidant" group to help each member develop an intimate relationship with another widow, and (3) a women's consciousness-raising (C-R) group for widows. Chapter II will review the cul tural milieu of widows, describe their demography, discuss the stresses they experience, consider the variables which are important in determining a widow's response to her environment, and describe the current institutional resources for the widowed. The methodology of the research will be presented in Chapter III, and the results of the pretest, posttest and follow-up in Chapter IV. The final chapters of the dissertation will discuss and summarize the research results and outline their implications for a responsible, problem-solving societal response. 2 CHAPTER II REVIEW OF THE LITERATURE The Cultural Context The denigration of women whose husbands have died is an almost universal dowry bestowed on millions of brides all over the world. This phenomenon is perhaps most impli cated in the ancient Hindu rite known as "suttee” whereby a woman achieves faithfulness by being burned alive with the body of her late husband. The custom was prevalent in India, especially among the Punjab, by the fourth century B. C. Although outlawed by the British government in India in 1829, cases have continued to appear (Thompson, 1928). In Hindu theology a woman being left a widow is an indica tion that her sins in her previous life merited the heaviest of all punishments--the loss of her visible God, or husband. Thus for justice to occur widowhood must be such a miserable experience that it is better to die in the flames surrounding the husband's corpse. That widowhood may in fact be associated with such misery is perhaps 3 evident in the following quotation: Today, when suttee is forbidden, the life of widows of chieftains in Central India is often too sordid for contemplation. There are fortresses packed with these wretched creatures, who exist there without ornament or amusement or pleasant food, and have no relief except squabbles among themselves and banding together to make another newly arrived consignment more unhappy than themselves. A lady who knew well many of these corralled unfortunates told me that she thought it would be a reform to reintroduce suttee. (Thompson, 1928, p. 49) Widow sacrifice has also occurred in some parts of New Zealand and Africa (Westermarck, 1921; Sumner & Keller, 1929). Perhaps less dramatic are society's strictures concerning a widow's escape from her status by remarriage. Patterns range from a prohibition of remarriage to manda tory remarriage to a prescribed mate. Marrying widows is taboo among certain castes in India, although there is evidence that some rural Brahmans are becoming more accept ing of such unions (Singh, 1969; Dubey, 1965). At the other extreme is the Hebrew custom of "levirate" which dictates that the brother of a man who dies without children must marry the widow, even if he already has a wife. According to the "sororate" a man must marry the sisters of his deceased wife. But almost universally cultures are more permissive in the case of remarried widowers than widows (Bernard, 1956). United States statistics for frequency of remarriages among widows and widowers (described below) are not inconsistent with this observation. Smith (1962) has distinguished three societal patterns of widowhood. Countries such as Egypt, Ceylon and India where women marry young and the mortality rate is high have a very high proportion of widows. In Canada, Australia, New Zealand, and the United States the relative number of widows is intermediate. In Mexico, Central and South America the high death rate is offset by a high proportion of men and women who never marry, and Smith considers the role of widowhood in the society as unimpor tant. This may be challenged however from another perspec tive. In Mexico, the status of divorcee is apparently so much less desirable than that of widow that families may frequently be persuaded to corroborate the fiction that one's husband has died. Unmarried mothers may claim both a former marriage and bereavement (Folan et al., 1968). In our own culture, the oppressive condition of widowed women can be observed in the countless numbers of widows who hate the word "widow," and in their complaints that others respond to them "as if I had an infectious disease.” A more subtle manifestation of the status of widows in this country resides unobtrusively in the impressive buildings of our public libraries. The number of comic plays about widows are apt to outnumber the col lection of lay books offering advice or relating personal experience (e.g., Owen, 1950; Egelson and Egelson, 1961; Richford, 1953; Champagne, 1964; Langer, 1957; Decker and Kooiman, 1973). Social scientists have only recently begun to assume responsibility for the study of widowhood. In an impressive review of the role changes in widowhood cross- culturally, Lopata (1972) has ranked the subsequent life style of the widow from that requiring the least change from her life style as a wife to that requiring the most. The United States norm is high on this hierarchy. The typi cal widow lives alone and undertakes new roles through her own initiation with a wide range of life styles available, but reengagement patterns are not adequately socialized. The often dramatic change in life style imposed on the American widow takes place in the context of the larger society which already grants fewer rights and privileges to women. There appears to be much potential power for social change arising from the discontent of our 6 large population of widowed women. The current level of oblivion in which many live perhaps accounts for the high frequency with which as the investigator in this research I was quizzically asked, "How did you get interested in that?" Demographic Features of Widows The description of the demography of widows that follows is based heavily on 1960 Census reports, since detailed analyses of the 1970 Census are not yet available. Prevalence In 1970 there were 12 million widowed persons in this country. Approximately 10 million of them were women (Abrahams, 1972). Widows thus outnumber their male counterparts almost five to one. They make up almost 5% of the total United States population. Their absolute numbers have been increasing by more than 100,000 a year, compared with an annual increase of about 80,000 in the 1920s and 50,000 at the turn of the century (Metropolitan Life Insurance Company, 1962). There is disagreement as to whether the relative occurrence of widows is increasing in the population. Although Smith (1962) claims the number of widows is increasing more rapidly than the general popula tion, due to the aging of the population, Carter and Glick (1970) compared census data for 1900, 1930, and 1960 and found an overall reduction in the percentage of widows in the country, offering higher remarriage rates as an explanation. The rate at which widows outnumber widowers has been increasing; sixty years ago there were only about two widows for every man in widowhood (Metropolitan Life, 1962). Probability Most women get married; despite high divorce rates, many of them will be widowed at some time in their lives, since currently women outlive men by six to seven years, and they tend to marry men older than themselves. Of those who find a partner the same age, 60% will be widows. The chances are 70%> when the husband is at least five years older than the wife (Bell, 1963). Only if the wife is at least five years older than her husband is it probable that she will die before him (Cosneck, 1966). The annual rate of widowhood is about 12 to 14 widows for every one thousand married women. The rate for widowers is 6 to 8 per one thousand married men. About 8 twice as many persons became widows as widowers in the period 1940-1963 (Carter & Glick, 1970). Age The median age of widows is 64 (Carter & Glick, 1970). In 1950, the median age of white widows was 66.4 years; for non-white widows the median age was 58.8 years. Of women aged 65-74, 407o are widows. For women over 75, 69% are widows (Cosneck, 1966). Yet, about one fourth of all widowed persons are younger than 65. Duration It is known that mortality rates for widowed persons from all causes, for every age group, for both sexes, and for both whites and non-whites are higher than those for married persons (Kraus & Lilienfeld, 1959). Despite this fact, the average widow who dies of a natural cause will have spent eighteen and one-half years of her life as a widow. (The average period of widowerhood is 13.5 years. Carter & Glick, 1970.) According to 1960 computations, Three fourths of the women who become widows at age 45 can expect to live an additional 25 years. The same number of years remain to 9 out of every 10 women widowed at ages 35 or earlier. Of the widows bereft of their spouse at age 65, somewhat more than half can expect to live 15 years longer and about a third still have 20 years of life before them [Metropolitan Life, 1962]. One fourth of all widows remarry within five years. (One half of widowers do so.) The likelihood of widow remarriages decreases sharply with increasing age (Bernard, 1956). Each year, the number of new widows is about twice the number of widows who remarry. For those who do remarry, the average duration of widowhood is three and one-half years. The average current duration of widowhood nation ally is not known, but can be estimated, The median length of time since the husband's death was about four years for widowed families entitled to survivors1 benefits from the federal government in 1962 (Palmore et al., 1963). (This may be a slight overestimate for the entire population of widows, since those with children do not remarry as fast as those without.) In an area probability sample of Chicago widows over age 50 (Lopata, 1973), 68% were widowed five years or more ago. Ethnicity Due to differential mortality rates among ethnic groups, racial minorities are overrepresented in the ranks of the widowed. Carter and Glick's (1970) presentation of 10 the percentage of women widowed in various ethnic groups is reproduced in Table 1. Notice that the highest rate is endured by Black women. The low Filipino widowhood rate is attributed to a very high remarriage rate. Comparable data for Chicanos will not be available until the 1970 Census can be analyzed with respect to Spanish surname persons. Among foreign-born persons or children with one foreign- born parent, the highest widowhood rates are among Mexicans and Irish. Education There is no statistic available on the amount of education of the average widow. Palmore et al. (1963) found that the educational level of a large sample of widowed mothers was similar to that of women in the general population. However, Glick (1957) noted a "simple correla tion between median age at widowhood and amount of educa tion because those with less education no doubt have higher mortality rates and therefore are at a greater risk of losing their husbands through death at a relatively young age." The 1950 Census data revealed higher widowhood rates among non-high school graduates than among high school graduates at all ages 14-54. And Carter and Click's (1970) • 7 0 Table 1 Age-Standardized Percentiles of Women Who Are Widowed, by Ethnic Group Ethnic Group Percentage of Women Widowed Black 17.2 American Indian 13.2 Japanese 13.2 Chinese 13.1 White 11.6 Filipino 7.9 Total, women aged 14 and over 12.2 *Reproduced from Carter and Glick (1970). 12 analysis of 1960 Census data showed a high negative corre lation between percentage of the population widowed and amount of education for persons aged 35-54. In addition, the rate of decline in percentage of the population widowed between 1940 and 1960 was more rapid for the well educated. Employment Widows are much more likely to work than married women of the same age. (Widowers are less likely to be employed than same-age married men, but still more likely than widows to work.) "A substantial majority of widows in the main working ages are found to be working at paying jobs [Carter and Glick, 1970]." Mothers receiving survi vors' benefits are more than twice as likely to work as mothers with husbands (62% versus 28%) even controlling for age and number of children. Yet their unemployment rate is about three times as high as that of other women (Palmore et al. , 1963). Overall, 26.97, of white widows are employed; 35.5% of non-white widows are employed (Carter and Glick, 1970). In an analysis of type of jobs held by working women aged 45 to 54, Carter and Glick (1970) found an over representation of widows in the service trades such as r '( private household workers and in the managerial and proprietary occupations including farming. They postulated that many of the widows' managerial positions are in retail stores formerly managed by the husband. Low percentages of widows are found among women in professional, clerical and factory jobs. One half of the employed white widows 45 to 54 were in white-collar positions, where as nearly one half of the employed non-white widows were doing private household work. Income Although widowhood can generally be assumed to result in greatly reduced income, there is no statistic available for the average income of widows. In a study of 1,774 women who had been widowed approximately two years and whose husbands died before reaching the age of 65, Nuckols (1973) found that family incomes were reduced an average of 447, from predeath levels. The change varied from a 4% increase among families used to less than $3,000 to a 57% decrease among families formerly receiving over $15,000. Further data on the income of widows is available on two subgroups: those who are employed, and those 14 widowed mothers who receive survivors1 benefits. According to Glick (1957) the aged widow who personally earns an income will still be poorer than her divorced, married or single sisters. But in a more recent analysis, Carter and Glick (1970) report that the median income of middle-aged widows with an income (assumed to mean personal earnings) was around 10 to 207, higher than that of married women of comparable age with an income. (A larger proportion of these widows worked full-time during the week and during the year than was true of the married women.) The modal income bracket for working white widows was $1,000 to $2,999. They averaged two and one-half times as much as working non-white widows. To date, widows without any dependent children will not be aided by the federal government until they are 62, and then only if the widow or her husband was insured under Social Security. (Nine out of ten workers are insured.) There are about two million such under-age widows without dependents. An extensive study of those widows and children who managed to meet the requirements for survivors' benefits in 1962 has been made (Palmore et al., 1963). There was a disproportionately large number of non-whites in the study 15 group, and an overrepresentation of families in the South, reflecting ethnic and geographical differences in mortality rates. One fourth of the families had incomes below the poverty level established by the Social Security Adminis tration. Two fifths are below the "low-income" level. These figures are about twice as high as those for the average United States family. The median total income of eligible families is $3,570. Even on a per capita basis this is less than three fourths of that available to the average family in the United States. Old-Age, Survivors, Disability and Health Insurance (OASDHI) program was the major source of income for almost two thirds of the families. Other sources of income included employment earnings, government employee survivor payments, Veterans Administration benefits, insurance, income from investments, public assistance (only 4% of the families qualified) or contributions from relatives and friends. Three fourths of the families supplemented their income with earnings. Few worked full time all year. The average earnings for employed widows were only about three fourths that earned by the average woman worker. The hypothesis that the children would be able to make a substantial financial contribution by working was not confirmed. Only one in 16 twenty families had received any cash aid from relatives. Geographic Distribution There are disproportionately high rates of widows in New England, the Middle Atlantic states, Florida, Georgia, California and Washington, D. C. The Western states (except California) generally have disproportion ately low rates (Smith, 1962). Palmore et al. (1963) found the South the most heavily represented of the four major geographic regions among widowed families receiving survivors' benefits. In terms of the urban-rural dimension, the ratio of widows to other women is lower in farm areas than in small towns and cities, except for women over age 75; among them the urban and rural ratios are about the same (Metropolitan Life, 1962). Family Size In a Chicago area sample of 301 widows over age 50, one fourth had no living children (Lopata, 1970). The children of most widows are no longer dependents. An estimated 627,000 widows have dependent children (Palmore et al., 1963). OASDHI beneficiary widow families tend to to be small; almost one half have only one child under 17 age 18. The average age of these mothers is 46, of the children, 14. (This is older than the typical family in the United States.) Only 8% of the children are preschool age. Dwelling Unit In a Chicago sample, 49% of widows over 50 live alone (Lopata, 1971); 297. share their residence with one other person; 127. with two others; and 10% with more than two. Most (69%) of the widows who share their household are the heads of the household: unmarried children are still there, separated, divorced or widowed children return to it, or they take in other relatives, friends or roomers. Only 10%, of the total sample live in a household headed by their children or children's spouse. The 1960 Census data confirm the fact that the majority of widowed persons are heads of their own households (Carter & Glick, 1970). However, OASDHI beneficiary widowed families are more than twice as likely to share households than are other families, probably for financial reasons (Palmore et al., 1963). One half of the widows living alone and two thirds of the widows who head larger households in Chicago are 18 still living in the same dwelling where they lived prior to becoming widows. In contrast, most widows in households headed by someone else moved there from another community (Lopata, 1971). Of the widowed, 5% live in institutions. Over one half of this group live in homes for the aged or dependent, such as nursing homes. About 75,000 persons, or 15% of the institutionalized group reside in mental hospitals. Cone1usion From the preceding data we can describe the typical widow in this country as a white woman about 64 years old whose husband died about four years ago. She will probably be widowed until her death another fifteen years from now. She still lives where she and her husband made their home in a large metropolitan area. She is about equally likely to be living alone as with someone else. If she does share her household, she is apt to be its head. She may have somewhat less education than other women her age. She is probably not employed and her income is definitely not substantial. 19 The Stresses of Widowhood The death of a spouse is consistently seen as a major life stressor requiring more readjustment than absolutely any other life event. This finding has been replicated with subjects of varying ages and diverse cultural backgrounds by a group of researchers at the University of Washington attempting to tease out life changes associated with illness susceptibility (Holmes & Masuda, 1970). Among social scientists, the question of whether this stress is greater for women or for men has been answered differently, based primarily on an analysis of sex roles in the culture and role disequilibrium at widowhood. But, as Berardo (1970) has pointed out in a review of the situation facing elderly widowers, "there is little consistent evidence that survivorship status demands more drastic role changes for either sex [p. 14]." It is clear that some of the stresses are different, and the focus of this chapter will be on those most relevant to the widow. Grief The pain of bereavement itself is the first stress facing the widow. The grief process has been described by 20 a number of writers, most notably Lindemann (1965), Harris (1958), and Parkes (1970). Marris' study of 72 British lower-class widows recognized the following frequent phenomena: a sense of futility that anything in life is worthwhile, an inability to understand the loss, reliving of shared experiences, a sense of injustice at one's fate, and a need to blame. A wide range of physical symptoms which the widow herself or her physician thought were caused or aggravated by the shock of the husband's death were reported, most frequently an initial inability to sleep. In the initial struggle to accept death some were bothered by obsessive memories of the circumstances of the death or illusions of the husband's presence. Some widows appeared to cultivate a sense of his presence, i.e., by talking to his photograph and imagining him advising her. For others any reminder of the deceased revived the grief. In contrast to others' studies, Marris found little self accusation. (Only 8 in his sample. It is possible that he missed this phenomenon because it occurred prior to his observations, which were on the average about two years after the death.) Rather, many elaborated their sacrifices and hardships as if defending against unspoken criticism. Marris also found a tendency for the widows to withdraw 21 from people and reject consolation. In a milestone paper in the study of grief, Parkes (1970) reported the experience of 22 London widows under age 65 studied longitudinally for thirteen months after bereavement. Among his findings were that most Ss failed to accept warnings of their husbands' impending deaths, and the immediate reaction to the death was a phase of numbness, followed by a phase of yearning in which the "pangs" of grief occurred. The fact of the loss was often avoided or disregarded, and several identification phenomena were experienced, such as a tendency for the widow to behave or think more like the spouse, to develop symptoms closely resembling the husband's last illness, or to experience the husband as inside oneself or one of the children. Rest lessness and a perceptual set for the dead husband with evidence of an urge to search for him were also observed. Lindemann (1965) described what he considers to be pathological reactions in our culture among some of his 101 bereaved patients. These include taking on traits of the deceased, especially symptoms of the last illness, and exaggerations of the normal reactions such as guilt, feel ings of going insane, and hostility toward friends and relatives trying to help. He believes the "grief work" 22 will go more quickly if the tremendous pain and its expres sion are not avoided. A question which has only begun to be asked is the extent to which differences in grief reactions may predict later adjustment to widowhood on a variety of dimensions. Several writers have postulated that bereavement is less distressing in cultures which sanction elaborate public displays of grief. Marris (1958), for example, has pointed to the perfunctory style of mourning in Western cultures, and Gorer (1965) has described it as deinstitutionalized. To this observer it seems that the ambiguity with regard to "proper" mourning in the United States makes early widowhood especially difficult. Thus in cultures where a specific mourning period is prescribed, the widow deserves to give in to her despair, but also has clear expectations of when to begin functioning normally again. In a retrospective study of aged Americans who cited the death of a loved one as a major stress to them, Anderson (1965) found that those later judged as psychi- atrically impaired tended to be unable to act at the time of the stress compared to a more normal group who took more initiative in coping with the stress. The impaired sample also seemed more "strain-directed," that is, oriented toward assuagement of the physical or psychological dis comfort of grief, for example, by getting away for awhile or talking about it. The unimpaired group in contrast was "stressor-directed"--oriented toward removing, resolving or alleviating the distressing circumstances themselves, for example by finding employment. The Economic Burden Virtually every study of widows has commented on the reduced financial resources characteristically ensuing after the husband's death. Aside from the loss of income from the husband's employment, there are apt to be large bills associated with prolonged illness and funeral arrangements. Widowed families receiving survivors' bene fits in 1962 spent considerably more of their income on food and housing than the average family (Palmore et al., 1963). Thus, fewer had enough money for other needs including life or health insurance, or cars. Relatives of working widows frequently cared for the children in the widow's own home, but many widows found it necessary to leave their school age children unattended while they worked (157o of children aged 6-11 were not cared for) . Almost none of the children in the study were cared for at a day 24 nursery or other group care center, A recent study has extensively described the financial situations of 1,744 women who were widowed "prematurely"; e.g., their husbands died before reaching age 65 (Nuckols, 1973). The subjects, 5%, of whom had remarried, were obtained from the death certificate files in Boston, Houston, Chicago, and San Francisco, and inter viewed approximately two years subsequent to the onset of widowhood. The typical widow faced $2,860 worth of final expenses; the average final expense was $3,900. Over half of those receiving life insurance benefits received their first check within two weeks of filing a claim, whereas social security payments typically took more than three months to be processed. Few widows had had anyone to talk with concerning options for settling life insurance claims, such as a lump sum payment or annuity, and many were not aware they had any options. Two thirds had proceeds remaining from insurance after meeting their immediate living expenses and paying final expenses from the husband's death. The average per capita monthly income payment from all sources was $155. The majority of husbands (71%) had died without a will, and only one fourth had individually programmed life insurance. In response to the question, "Can widows maintain their living standards following the premature deaths of their husbands?" the study concluded that "about one half will be able to do so but that only one fourth will be living comfortably and reasonably free from financial worries." The highest source of income received by the widows was their own earnings (40%). Whereas 47% of the Ss were working prior to the husband's death, two years later 567» were working. For those with lowered standards of living, the most frequent cutbacks were for clothing, social and recreational activities, and food. In this sample of widows, financial problems were rated as the second most serious stress of widowhood; loneliness was the first, and the difficulty of raising children alone (especially sons) was third. Social Implications The most frequently reported problem of widows is, consistently, loneliness. Lopata (1969) has conceptualized ten forms of loneliness based on her interviews with 300 widows in the Chicago area: (1) missing the particular other, with whom interaction is no longer available; (2) lacking the feeling of being loved; (3) lacking someone to care for, (4) longing for an in-depth relationship with another human being; (5) missing the presence of the other in the dwelling unit; (6) absence of the other to share work with; (7) homesickness for a certain style of life; (8) drop in status of the unescorted female; (9) strain in other relationships; and (10) inability to make new friends. Implicit in this schema is the assumption that widowhood brings with it a reduction in a wide variety of social interactions. More recently Lopata (1970, 1971, 1973; Lopata et al., 1970) has elaborated on widowhood as an event marking reduced social roles. Roles apt to be disorganized or severed with the husband's death involve his prior roles as sexual partner, father of the children, partner in couple-companionate leisure activities, contrib utor to management of the home in which she is housewife, and co-member in voluntary association groups. His death may remove her link to his relatives, the community, his work associates and mutual friends. Society is seen as derelict in training widows to establish new roles--to convert strangers into associates. The evidence that widowhood does indeed lead to social isolation is mixed, but most of it concurs. Lay books on the subject abound with the difficulties of finding satisfying social events which widows may enjoy. Interview data indicate that the widow does indeed see far less of her husband's relatives than before his death (Lopata, 1970; Marris, 1958). Marris1 (1958) study found the degree of contact with the widow's own relatives to be essentially unchanged after the death. Among the reasons he assembles to explain why contact with the family is not increased are the widow's desire to be independent, her resentment of pity or patronage, the social apathy of grief, the family's frustration at trying to console some one who refuses to be consoled, and the side effects of poverty, such as reduced funds for transportation and entertainment. A more detailed study by Adams (1968) com paring 263 middle-aged adults and their widowed or married mothers indicates that the widow's contact with her sons diminished, while visits with her daughters increased. Adams also looked at the type of interaction between widows and her children: daughters tended to increase almost all types of interaction including those of a ceremonial nature, home visits, social activities, and aid to the parent. For sons, however, the only increase was in the category of aid to the parent. Recent research reported by Berardo (1970) suggests that the frequency of extended family interaction 28 enjoyed by widowers is even less than that for widows. The major dissenting vote for the hypothesis that widowhood increases social isolation comes from a study by Lowenthal (1964). A large group of mental patients and a stratified random sample of San Francisco residents over age 60 were classified as isolates or interactors on the basis of their reports of the number of contacts with friends or relatives in a two-week period. There were proportionately more widows and widowers among the inter actors than among the isolates, both within the hospital and non-hospital groups. (Actual frequency data were not presented.) Inasmuch as all the evidence on degree of isolation in widowhood has been based on after-the-fact interview data, it may take a longitudinal method of time- sampled behavioral observations to clear up the confusion. Part of the importance of the hypothesis lies in its relevance to the relationship between mental illness and widowhood, which will be discussed in the subsequent section. Remarriage may be seen as a solution to the apparent isolation of widowhood. Second marriages in general are not described as less happy than first marriages (Bernard, 1956). Yet the widowed do not remarry as often as the divorced. Three fourths of all divorced persons remarry within five years; only one half of widowers and one fourth of widows remarry by that time. The likelihood of remar riage decreases more sharply with advancing years for widows than for widowers. There is a tendency for older widows to remarry widowers (Marris, 1958). (This choice of mate is popular with older women generally, but particu larly so with elderly widows.) Interestingly, in the remarriages of the widowed, religious preferences tend to be consistent, whereas remarriages of divorced persons tend to be religious intermarriages (Rosenthal, 1970). A number of studies have found that the majority of widows do not want to remarry (see particularly Cosneck, 1966 and Lopata, 1973). Marris (1958) and others have described the guilt associated with bereavement and subse quent feeling of loyalty to the deceased as an obstacle to remarriage. Marsden (1969) has poignantly pointed out that remarriage may be a financial risk for the widow who may lose her albeit meagre government benefits. We still do not know whether the widow's typical denial of interest in remarriage is an honest reflection of what she desires or a rationalization for the likelihood that she cannot get married, given the sex bias in the mortality rate, the 30 * C \ cK tendency for males to marry females younger than themselves, and her rusty skills at mate-finding. Physical and Mental Health That the physical and mental health of widows is impaired relative to their married sisters has been repeatedly demonstrated. The reasons for this state of affairs are less clear. Woolsey (1952) presents percent- of-population data showing higher rates of disabling illness among unmarried females (primarily widows) than among married females. The gap is large by age 30 and continues to be substantial through age 60. (Data after age 60 are not shown.) Widowed, divorced or separated women also have a greater number of days in disability per person and a higher frequency and duration of hospitalization than married women (Rosenfeld et al., 1952, 1957; California Department of Public Health, 1958). In their summary of health and age relationships, Confrey and Goldstein (1959) state, "It is clear that illness among unmarried women (widows, primarily) may represent a large part of the total problem of illness in later life." Two studies have documented health decrements in younger widows relative to matched, married controls 31 <Y\ i ri 1 (Maddison & Viola, 1968; Parkes & Brown, 1972). Higher levels of physical complaints were observed in both Bostonian and Australian widowed samples. Heyman and Gianturco (1972) did not find a health deterioration associated with bereavement in their longitudinal study of 41 elderly Ss who became widowed. Even more striking than the health decrement in young widows is the fact that widows die sooner than married women. The mortality rates for widowed persons from all causes for every age group for both sexes and for both whites and non-whites are higher than those for married persons (Kraus & Lilienfield, 1959). (The rates for single persons are slightly lower than widowed rates; rates for the divorced are even higher.) The relative excess mortality in unmarried groups is greater at younger ages. Persons aged 20 to 75 were considered; the highest death ratios were for the widowed aged 20-34. Persons in this group are more than twice as likely to die as married persons of the same age, sex and race. Kraus and Lilienfield also broke down the death rates in the 20-44 age groups by the twenty leading causes of death. The variance was large, but all ratios of deaths of widowed to deaths of married were greater than one. 32 Particularly noticeable were the consistently high ratios for arteriosclerotic heart disease and vascular lesions affecting the central nervous system. Some of the ratios for influenza and pneumonia (which may be contracted from an affected spouse) and suicide are also substantial. That suicides occur more frequently among the widowed than the married has been documented elsewhere (Cosneck, 1966; Segal, 1969). The causes of death in which widowed females outdo widowed males in spite of a lower overall mortality rate are arteriosclerotic heart disease and accidents other than by motor vehicle. The authors did an excellent job of indicating possible artifacts which may contribute to the high death rates in widowhood. Their analysis indicates, however, that a great deal of variance is left to be explained. Several hypotheses are offered. One is that for a variety of reasons persons with poor health risks tend to marry each other. Secondly, widowed individuals may have shared an unfavorable environment with the deceased spouse which contributed to his death. Finally, there is the hypothesis that widowhood itself has such deleterious consequences that an excessive death rate ensues. It is the writer's hope that this chapter on the stresses of widowhood will 33 /O' make that hypothesis seem plausible. In conjunction with the difficulties already described, widowhood may result in an altered diet and less stringent standards of personal hygiene. Frank (1963) has documented a phenomenon in other cultures of deaths being produced literally by group pres sure and individual decision to die. Perhaps the semblance between the pressures of widowhood in this society and the circumstances of such "inexplicable" deaths is too painful to contemplate. A relationship between mental illness and widowhood has also been observed (Beilin and Hardt, 1958). Higher suicide rates have already been mentioned. An increase in psychiatric symptoms such as anxiety, depression, insomnia, and tiredness was reported by Parkes (1964) in widows under age 65, but, again, older widows escaped this phenomenon. The high proportion of widowed in-patients in mental hospitals is known, but further analysis of mental illness and marital status is needed controlling for possibly con founding variables such as age and income level. A number of recent articles by Lowenthal and her co-workers in San Francisco have attempted to tease out the intervening social variables in the relationship between old age and mental illness. These studies have emanated 34 from the stress theory of mental illness. Simon (1970) found much stronger relationships between socioeconomic status (S.E.S.) or retirement and psychiatric impairment in an elderly hospitalized sample than that found for widowhood per se, but the methodology and presentation of data were inadequate. Blau (1957) has suggested that the stresses of retirement for men may be similar to the stresses of widow hood for women. The results of a second series of studies imply that it is unlikely that social isolation per se leads to hospitalization for mental illness in the aged (Lowenthal, 1964, 1965; Lowenthal & Boler, 1965). However, a study of the possible etiology of suicide in the elderly (Bock & Webber, 1972) should serve to keep us alert to the possi bility that the characteristic social isolation of widow hood leads to emotional disturbance. The study was correlational in nature and the authors were liberal in making causal interpretations. Despite this, their evi dence that elderly widowed with meaningful interpersonal relationships (as measured by membership in formal organizations and proximity of relatives) are less prone to commit suicide suggests that social interaction may serve a preventive mental health function for the widowed. An intriguing study of intimacy as a critical 35 variable in adaptation to old age is relevant to this hypothesis (Lowenthal and Haven, 1968). Two hundred and eighty San Franciscans over age 60 residing in the commu nity were evaluated in terms of number of social roles (spouse, parent, worker, organization member, or member of religious group--the roles of housewife, friend or neighbor discussed by Lopata were not included) and level of social interaction (rating by interviewer on an eight-point scale ranging from "no contact with others" to "contributes to social goals"). In addition, they were asked the ques tion, "Is there anyone in particular you confide in or talk to about yourself or your problems?" to determine the absence or presence of a "confidant." Three measures of adaptation were used. A morale test of eight questions pertaining to satisfaction with life, happiness, usefulness, mood and planning for the future allowed the division of subjects at the median into two groups--satisfied or depressed. Secondly, psychiatrists made ratings of degree of impairment from interviewers' reports--they did not actually see the patients. Finally, Ss were asked whether they considered themselves young or old for their age. The results indicated a relationship between both measures of social resources and all three adaptation 36 measures. (Unfortunately, this group of researchers characteristically presents frequency data only, and sta tistical analyses are omitted.) At follow-up one year later social losses were related to poor morale, but gains were not related to high morale. Maintaining one’s status quo with regard to number of social roles and level of interaction is related to all three measures of adaptation. The current presence of a confidant is positively associ ated with all three measures of adaptation. Absence of a confidant is related to low morale, though not to psychi atric rating or how old S feels. The great majority (70%) of those who lost their confidant during the year are depressed; whereas 68% of those who maintained one are satisfied. "Gaining one helps, but not much, suggesting again the importance of stability." The researchers hypothesized that the maintenance of an intimate relationship can serve as a buffer against the depression associated with decrements in social roles or interaction. This was confirmed. If you have a confidant, you can decrease your social interaction and run no greater risk of becoming depressed than if you had increased it. Further, if you have no confidant, you may increase your social activities and yet be far more likely to be depressed than the individual who has a confidant but has lowered his interaction level. Finally, if you have no confidant and retrench in your social life, the odds for depression become overwhelming. Similar but not as dramatic findings for number of social roles were evident. Particularly striking was the role of the confidant in mediating the morale of the widowed in the sample, as Table 2 demonstrates. A widow with a confidant has higher morale than a married person who lacks one. (The research ers admit that married Ss may have thought the confidant question pertained to someone other than the spouse. This might bias the data.) A similar "buffer effect" for the morale of retired persons in the sample was observed. A less hopeful alternative interpretation to the "buffer" hypothesis is that depressed persons are simply undesirable as friends and hence do not have confidants. The buffer effect was not observed for the sub sample with serious physical illness. A number of explana tions for the failure of the intimate relationship to ward off depression in the case of illness were described. To this reviewer, the most appealing answer is that the depen dence in prolonged illness (Clark, 1969) destroys the reciprocity in one's intimate relationships, which may be a crucial factor in morale. Table 2 Frequency of Depression Among 280 Aged, Widowed and Married Persons With and Without a Confidant* Percent Satisfied Percent Depressed Widowed within 7 years Has confidant 55 45 No confidant 27 73 Married Has confidant 65 35 No confidant 47 53 *Reproduced from Lowenthal and Haven (1968) . 39 Variation in Response to Widowhood In the previous section the major stresses of widowhood were discussed. Lest the reader assume that the entire experience of widowhood is negative, it should be noted that half of the 301 widows studied extensively by Lopata (1973) reported some compensations. The primary advantages of being widows as they saw it were enjoying living alone, having less work (e.g., housework) to do, and being independent. Although social scientists are developing an appreciation for the problems of the widow, we are far from being able to predict the adjustment pattern of a given widow to her situation. Hypotheses about potentially relevant subject characteristics are tentative at best, since so little data are available. Research subsequent to this dissertation will examine the relationships between a variety of demographic variables and measures of adjust ment to widowhood. Previous Social Class The widow's social class prior to her husband's death is likely to be an important determiner of her response to widowhood. Perhaps the most obvious hypothesis 40 is that the lower the social class, the fewer resources available, and the increased likelihood of poorer adjust ment to widowhood. Low income means fewer cars for trans porting widows to new role locations, less adequate health care, and less money for entertainment and social life. Increased education may facilitate assertiveness in enter ing new social roles (Lopata, 1972) and low education will restrict entry into an already discriminatory job market. However, the low S.E.S. widow is likely to be more experienced as a job-holder than her higher S.E.S. sister, which may be an advantage since both are likely to seek employment as widows if they are in the main working ages for women. And since in this society a woman's social status is largely determined by her husband, the previously high social class woman may experience a greater reduction in status at his death and concomitant greater disorganiza tion of life style than the woman already in a low social class. An examination of marriage role patterns for differ^ ent socioeconomic groups adds further complexity to the prediction of adjustment in widowhood. Rainwater (1965) classified marital role patterns in three types. In the "joint conjugal role relationship" activities are often 41 shared by both husband and wife or the same activity is carried out by either partner at different times inter changeably. Leisure as well as tasks are apt to be spent together. Sharing of activities is valued for the interest of each in what the other is doing, not just for the sake of efficiency. The husband and wife tend to relate as persons, instead of as "husband" and "wife." In the "segregated conjugal role relationship" the wife's and husband's activities are often very separate, and there is a formal division of labor. The "intermediate conjugal role relationship" lies in between the other two. Sharing is valued, but a formal division of labor is preserved. The husband and wife relate out of husband-wife role expectations. Characteristically, leisure time is spent watching television or reading, in which there is little talk about the activity, but the couple feels it reflects "togetherness." Rainwater interviewed 409 subjects including 152 married couples and on this basis assigned them to one of the three role patterns. Of the upper-middle class sub jects, 88% had a joint relationship, whereas 12% of the lower-lower class subjects had a segregated relationship. The majority of the upper-lower and lower-middle groups had intermediate conjugal role relationships. In conclusion, The lower-lower class husband, then, tends to be tangential to family functioning . . . the lower-lower class wife seems to find handling the family on her own to her liking, or at least consistent with what she has learned to expect from living in her particular social world. Her objections to her husband's separateness . . . are likely to have more to do with dissatisfac tion over money or socially demeaning behavior than over his failure to contribute interpersonal resources and skills to coping with the internal tasks of the family. One implication of such variations in marital role patterns is that lower class women may be more used to functioning independently of their husbands--at least in the realm of family management--than their middle and upper class sisters and consequently experience less change in life style as widows. Furthermore, Lopata (1969, 1970) has commented on the trend in urban middle and upper classes to replace sex-segregated groups with couple groups. With higher education, the wife is more likely to have multi-dimensional relationships with her husband, and to experience more complex feelings of loneliness as a widow. The more functional the husband-wife relationship, and the more dimensions on which the husband is involved in the wife's life, the more disorganized become her other social relations at his death. Safilios-Rothschild (1970), in a review of the 43 research on the family power structure, reports that lower class wives enjoy less decision-making power relative to their husbands than wives married to more highly educated and wealthy husbands. Despite methodological and concep tual limitations of this research, it may be that higher S.E.S. widows are more practiced at making decisions than their lower class counterparts, and in this regard can more easily cope with the independence of widowhood. Social class is one of the few variables for which there is some actual data about response to widowhood. Cosneck's (1966) dissertation on Jewish widows and widowers did not find a strong relationship between social class and adjustment as rated by the interviewer. Formerly high income widows are more apt to report loneliness as a major problem, whereas lower income widows are more apt to regard finances and health as their major problems (Nuckols, 1973). Lopata's (1973) in-depth interviewing of Chicago widows over age 50 permitted an analysis of their roles as wife, mother, kin member, friend, and participant in the community. She found more socially isolated women among those with the least education, whereas highly educated women tended to be active in a variety of roles. Both the widow's and her late husband's education and their financial 44 and occupational status heavily influenced the widow's involvement in her community, whether through neighboring, voluntary associations, or work. Current Employment Abrahams (1972) indicates that 39% of the 570 widowed persons who responded to a volunteer staffed hot line service in the Boston area ("Widowhood Service Line") were employed either full time or part time. This per centage is somewhat higher than for widows as a group, but the sample includes some widowers as well, so it is not possible to tell whether employed widows are more or less likely to need help. The callers' requests for help were categorized as either (1) lonely--wants a listener, (2) lonely--wants to meet people, or (3) requests specific information. Table 3 reports the type of help requested by employment status. Employed widowed persons more fre quently ask for help in meeting people than the unemployed. The author interprets this as an indication that employed callers may be even more lonely than the unemployed. Further, "the work situation does not provide the social intimacy necessary for a healthy recovery after the disrup tion of a way of living. Employment may even to some Table 3 Percentage of Widowed Callers Requesting Various Types of Help, by Employment Status* Type of Help Requested Employed Not Employed Full time Part time Lonely, wants a listener 19 14 21 Lonely, wants to meet people 50 34 31 Requests specific information 31 52 48 *Reproduced from Abrahams (1972). 46 extent deter the widowed from facing the realities of their social disengagement." This punitive interpretation seems unnecessary. Unemployed widows may be struggling with serious financial, child-care and unemployment burdens, and therefore they may be less likely than the already employed to ask about meeting new people. Duration of Widowhood There are data indicating that the more recent the death of the spouse, the greater the likelihood of reported loneliness (Townsend, 1957). The type of help requested by Widowhood Service Line callers of varying duration of widowhood (which is correlated with age) is shown in Table 4. Those calling within the first year of bereavement were most likely to need an understanding listener. Requests from two- to six-year widows were more often related to re-engagement. After seven years, there was an increase in requests for specific information, perhaps reflecting another critical period for the widowed when children leave home. Age A working hypothesis is that increasing age augments the difficulties of widowhood. After a look at widowhood Table 4 Percentage of Widowed Callers Requesting Various Types of Help, by Duration of Widowhood* Type of Help Requested Duration of Widowhood Less than 1 year 1-2 years 3-6 7 years or more years Lonely, wants a listener 42 12 16 16 Lonely, wants to meet people 25 52 51 42 Requests specific information 33 36 33 42 *Reproduced from Abrahams (1972). 48 in different societies Lopata (1972) indicated: The older widow is usually given much more freedom in daily action than is her younger counterpart, but her own circumstances may place heavy limitations on life styles. Remarriage, entrance into economic roles enabling self-sustenance, development of new relations, movement into more advantageous locations, and other role shifts are often made difficult by health problems, financial shortages, and societal inadequacies in services for the aged. The probability of remarriage does decrease with age (Bernard, 1956) and in the longer marriage period of the aged widow, the wife-husband relationship may have become more functional, implying greater disorganization in widow hood (Lopata, 1970). In spite of the conceptual support for the hypothe sis that aged widows are "worse off" than younger ones, it is the younger ones who incur the greatest physical and mental health deficits. Blau (1957) has advanced the hypothesis that the prevalence of widowhood in the sub culture has implications for adjustment. Specifically, the associates of older widows are more likely to be widows themselves, and a social group is more available. She studied men and women over 60 years of age; only among women over 70 were the large majority widowed, and in this age group widows had more extensive friendship ties than married women. Abrahams' (1972) data indicate that Widowhood Service Line callers aged 40-60 are most likely to have social problems (lonely--want to meet people). Both those under 40 and those above 60 were more likely to make specific requests for information— the older ones for handling financial, legal and health problems, and the younger ones for the problems of children and single parenthood. To control for duration of widowhood she selected 135 persons widowed over seven years, 21 under age 50 and 114 over 50. The younger group, even though widowed a long time, were more likely than the older ones to be seeking new relationships. Family Size Whether or not a widow still has children at home would seem to make a difference in her response to widow hood. In view of the emphasis on role loss the widow with children may be unique in experiencing an automatic role addition--that of father. This burden is increased by the level of children's needs at the time of the father's death. Financial problems are apt to be greater for larger families. A full 30% of the widowed callers responding to the Widowhood Service Line lived with children under age 16 50 (Abrahams, 1972). This is a large overrepresentation of widows with dependent children, and suggests that their needs may be greater than that of other widows. According to a lay advice book for widows (Owen, 1950) those with children have several advantages. They have someone to share their grief with, someone who will give direction and focus to their lives, and a living monument to the husband's memory. Children do seem to be an aid in facilitating social encounters between adults in urban neighborhoods. Abrahams (1972) concluded that "the widowed, while raising young children, need less help in making new friendships." Loneliness is more often reported as a major problem by those widows without children than by those with children at home (Nuckols, 1973). A widow's contact with adult children depends on the sex of the children, with daughters making more and varied visits than sons (Adams, 1968), perhaps because of the greater familial responsibility into which girls in our culture are socialized (Komarovsky, 1950). The size of the extended family may also have an impact on the experience of widowhood. According to Lopata (1972), "Generally, the smaller and 'simpler' the family unit, the more important each role and each person within 51 A it, and the more traumatic the experience of widowhood." Widows with large extended families seem to do better on two counts: widowhood is both less traumatic because less of one's life has been altered, and there are more resources for coping with it, making an easier transition to an altered life style. Living Arrangement Although living alone implies relative social isolation, most widows prefer living alone (Lopata, 1971a). This may be in part a rationalization but probably also reflects a realistic assessment of the difficulties of I 1 living with relatives who head the household. In particu lar, the important role of housewife (Lopata, 1966, 1971b) would be drastically reduced. Cosneck (1966) postulated that Jewish widowed living in a Hebrew home for the aged would be better adjusted than those outside the home (some of whom were peripherally associated with it) but she found little difference between the groups. Urban-Rural Residence Although in general the husband's occupation deter mines the social status of the wife and family, according to Parsons (1971) the status of a farm family is jointly 52 determined by both wife and husband. The urban widow is apt to experience a sudden break in the affiliation with her husband's employment, whereas a farm widow could con ceivably carry on the work with children and hired help. However, Smith (1962) found that few widows elect to stay on farms. Egelson and Egelson (1961) indicate that the social life of a widow in a small town is very difficult. How ever, one could argue that the increased mobility of possible associates in an urban environment would make social life there more unstable. Rural widows (in Washington and Missouri) are more likely to live alone than urban widows (in Chicago) (Lopata, 1970). Religion Because religions are concerned with the meaning of death and life after death, we can postulate that religious widows will have an external explanation for their circumstance and as a result experience a less traumatic period of mourning than others. The management of grief can be aided through institutionalized mourning rituals and the person of the minister, rabbi or priest. Liberman and Raskin (1971) raise the question of whether certain mourning rites may offer protection from depression whereas others may in fact reinforce depression. Cosneck (1966) found no suicides among a sample of 103 Jewish widowed persons over 60 years of age (period of study unspecified) and attrib uted this to the Jewish religion. Marital History It has already been pointed out that a long marriage may result in greater social role disorganization at widowhood than a briefer one. Bart (1968) compared women admitted to psychiatric institutions in Los Angeles and diagnosed as depressives with those receiving other diagnoses. Although 75% of the widows in her sample were depressed, women with two marriages were less likely to be depressed than those with only one, and women with more than two marriages were least likely to be diagnosed depressed. This suggests that the widow with multiple marriages behind her may be less inclined to respond to her situation with depression, although other forms of psycho pathology may develop. Services for the Widowed Silverman (1970) surveyed members of the National Funeral Directors Association to obtain information about 54 community-sponsored psychological and social services for the widowed. She also surveyed professional social service agencies and mental health clinics in a large urban area to determine the nature of professional resources available to the widow (Silverman 1966, 1967). The results of these surveys revealed that professional mental health resources, while plentiful, emphasize circumscribed prob lems, and are therefore not specifically oriented to groups making critical transitions. They are also greatly under utilized by the widowed in crisis. Furthermore, non professional resources for the widow are rare. At the present time, the major organizations offer ing help to the widowed reflect a self-help intervention model. Parents without Partners (PWP) is a voluntary dues- paying organization for widowed, divorced and separated adults with dependent children started in 1957 by Egelson and Egelson (1961). Lectures, discussions, a lending library, family outings and adult social activities are available to members, and there are chapters in many large cities. There are a variety of Catholic church groups for the widowed, such as NAIM in Chicago. A non-denominational but spiritually oriented organization primarily for the recently widowed called They Help Each Other Spiritually (THEOS) was founded in Pittsburgh by Decker (Decker & Kooiman, 1973). In New York the more recently formed Widows Con sultation Center directed by Horowitz employs a staff of psychiatrists, social workers, a financial adviser and a legal adviser, and has served over 2,000 widows. The Harvard Medical School Laboratory of Community Psychiatry developed a Widow-to-Widow program in a Boston community in which five widows contacted young, newly bereaved women and offered a listening ear, advice or direct help such as in locating a job (Silverman, 1969, 1972). Another part of the program was the Widowhood Service Line, a telephone crisis intervention program staffed by non-professional widows and widowers (Abrahams, 1972). Unfortunately, due to lack of funds the Boston program has since been dissolved. In San Diego Antoniak (1973) has developed a training program for volunteers who make home visits to recently widowed persons; a telephone hot line is also available. Similar volunteer outreach programs are devel oping in Phoenix, Houston, San Francisco, and Oakland. Although the Boston Widowhood Service Line had a built-in evaluative research component by requesting the 56 aides to record limited data after each call (Abraham, 1972), none of the programs described has systematically attempted to determine the outcome of the intervention on the recipient's life. The need for thorough program evaluation and outcome research in a field possibly about to burgeon with fledgling organizations is tremendous. Social scientists and mental health professionals concerned with the widowed have made a variety of recom mendations for future services and suggestions for allevi ating their distress. Silverman (1966, 1967) urges programs to vary their offerings depending on the stage of the widow's mourning ("impact," "recoil" or "recovery"). She also believes preventive mental health services will do best outside the auspices of social service agencies or mental health clinics. In 1970, Silverman outlined the advantages of using widows themselves as the primary caretakers in these interventions. Lopata (1973) has stressed the need for social interaction training among the elderly widowed, many of whom are prevented from re-engaging in new social roles because they lack skills which were not needed in former marital roles. 57 It is the conclusion of this analysis of the role of friend that the society could contribute to the welfare of its older widows by increasing the opportunity for them to become involved in primary relations. To facilitate such relationships, she recommends organizing activities of interest, arranging opportunities for contact with similar others (e.g., providing transportation) and psychologically convincing the widow she is able to make new friends. She further suggests a meal sharing program to combat the worst hours of loneliness, and advises that strain in the friendships a widow maintains after her husband1s death can be diminished by public information about the inevitability of this strain, and the reasons for it. Nuckols (1973) has recommended that life insurance sales persons provide general financial counsel to poten tial widows at the time of the sale and subsequently, during review of the client's needs. He has underscored the need for more sensitive training of the insurance agent with regard to claim settlements. In a broader vein, Bernard (1956) has suggested that attitudes conducive to romance in old age be fostered to increase the possibility of remarriage among the widowed. Cosneck's (1966) review of suggested solutions includes a 58 recommendation to advise young women to marry men younger than themselves, and to educate them for self-reliance. A few investigators have urged legislative reform of the OASDHI laws to include all bereaved persons in financial distress, to increase the level of aid, and to reduce the penalties for employment. In considering the alternative of job training for the widowed, Palmore et al. (1963) have made the absurd rationalization that increased pensions "might reduce the prospects for the family to share through earnings in the economic growth of the country." The problem of inadequate day care facilities for the children of working mothers including the widowed is a large one. Harcourb (1969) has boldly recommended a campaign to readjust working hours to fit the schedules of widows with children. The government itself could model such employment practices. She also suggests training widows in home repairs and providing advice on legal and financial matters. Finally, there are the recommendations made by widows themselves: their most frequent advice to the potential widow was to be active (Nuckols, 1973), 6% encour aged her to "have faith"; 4% wanted someone to tell her 59 > friends and relatives not to forget her; a full 4% spon taneously suggested a special counseling service, and the same number urged preparation for widowhood. A substantial percentage recommended expansion of social security benefits. 60 CHAPTER III THE RESEARCH METHODOLOGY The Design of the Study As the previous chapter emphatically demonstrates, widowed persons experience multiple problems. The most plausible unitary explanation for this predicament is that widowhood itself is a major stress, affecting a substantial portion of the total population. We did not yet know how to predict a widow's response to this stress, although several hypotheses have been enumerated. Institutionalized resources for the widowed are rare, and no outcome research on the available services has been attempted. The purpose of this dissertation was to develop a model psychological and social program for widowed women of all ages and duration of widowhood, in order to compare the efficacy of three different group interventions. Thus the research addressed itself to the question of how to best facilitate the successful negotiation of the stresses a widow encounters. Secondly, the research provided informa tion as to whether the outcome of the therapeutic 61 approach varies with widows of different characteristics. To respond to these issues, a quasi-experimental research design with a waiting list control group and a pretest, posttest and three and one-half month follow-up was implemented. Urban widows participated in a seven-week group experience with one of two leaders, each of whom was responsible for three types of intervention: (1) a self- help group, (2) a confidant group, and (3) a women's consciousness-raising group. The waiting list control group began meeting at the collection of the posttest data; hence, their data at follow-up would not provide a useful control and were not evaluated. Since the study was exploratory in nature, few a priori predictions were made. However, a wide array of demographic and dependent variables were available for the assessment of responses to the therapeutic interventions. Primary variables included the widow's age, duration of widowhood, cause of husband's death, degree of preparation for widowhood, quality of general psychological functioning, specific reactions to widowhood, changes in life style as a result of the program and attitudes toward women. Subjects were further asked to assess the program itself, 62 describe positive and negative life changes occurring during the program, and report the amount of contact they had with other group members both during the treatment and follow-up periods. The Subjects Subjects for the study were obtained by a news release (see Appendix A), shortened versions of which appeared in the Los Angeles Times and several local commu nity newspapers including The Valley News and Green Sheet (Van Nuys, circulation approximately 270,000), The Post Advocate (Alhambra), The Burbank Daily Review, The Evening Outlook (Santa Monica), The Temple City Times, and the Brentwood Westwood Hills Press. The news items, published in August or September of 1973, asked widows of any age or length of widowhood to volunteer, and requested potential participants to send a card or letter to the investigator stating their name, age, length of widowhood, address, and telephone number. Approximately 320 widows responded from all parts of Los Angeles County. Approximately 234 of the first widows who wrote in were sent a letter inviting them to one of four orientation meetings (see Appendix B). Five additional Ss were 63 contacted later by telephone to fill a group scheduled at an unpopular time. The remainder were sent a letter explaining that the program had been filled, and referring those who wanted professional help to the Los Angeles County Department of Mental Health for the nearest commu nity mental health center. Altogether, approximately 121 women attended one of the two-hour orientation meetings. (Those who missed one of the first three meetings were given an opportunity to attend a later orientation session.) At these meetings, Ss were informed of the nature of the program, and the three types of group format, given a brief lecture on known facts related to widowhood, and introduced to the group leaders. They completed the pretest dependent measures using a subject code number, signed a general consent form (see Appendix C) and indicated the times they would be available for discussion groups. Although Ss knew they would be attending one of three types of discussion groups, focusing on "the development of friendships" (the confidant groups), "specific problems of widowhood" (the self-help groups), or "the roles of women in society" (the consciousness-raising groups), they did not know at which times these particular groups would meet. A $15.00 check 64 for a "registration deposit," payable at the first group session and to be refunded in full at the follow-up session, was required to reduce the incidence of dropouts. Although the preferred procedure for assignment of Ss to treatment group would have been random assignment, several factors contributed to a departure from the ideal. The first was a constraint due to availability of subjects during pre-designated group meeting times. The second was the advantage of pairing subjects from the confidant treat ment with persons who lived in the same general vicinity, to increase the likelihood of continued relationships after the end of the program. The third was the decision to begin the groups as soon as possible after the first letters had been written, to avoid subject attrition due to delay; this meant the first subjects had to be assigned before the availability of others (or even the exact number of potential Ss) was known. To ensure geographic proximity among confidant group participants, these group assignments were made from among the two most represented regions of the five Los Angeles County regions partitioned by the telephone company as toll-free telephone call areas. These were the North western Region (the San Fernando Valley) and the Los 65 Angeles Central Region. To prevent confounding of treatment group with employment or other subject variables influencing avail ability, one group of each of the three types of interven tion met on a weekday afternoon, and one of each type met on Saturdays. (The waiting list control group was sched uled at the most popular meeting time, which was Saturday morning.) The general procedure for assignment of Ss to group was made as random as possible with the following stipula tions. Approximately 14 Ss were originally assigned to each treatment group, and 28 to the control group. All Ss who attended the orientation were assigned to a group, except 4 who withdrew after the meeting. Ss from the first orientation meetings were assigned to the first group leader or the control group; Ss from the final orientation meetings were assigned to the second leader or the control group. Ss in the first leader's confidant group all lived in the San Fernando Valley region. The second confidant group was composed of widows living primarily in the central Los Angeles region. No subject was assigned to a group she had indicated at the orientation it would be impossible for her to attend; however, once assigned to a 66 group some Ss were unable to attend and their group assign ments were changed. Those subjects who for reasons such as vacations or relocation of homes could only attend the waiting list control group were permitted to attend this group, but were excluded from the data analysis because of possible sample bias. (Only 4 of these 8 Ss actually attended a widows' discussion group.) The occasional subjects who requested to be in the same group with a friend were assigned together to facilitate transportation. All of the Ss assigned to treatment groups began their small group sessions within one to two weeks of the orientation meeting. The waiting list controls were sent the letter which appears in Appendix D, and were given a seven-week group discussion experience following the collection of posttest data during the follow-up period. Twenty-five of the Ss assigned to a discussion group never attended; 10 control Ss dropped out prior to collection of posttest data. A minimum attendance of four small group sessions including the final data collection session was the criterion for inclusion of Ss in the data analysis; this resulted in the elimination of 8 Ss, 4 from confidant groups, 2 from self-help groups, and 2 from consciousness-raising groups. One divorced subject 67 assigned to the control group was also eliminated. This left the following subjects for inclusion in the data analysis: Self-help group 1 8 Self-help group 2 10 Confidant group 1 8 Confidant group 2 7 Consciousness-raising group 1 11 Consciousness-raising group 2 9 Waiting list control group 17 Total 70 The Group Interventions Three different group interventions were developed by the investigator and compared to a waiting list control group. All treatment groups met for two hours once a week for seven weeks following the orientation meeting. Post test data were collected in the last half of the final group meeting. The final data forms were completed in a meeting of all subjects held three and one-half months after the final sessions of the small group discussions. (The period varied from 13 to 15 weeks.) At this time the regis tration deposits were returned, and participants were given feedback about the program based on a preliminary data analysis. They also had an opportunity to discuss the possibility of developing an ongoing organization for widows in the Los Angeles area. All of the research meetings and group sessions were held on the University of Southern California campus in one of two locations--the Ethel Percy Andrus Gerontology Center or Kerchkoff Hall. There were two group leaders for the research project; one was the investigator and the other was a female doctoral student in clinical psychology and aging. (Neither was herself a widow.) The leaders met each week during the treatment phase of the program to review the strategies in each group intervention and discuss any difficulties encountered in operationalizing them. The self-help group modality incorporated many elements of self-help therapy groups organized around other problem areas (Hurvitz, 1970). An assumption of this approach, stated directly to the members, was that the widowed are the experts on widowhood. The purpose of the group was to facilitate a discussion of the problems of widowhood and encourage members to help each other toward solutions. The problems discussed were those initiated by the members themselves, and included loneliness, grief, 69 single parenting, reduced financial resources and employ ment difficulties, decisions about living arrangements, strained relationships with relatives and married friends, new relationships with men, and legal problems. The role of the leader was that of discussion facilitator--direct requests for advice were referred back to other group members, and reticent members were encouraged to share their experiences. Helping statements made by one member to another were reinforced by the leader's praise. The idea for the "confidant group" originated from the research of Lowenthal and Haven (1968), demonstrating that the presence of a confidant is predictive of subse quent mental health in an elderly urban population. The purpose of these groups was to facilitate the development of an intimate friendship between pairs of widowed women who could use their relationship as a buttress against the stresses of widowhood. The recent emphasis on peer coun seling techniques in psychotherapy enhanced the rationale for this intervention strategy. Subjects in these groups were paired with another widow in one of two ways. At the end of the first group session (following a general discussion used in all treat ment groups as to whether widows or widowers are exposed to the most stress), Ss could confidentially indicate their preferences for pair assignments. Those who did not elect this option were assigned primarily on the basis of matched age and duration of widowhood. The subsequent irregular attendance of some of the confidant Ss meant that the pairs were not as stable a unit as was planned. The format of these groups consisted of an intimacy training task in pairs, during which the leader moved about the room helping where needed, followed by a whole group discussion of the experience. The interpersonal tasks began with finding three things the couple had in common, and proceeded gradually over the seven-week period to the sharing of personal problems as a widow and offering help to one's confidant. A planned higher-order intimacy task was the sharing of telephone numbers and planning of a joint activity outside the group. However, by mid treatment many of the groups had spontaneously shared telephone numbers. The assigned intimacy tasks were facilitated by group activities such as an informal birthday party for one of the members and a walk through a nearby garden. The third group intervention was a women's consciousness-raising (C-R) group for widows, based on the 71 structure developed for C-R groups by the C-R Committee of the Los Angeles chapter of the National Organization for Women (Freeman, 1973). Each group was given a list of possible topics reproduced in Appendix E at the first session; topics for subsequent sessions were agreed on by the members a week in advance. The format of each session iwas: (1) introductory comments about the topic from the leader's own experience, (2) 5 to 10 minutes for each subject in sequence around the group to express her reac tions to the topic without interruption, and (3) open discussion by all participants. An enforced rule of the C-R groups was "no confrontation," so that a trusting climate for exploration of new ideas could be created. Following each member's individual reactions, others could ask questions, but they were not permitted to make comments until their "turn." Subject and Dependent Variables A variety of kinds of data were obtained from the Ss using a pencil-and-paper, self-report format. These included subject demographic variables, dependent measures reflecting general psychological functioning, specific reactions to widowhood and life style indices, attitudes 72 toward women, several behavioral measures, and evaluations of the group interventions per se. A check on the manipu lation of the group interventions was also undertaken. A multitude of information was available for each S. Only those variables reported and discussed in this dissertation will be described here; other measures which appear in the data forms in the Appendices will be analyzed in subsequent research. Subject Variables The demographic variables were obtained at the pretest using the General Information Form (see Appendix F). Ss reported their age, duration of widowhood, number of prior marriages, ethnic identity, religion, education completed, and current employment status. The total current monthly income from all sources and the total household income prior to widowhood were indicated by selecting one of eight income brackets (see questions 59 and 60, Appendix F). Data were obtained on type of dwell ing unit (question 10) as well as living arrangement (question 11). The question, "Did your husband pass away: (a) very suddenly, (b) somewhat unexpectedly, or (c) after a long illness" yielded a three-level index of the variable cause of death. 73 I V The yes-or-no answers to four questions (number 42 to number 45) contributed to a summed score for degree of preparation for widowhood. These questions tapped the widow's thoughts about widowhood prior to the event, her expectation of widowhood, her ideas about what to do if widowed, and the presence of specific plans for widowhood such as a will. Lowenthal and Haven's (1968) exact wording for determination of the presence or absence of a confidant was employed (question 30). In addition, Ss were asked whether they had ever been hospitalized for psychiatric reasons, and whether they owned or had access to a car. Psychological Functioning To assess general psychological functioning, three standardized psychological instruments were administered at pretest, at posttest, and at follow-up. These were Rotter's (1966) I-E scale to measure internal versus external locus of control, Rosenberg's (1962) self-esteem instrument, and Neugarten et al's (1961) Life Satisfaction Index A. 74 Reactions to Widowhood Three measures of Ss1 reactions to widowhood were developed, and administered at all three data collection points in the study. The first of these, degree of physi cal complaints, tapped the frequency of six physical symptoms commonly experienced in conjunction with grief. On a three-point scale from "hardly ever" to "frequently," Ss reported the frequency of headaches, digestive distur bances, aching muscles or limbs, inability to sleep well, oversleeping, and poor appetite. Their answers were summed to produce a single score for frequency of physical symptoms. The second variable was intensity of grief, and consisted of the sum of twelve ratings made on a seven- point scale. The widow assessed her current degree of loneliness, desire for help, feelings of crisis, guilt, anger at her husband, at friends or family, at God, or at her husband's physician, ability to cope, frequency of being overcome with grief and of avoiding thinking about her husband, and degree of depression. (See Appendix G, items 9, 10, 11, 13, 14, 15, 16, 17, 18, 20, 22, and 23.) The third variable in this group measured attitude toward widowhood on a seven-point scale from "Being a widow is a terrible experience" to "Being a widow is a wonderful opportunity" (Appendix G, item 24). A fourth measure for degree of stress in widowhood was obtained only at posttest, when Ss were asked to recall the stress experienced at the start of the program, and at follow-up. It consisted of a sum of 9 seven-point ratings for the stresses of loneliness, finances, unresolved legal issues, family difficulties, health problems, work related stress, and two additional personal stresses which S iden tified. (See Appendices H and J, items 1 and 3 through 9.) Life Style Variables Three measures were devised to assess changes in the widow's actual or anticipated style of living at pre test, posttest and at follow-up. The first of these asked her to predict whether her health in the next five years would be: (a) excellent, (b) good, (c) fair, or (d) poor (Appendix G, item 3). Her attitude toward remarriage was elicited (Appendix G, item 31). Finally, Ss rated their degree of involvement in eight major social roles: mother, grandmother, daughter, friend, date, employee, volunteer, and religious person. Scores on the seven-point scale were added to yield a sum of social role engagement. 76 Attitudes toward Women The factor-analyzed questionnaire reported by Gump (1972) was modified slightly and used to assess atti tudes toward women at all three points in time. (See Appendix I, items 1 through 23.) The questionnaire yields two scores which are the sums of several independent factors. "Other orientation" is a combination of attitudi- nal factors reflecting a home-oriented duty to children and the views that women's role is submissive and that one's identity is derived through traditional roles. "Self orientation" includes factors of need for individual achievement and sense of autonomy, and reflects the views that the traditional role for women implies some relin quishing of needs and that the family is inadequate to fulfull one's needs completely. In addition to these two measures, 13 items were selected from Spence and Helmreich's (1972) "Attitudes Toward Women" scale which seemed to the investigator both to represent content areas not included in the other scale and to be particularly germane to the experience of widowed women. These questions appear in Appendix I as items 24 to 36. The four-point answer code provides a summed score for degree of radical versus conservative attitudes. 77 Behavioral Measures Two behavioral dependent measures were obtained at posttest and at follow-up, and two others at follow-up only. Subjects' responses to item 11, Appendix H, and to item 15, Appendix J, revealed the number of group members they had contacted outside of the group during the treat ment or follow-up period, and the total number of individ ual contacts. The posttest scores refer only to group members not previously known to S; the follow-up scores include contacts with all group members. The number of meetings of each widows’ group as a whole during the follow-up period was also obtained. The final behavioral measure was an open-ended question which asked S to "describe the ways in which you or your life have changed since the start of the widowhood research project. Please include changes of all kinds, positive or negative, whether they seem related to the program or not." Responses to this question were rated by an experienced psychotherapist with a D.S.W. degree, who was blind as to treatment condition, on a five-point scale from negative to positive. 78 i. <V> Group Evaluation The experimental subjects made ratings of how helpful their widows' discussion group was (Appendix H, item 18) and how much they learned from it (Appendix H, item 21). These measures of helpfulness and educational value were obtained at posttest and again at follow-up. Members of the confidant groups only completed an additional evaluation form at posttest and follow-up (see Appendix K). Seven-point scales were utilized in opera tionalizing the confidants' knowledge of self, liking for the confidant, ease of discussing personal problems, nature of the current relationship, and anticipated relationship in three months. The final group evaluation measures consisted of the experimental Ss' posttest recommendations for the composition of future widows' groups. (See Appendix H, item 29.) Three variables were assessed: uniform versus diverse age group preference, uniform versus diverse duration of widowhood preference, and sex-integrated versus sex-segregated group preference. Manipulation Check At posttest two questions were asked of the experi mental Ss to determine whether the group interventions had 79 been successfully operationalized as planned (Appendix H, items 15 and 16). They selected the role description which best described their leader and that which best described their own role as participants. The "correct" responses for the leader's role were "helping us to help each other with problems" for self-help Ss, "helping us to get better acquainted" for confidant Ss, and "direct sharing of ideas" for C-R Ss. The anticipated member roles were "giving and receiving help" for self-help Ss, "developing friendships" for confidant Ss, and "sharing my experiences as a woman" for C-R Ss. Method of Data Analysis Data were available on 70 Ss at pretest and at posttest, and on 51 Ss at follow-up. (Two experimental Ss did not attend the follow-up session, and the control group data were excluded since these Ss had themselves received treatment prior to follow-up.) The basic statistical design for the research was a two-way, leader (2 levels) by treatment (3 levels), multivariate analysis of variance. By ignoring the leader factor, a one-way multivariate analysis could be performed on the three experimental manipulations and the control group. Change scores were utilized wherever pretest levels of dependent variables were available. It was possible to ascertain whether the mean change from pre- to posttest or from pretest to follow-up was significantly different from zero across all subjects by testing the "error effect." By using planned comparisons, the three experimental condi tions could be lumped together and compared to the control group. Another planned comparison examined differences between the treatments themselves. Four of the subject variables were chosen as covariates in the major analyses of variance; these were age, duration of widowhood, cause of death and degree of preparation for widowhood. It was postulated by the inves tigator that these variables might be important predictors of the response to the various group interventions; however no specific directional hypotheses were formulated. All analyses of variance and analyses of covariance were computed using the MANOVA computer program (Cramer, 1970). Chi-square analyses on the qualitative variables and calculations of sample means, medians, and standard deviations for the demographic variables were computed using programs in the Statistical Package for the Social Sciences (Nie, Bent and Hull, 1970). 81 § \ Few a priori predictions were made with regard to differential outcomes in the three group interventions developed, since the study was conceived as an exploratory trial of a new program for widows, and each of the three interventions had an independent rationale. It was predicted that self-help Ss would demonstrate the greatest reduction in specific stresses of widowhood, that confidant Ss would achieve the highest levels of contact with new friends from their groups, and that C-R Ss would become more radical and self-oriented in their attitudes toward women. General predictions were made with regard to post test differences between those Ss who were assigned to a widows' discussion group and the waiting list controls. It was hypothesized that changes in psychological function ing and reactions to widowhood would favor the experimental group. Because of the likelihood of correlation among these variables and the life style variables (e.g., health prediction may be related to current physical symptoms), multivariate analyses of variance were performed on the combined measures comprising psychological functioning, reactions to widowhood and life style. 82 CHAPTER IV PRESENTATION OF RESULTS Sample Description The sample ranged in age from 32 to 74, with a mean age of 55.7, median age of 56.8, and a standard deviation of 9.2 years. Subjects aged 65 and over comprised 14.3% of the sample. The age distribution appears in Table 5. The duration of widowhood among the Ss ranged from one month to 22 years, with a mean duration of 4 years 9 months, and a median of 3 years 9 months. The standard deviation was 4 years 7 months. Subjects widowed for 2 years or less comprised 32.97, of the sample. One out of seven subjects had been a widow 10 years or more. Table 6 describes the distribution of this variable. Most subjects (87.1%) had only been married once. Nine subjects reported two marriages. Only three subjects were ethnic minority members--two Chicano women and one Black. Of the sample, 40% were Jewish. Table 7 describes the religious affiliation of the sample. 83 Table 5 Age Distribution of the Sample Age Range Number of Subjects 32-39 4 40-49 11 50-59 27 60-69 25 70-74 3 70 Table 6 Duration of Widowhood in the Sample Duration of Widowhood Number of Subjects 1 year or less 18 1 to 2 years 5 2 to 3 years 7 3 to 4 years 9 4 to 5 years 7 6 to 10 years 14 Over 10 years , 10 70 84 Table 7 Religious Affiliation of the Sample Religion Number of Subjects Jewish 28 Protestant 23 Catholic 10 Other 9 70 85 sV \ The educational level of the subjects ranged all the way from grade school (2.9%) to graduate and profes sional degrees (15.7%). Women with at least some college education comprised 71.4% of the sample. Table 8 describes this variable. More than half of the sample were employed either full time or part time (52.9%). An additional 17.1% wanted jobs but were not able to find employment. Table 9 describes the employment pattern among subjects. The current modal income bracket among the subjects from all sources of income, including wages, survivors' benefits, social security, life insurance, retirement annuities, private pensions, and investments was $600 to $900 a month. The next most frequent income bracket was $450 to $600 a month. By contrast, the most frequently reported total household income of the marital unit prior to widowhood was "more than $1,500 a month," followed by $1,200 to $1,500 a month. Table 10 compares the subjects' current and marital incomes. More than half of the Ss owned their own homes (54.3%). Most of the rest rented apartments (38.6%). Five subjects lived in other types of dwellings. More than one half of the Ss lived alone (57.1%). Most of the others 86 'I X Table 8 Education of the Sample School Year Completed Number of Subjects Grade school 2 Some high school 2 High school 16 Some college 32 College 4 Some graduate school 3 Graduate or professional school 11 70 87 > » Table 9 Employment Among the Subjects Work Status Number of Subjects Full time 21 Part time 16 Unemployed 12 Not employed by choice 11 Retired 10 70 88 Table 10 Subjects' Total Household Income Prior to and At Present Widowhood Number of Subjects Prior to Widowhood At Present Less than $150 1 0 $ 150-$ 300 2 8 300- 450 1 9 450- 600 7 15 600- 900 8 23 900- 1,200 12 4 1,200- 1,500 17 5 More than $1,500 22 6 Total 70 70 89 lived with children still at home (34.3%). Six subjects had other types of living arrangements. Almost one half of the Ss reported that their husbands had died very suddenly (34, or 48.6%). The husbands of 20 subjects died of long-term illnesses (28.6%). The remainder of the Ss indicated their husbands died "somewhat unexpectedly" (16, or 22.97o). Exactly one half of the subjects reported having thought about the possibility of widowhood prior to the husband's death; the other one half had never thought about it. Similarly, about one half of the subjects indicated they had expected to be widows prior to the event (34, or 48.67o). Forty-one or 58.6% had some plan for widowhood, such as a will or knowledge of husband's financial affairs; 29, or 41.47. had no such plans. Only 16 subjects (22.9%) reported having any ideas about what they would do if widowed. Most subjects at the pretest had at least one confidant, that is, they had someone with whom they could discuss personal problems (57, or 81.47o). However, 13 subjects (18.6%) did not have a confidant. Six subjects had previously been hospitalized for psychiatric reasons. All but 8 Ss had access to an automobile. Pretest Analysis Multivariate and univariate analyses of variance on the dependent variables obtained at the pretest (three attitudes toward women variables, three general psychologi cal functioning variables, the three available specific reactions to widowhood variables, and three life style variables) resulted in no significant differences between the four treatment groups (self-help, confidant, conscious ness-raising, and waiting list control groups). A multivariate analysis of variance on the four subject variables selected for covariates in subsequent analyses (age, duration of widowhood, cause of death, and extent of preparation for widowhood) did result in a significant main effect for treatment. As the univariate tests on these variables shown in Table 11 demonstrate, this result was largely dependent on a significant treat ment group difference in the cause of death variable. The group means indicate that husbands of more Ss assigned to self-help groups and consciousness-raising groups had died very suddenly, whereas the control and confidant groups included more women whose husbands had died of long-term illnesses. 91 o \ 5 ( 0 Table 11 Analyses of Variance on Age, Duration of Widowhood, Cause of Death, and Preparation for Widowhood Treatment Age Duration Cause of Death Preparation Self-help (N=18) Mean 54.2 76.4 1.4 2.0 S.D. 11.4 63.4 0.7 1.3 Confidant (N=15) Mean 56.7 55.1 2.3 1.9 S.D. 8.0 33.7 0.6 1.4 C-R (N=20) Mean 55.4 43.5 1.5 1.9 S.D. 9.8 60.1 0.8 1.2 Control (N=17) Mean 56.8 57.0 2.2 1.4 S.D. 6.8 56.5 1.0 1.3 F df P Multivariate (N=70) 2.1 12,167 .018 Age 0.3 3,66 NS Duration 1.1 3,66 NS Cause of death 5.7 3,66 .002 Preparation 0.7 3,66 NS 92 The univariate tests of treatment differences for age, duration of widowhood, and extent of preparation for widowhood were not significant. Although the mean duration of widowhood for the four treatments was highly variable, ranging from a low of 3 years 6 months in the C-R groups to a high of 6 years 4 months in the self-help groups, the large variance within treatments prevented significance. Manipulation Check A chi-square analysis on the perceived role of the group leader at posttest by treatment group was almost significant, as shown in Table 12. The subjects' percep tions of their leader's role were more likely to be congruent with the specific role attempted in each treat ment than they were likely to resemble a role attempted in another treatment. An exception to this pattern occurred in the C-R treatment, with more Ss choosing the leader's role as that of "helping us to help each other" than the treatment-designated role of "direct sharing of ideas." A chi-square analysis on the members' perceptions of their own roles in the group by treatment (the second manipulation check) was not significant. More Ss in all three treatments saw their role as "sharing my experiences Table 12 Chi-Square Analysis of Perceived Role of Group Leader by Treatment Treatment Self-Help (N=18) Confidant (N=15) C-R (N=20) N=53 Chi-square = 11.67, with 6 df significance level = .069 Perceived Leader s Role Other or Self-Help Confidant C-R Missing Data 7 2 5 4 4 6 2 3 10 0 6 4 21 8 13 11 94 H-tt as a woman" (the C-R role) than either of the other two roles the leaders attempted to engender in the self-help and confidant treatments. Analysis of Change in Psychological Functioning, Reactions to Widowhood and Life Style A two-way multivariate analysis of variance on the change scores from pre- to posttest for nine major depen dent variables (three psychological functioning variables, three reactions to widowhood variables, and three life style variables) was performed on the 53 experimental Ss only with treatment (3 levels) and leader (2 levels) as independent factors. There was no significant main effect for leader; nor was there a significant leader-treatment interaction. In subsequent univariate analyses on these nine variables and on the variable of extent of stress at pretest recalled at posttest, only one significant effect for leader emerged: the change in Ss' attitude toward remarriage varied as a function of the group leader (see Table 13). No significant leader-treatment interactions occurred on any of these univariate analyses. A one-way multivariate analysis of variance on the pretest change scores of the same nine dependent variables 95 Table 13 Change in Attitude Toward Remarriage at Posttest as a Function of Group Leader Treatment Mean Change S.D. Leader 1 Self-help -0.25 1.67 Confidant 0.13 1.55 C-R 0.00 0.92 Leader 2 Self-help - .67 1.32 Confidant -2.17 2.56 C-R -1.00 1.16 Two-way analysis of variance Test for leader: F = 6.44 with (1, 40) df, p < .015 Test for treatment: NS Test of leader-treatment interaction: NS 96 was then undertaken on all Ss including the waiting list Ss, but eliminating the leader factor. By partitioning the degrees of freedom in the treatment factor, planned comparisons were used to assess (1) whether any non-zero change in the variables occurred from pre- to posttest across all subjects, (2) whether the change scores among the three treatments as a whole differed from change scores in the waiting list control group (Contrast 1), and (3) whether there were any between-treatment differences in the pre-post change scores (Contrast 2). The dependent variables included in this analysis were: locus of control, life satisfaction, self-esteem, attitude toward widowhood, intensity of grief, physical symptoms, five-year health prediction, attitude toward remarriage, and social role engagement. As indicated in Tables 14 and 15, this multivariate analysis resulted in a significant non-zero change across Ss (the error effect); neither Contrast 1 nor 2 was significant. Subsequent univariate analyses revealed significant error effects for increased self-esteem, increased intensity of grief, and more negative attitudes toward remarriage. The only variable among this group of nine on which the combined treatment groups differed from 97 i Table 14 Mean Change in Self-Esteem, Grief, Health Prediction, and Attitude toward Remarriage at Posttest Variable Treatment N Mean S.D. Self-esteem Self-help 17* 1.3 5.6 Confidant 14* 2.8 5.0 C-R 15* 2.5 6.1 Control 17 3.8 5.2 Intensity of grief Self-help 1.7 9.9 Confidant 5.8 9.4 C-R 6.5 14.4 Control 4.6 10.8 Health prediction Self-help - .06 .56 Conf idant .14 .54 C-R .20 .41 Control - .24 .66 Attitude toward remarriage Self-help - .47 1.5 Confidant - .86 2.3 C-R - .47 1.1 Control - .65 1.7 *Total N = 63 for each variable. Seven Ss omitted due to missing data on one or more variables: 5 from C-R, 1 from self-help, and 1 from confidant. 98 £ Table 15 Analysis of Variance on Change in Psychological Functioning, Reactions to Widowhood and Life Style at Posttest Effect F df P Multivariate on 9 variables* Non-zero change 3.09 9,51 .005 Locus of control No significant effects Life satisfaction No significant effects Self-esteem Non-zero change 14.07 1,59 .001 Attitude toward widowhood No significant effects Intensity of grief Non-zero change 9.92 1,59 .003 Physical symptoms No significant effects Health prediction Treatment vs. control 4.12 1,59 .045 Attitude toward remarriage Non-zero change 8.07 1,59 .006 Social role engagement No significant effects *Total N = 63. Seven Ss omitted due to missing data on one or more variables: 5 from C-R, 1 from self-help, and 1 from confidant. 99 the control group was the five-year health prediction, with experimental Ss evidencing a more positive change in evaluation of future health. Univariate tests of between- treatment differences were not significant. The preceding multivariate and univariate analyses of variance on the pre-post change scores using planned comparisons were then repeated, covarying one at a time for the variables of age, duration of widowhood, cause of husband's death and preparation for widowhood, to determine the impact of these subject variables on the response to treatment. The results of these analyses of variance with covariance are as follows: 1. None of the covariates itself had a significant regression effect in either multivariate or univariate tests. 2. Analyses of covariance did not result in any sub stantial change in the test of Contrast 1; that is, the significant treatment versus control group differential change in health prediction was main tained with all four covariates, and no new signifi cant differences emerged. 3. Analysis of covariance did not alter significance patterns for Contrast 2; that is, even with 100 covariate analyses, there were no between treatment differences at posttest on these nine variables. 4. Analyses of covariance did substantially influence the pattern of significant error effects (non-zero change across all subjects). Specifically, the highly significant outcome of the multivariate test of the error effect disappeared when covarying for age, for duration of widowhood, and for cause of death, and was much less significant when covarying for widowhood preparation (p < .031). Similarly, significant univariate error effects for self-esteem, intensity of grief, and attitude toward remarriage either disappeared, or were somewhat less significant. No new significant error effects emerged, although a trend toward increased external control on the I-E scale appeared covarying for cause of death (F = 3.57, 1 and 58 df, p < .064) and widowhood preparation (F = 3.14, 1 and 58 df, p < .082). A trend toward increased social role engagement also emerged when covarying for widowhood preparation (F = 3.46, 1 and 58 df, p < .068). A similar one-way analysis of variance and analyses of covariance with planned comparisons were performed on the variable extent of stress at pretest recalled at posttest. Neither treatment contrast was significant, even when considering the variance due to each of the four covariates. However, there was a significant regression effect for age (F = 5.4, 1 and 65 df, p < .023). The regression coefficient for age was -0.26, indicating that age is negatively correlated with extent of stress. There was also a trend toward a regression effect for cause of death (F = 3.4, 1 and 65 df, p < .070). The regression coefficient of -2.49 indicates that longer causes of death are negatively correlated with extent of stress, or that sudden deaths are positively related to high stress levels. Finally, an analysis of change in psychological functioning, reactions to widowhood, and life style was made after the follow-up period, based on the experimental subjects' difference scores from pretest to follow-up. (Control group Ss were not included in this analysis). A one-way analysis of variance on the ten previously discussed variables including change in extent of stress was performed, with a test for the error effect in addition to the test for main effect due to treatment condition. Multivariate and univariate analyses were computed both without covariates, and with the individual covariates of age, duration of widowhood, cause of husband's death, and preparation for widowhood. The results indicate no significant main effect for treatment in any of these follow-up analyses. A comparison of the means for change in health prediction at posttest and at follow-up (Tables 14 and 16) shows a similar pattern in the treatment groups, with a somewhat reduced magnitude in the change toward positive health in the confidant and C-R groups by follow-up. The results of the test for error effects without covariates was a pattern of non-zero change strikingly consistent with the pattern at posttest. As Table 17 demonstrates, the multivariate F ratio was highly signifi cant, and Ss increased in self-esteem, reported increased grief reactions, became more negative in their attitudes toward remarriage, and tended to increase their involvement in social roles (non-significant trend). A comparison of the mean changes at posttest and at follow-up suggests that treatment led to a further increase in self-esteem and in intensity of grief reactions during the follow-up period per se. As in the posttest analyses on the entire sample, follow-up univariate tests indicated that change in locus of control, life satisfaction, attitude toward widowhood, physical symptoms, and health prediction were not Table 16 Mean Change in Self-Esteem, Grief, Health Prediction, Attitude toward Remarriage and Social Roles at Follow-Up Variable Treatment N Mean S.D. Self-esteem Self-help 16* 4.6 6.9 Confidant 11* 3.2 4.1 C-R 17* 4.7 7.3 Intensity of grief Self-help 6.1 6.7 Confidant 9.4 10.2 C-R 6.6 11.6 Health prediction Self-help - .06 .57 Confidant .09 .70 C-R .11 .60 Attitude toward remarriage Self-help -1.1 2.1 Confidant -0.5 2.2 C-R -0.8 2.3 Social role engagement Self-help 2.4 7.4 Confidant 1.6 5.5 C-R 1.2 6.0 *Total N = 44 for each variable. Seven Ss omitted due to missing data on one or more variables: 3 from C-R, 3 from confidant, and 1 from self-help. 104 Table 17 Analysis of Variance: Test for Non-Zero Change (Error Effects) in Psychological Functioning, Reactions to Widowhood and Life Style at Follow-up F df P Multivariate on 10 variables* 3.79 10,32 .002 Locus of control NS Life satisfaction NS Self-esteem 18.98 1,41 .001 Attitude toward widowhood NS Intensity of grief 23.7 1,41 .001 Physical symptoms NS Health prediction NS Attitude toward remarriage 6.94 1,41 .012 Social role engagement 3.32 1,41 .076 Extent of stress NS *Total N = 44. Seven Ss omitted due to missing data on one or more variables: 3 from C-R, 3 from confidant, and 1 from self-help. Two Ss, 1 from self-help and 1 from confidant did not attend the follow-up session. 105 £ ) significantly different from zero. Nor was the change in extent of stress from pretest to follow-up. Results of the analyses of covariance at follow-up departed somewhat from those obtained at posttest. Although the multivariate tests of regression effects for the four covariates were not significant, a few significant univariate regression effects appeared. Specifically, age, cause of death, and preparation for widowhood were all significantly correlated with change in self-esteem by follow-up (see Table 18). The regression coefficients indicate that the greatest positive change occurred in the youngest subjects, those whose husbands had died suddenly, and those with the least preparation for widowhood. Age was also significantly related to change in attitude toward widowhood, with more positive change attained by younger Ss. There was also a trend for widowhood preparation to be related to change in locus of control, with the least prepared Ss becoming more externally controlled, and the most prepared Ss becoming more internally controlled. As in the posttest data analysis, the removal of variance due to the four covariates decreased the signifi cance of non-zero change in the dependent variables, but substantial error effects were maintained. The multivariate 106 107 Table 18 Analysis of Covariance at Follow-up: Regression Effects Influencing Change in Self-Esteem and Attitude toward Widowhood Covariate Criterion Regression Coefficient F df P Age Self-esteem -0.22 4.56 1,40 .039 Attitude toward widowhood -0.06 7.68 1,40 .008 Duration of widowhood No significant regression effects Cause of death Self-esteem -3.41 6.97 1,40 .012 Preparation for widowhood Self-esteem -1.77 5.10 1,40 .029 Locus of control -0.71 3.60 1,40 .065 analyses of variance yielded a significant error effect when covarying for cause of death (F = 3.05, with 10 and 31 df, p < .008) and widowhood preparation (F = 2.2, with 10 and 31 df, p < .044). The multivariate error effect approached significance when covarying for age and duration of widowhood. In the univariate tests, non-zero change in self esteem was maintained in all four analyses of covariance. Increased intensity of grief was significant in all covariate analyses except for age. The regression coeffi cient of -0.08 suggests that change in grief reactions can be accounted for by the increased intensity of grief reported by younger subjects. The significant error effect for attitude toward remarriage disappeared when covarying for age, for duration, and for preparation, but was not affected by cause of death. In addition to the significant regression effect for age and attitude toward widowhood described earlier, a new error effect for attitude toward widowhood across experimental Ss appeared when the variance due to age was controlled (F = 7.19, with 1 and 40 df, p < .011). Inspection of the treatment means for this variable (S-H = -.25, CF = -.46, C-R = +.35, grand mean = -.07) suggests an overall small negative increment in attitudes toward widowhood over time. In the analysis of covariance for age, there was also a new trend for non-zero change in life satisfaction (F = 3.50, with 1 and 40 df, p < .069). The mean change in life satisfaction in all treatments was positive. Analysis of Change in Attitudes toward Women A multivariate analysis of variance on the three dependent variables measuring attitudes toward women was done using the pre-post difference scores of the experi mental subjects with leader (2 levels) and treatment (3 levels) as independent factors. In neither this analysis nor subsequent univariate tests were there any significant main effects for the leader factor; nor were there any significant leader by treatment interactions. Hence, in further analysis of change in attitudes toward women, treatment groups were collapsed across the leader variable. Pre-post change scores on all Ss including the control group were subjected to a one-way (four treatments) multivariate analysis of variance with planned comparisons to determine (1) the extent of non-zero change across Ss (the error effect), (2) the difference in pre-post change between the combined experimental treatments and the con trol group (Contrast 1) and (3) the extent of differences among the treatment groups in attitudes toward women at posttest (Contrast 2). The dependent variables in this analysis were other orientation, self orientation, and radical versus conservative attitudes. The results are presented in Table 19. There was a significant error effect in the multivariate test, and subsequent univariate tests revealed a significant reduction in other orientation across all subjects and an almost significant increase in self orientation (p < .054). The multivariate test of Contrast 1 was not significant. However, the experimental versus control group effect for change in other orientation was significant, with the experimental groups becoming less other-oriented and the control group becoming more other- oriented. There were no significant between-treatment differences, and none of the univariate tests on radical versus conservative attitude change was significant. Separate analyses of covariance for age, duration of widowhood, cause of death, and preparation for widowhood were performed on the three attitudes toward women variables with the same planned comparisons procedure. The multi variate covariate analyses resulted in significant 110 Table 19 Posttest Analysis of Change in Attitudes Toward Women Variable Treatment N Mean S.D. Other orientation Self-help 16* -2.38 4.4 Confidant 15 -1.53 4.4 C-R 18* -1.83 3.5 Control 17 0.29 3.7 Self-orientation Self-help 1.31 4.9 Confidant 0.47 4.8 C-R 0.67 3.0 Control 1.29 3.0 Radical versus conservative Self-help -0.69 3.0 Confidant 2.13 5.0 C-R 0.28 4.4 Control 1.00 3.4 Variable Effect F df P Multivariate* Non-zero change 4.20 3,60 .009 Other orientation Non-zero change 7.50 1,62 .008 Treatment vs. control 3.86 1,62 .054 Self-orientation Non-zero change 3.47 1,62 .067 Radical vs. conservative No significant effects *N = 66 for each variable. Four Ss omitted from analysis due to missing data on one or more variables: 2 from self-help and 2 from C-R. Ill regression effects for duration of widowhood and prepara tion for widowhood (see Table 20). Subsequent univariate tests indicated that duration of widowhood is significantly correlated with change in self-orientation. Specifically, the most recent widows increased their self-orientation the most. Preparation for widowhood is significantly correlated with other orientation; the least prepared widows showed the greatest reduction in other orientation. The pattern of significant error effects for non zero change in attitudes toward women at the posttest varied with the covariates. The multivariate error effect was maintained when covarying for duration, cause of death, and preparation, but was not present in the age covariance analysis. The univariate error effect for change in other orientation persisted when covarying for cause of death and preparation; it was reduced to a trend in the age covariance analysis and disappeared in the duration covariance analysis. Significant non-zero change in self- orientation occurred when partialing out variance due to duration of widowhood (F = 16.66, with 1 and 61 df, p < .001) and cause of death (F = 5.11, with 1 and 61 df, p < .027). In the duration covariance test, a non significant trend for increasingly radical attitudes toward 113 Table 20 Posttest Analyses of Covariance on Attitudes toward Women: Regression Effects for Duration and Preparation Covariate „ . Regression Criterion „ _.. . Coefficient F df P Age No significant regression effects Duration of widowhood Multivariate (N=66) 5.01 3,59 .004 Other orientation NS Self-orientation -0.04 13.24 1,61 .001 Radical vs. conservative NS Cause of death No significant regression effects Preparation for widowhood Multivariate (N=66) 3.04 3,59 .036 Other orientation 1.07 8.00 1,61 .006 Self-orientation NS Radical vs. conservative NS \ ■ V } 4 women emerged (F = 3.10, with 1 and 61 df, p < .084). The significant experimental versus control group comparison for other orientation held up with duration and preparation as covariates, was almost significant with age as a covariate, but disappeared when covarying for cause of death. Since the four treatments differed in cause of death at the pretest, the reduction in other orientation experienced by the experimental Ss is probably due to random error. No posttest between group differences in attitudes toward women resulted from the analyses of covariance. To assess changes in attitudes toward women at follow-up, the experimental subjects' change scores on the three variables from pretest to follow-up were subjected to multivariate analysis of variance with treatment (three levels) as the independent factor. The error effect, as well as the main treatment effect, were tested, and analy ses were performed both without and with each of the four covariates. The results of the tests for error effects were similar to the posttest data: the multivariate test and the univariate test for other orientation were signifi cant or near-significant in the initial analysis (see Table 21) and in the analysis of covariance for age and 114 Table 21 Analysis of Change in Attitudes toward Women at Follow-up Variable Treatment N Mean S.D. Other orientation Self-help 16* -1.81 2.6 Confidant 14* -1.21 3.0 C-R 18* -0.61 4.4 Self-orientation Self-help -0.50 4.3 Confidant 1.29 6.0 C-R 1.00 3.5 Radical vs. conservative Self-help -0.06 3.3 Confidant 2.00 3.7 C-R 0.28 3.5 Variable Effect F df P Multivariate* Non-zero change 2.57 3,43 . 066 Other orientation Non-zero change 5.57 1,45 .023 Self-orientation No significant treatment or error effects Radical vs. conservative No significant treatment or error effects *Total N = 48 for all variables. Three Ss omitted from analysis due to missing data on one or more variables: 2 from C-R and 1 from self-help. preparation, but these effects disappeared when covarying for duration of widowhood and cause of death. A trend for non-zero change in self-orientation appeared in the follow-up analysis using duration as a covariate (F * 3.54, with 1 and 44 df, p < .066). As in the posttest, no significant treatment effects developed either with or without covariates. The major discrepancy in the evaluation of changes in attitudes toward women at follow-up relative to the posttest was the pattern of regression effects for the covariates themselves. As Table 22 outlines, the signifi cant relationships between the attitudinal variables and duration and preparation vanished, whereas age became a significant covariate. The regression coefficient indi cates that younger Ss decreased the most in other orienta tion over time. Analysis of Behavioral Measures At the posttest, Ss reported the number of new acquaintances in their groups with whom they had contact outside of regular group meetings, and the total number of contacts. Due to an administrative error, Ss in the second leader's self-help and confidant groups, and 3 Ss from the 116 Table 22 Follow-up Analyses of Covariance on Attitudes toward Women: Regression Effects Covariate Criterion Regression Coefficient F df P Age Multivariate (N=48)* 3.55 3,42 .022 Other orientation .162 9.90 1,44 .003 Self-orientation NS Radical vs. conservative NS Duration of widowhood No significant regression effects Cause of death No significant regression effects Preparation for widowhood No significant regression effects *Three Ss omitted from analyses due to missing data on one or more variables: 2 from C-R and 1 from self-help. n p consciousness-raising group did not receive this data form. A multivariate analysis of treatment effect on the 33 Ss for whom data are available is reported in Table 23. The average subject in the confidant treatment contacted two persons outside the group, and made contact more than seven times during the treatment period. The greater amount of outside contact in this treatment is significant for the number of new acquaintances contacted, but the multivariate test and test for number of contacts is not significant with this small a sample. Data on all experimental Ss for the number of group members contacted during the follow-up period (including occasional prior acquaintances), the total number of contacts with individuals during follow-up, the number of group contacts which occurred after the end of treatment, and the rating of positive or negative life change since the pretest were subjected to a multivariate treatment (three levels) by leader (two levels) analysis of variance. The results are shown in Tables 24 and 25. The multivariate test for treatment was significant at the .004 level. Univariate tests on number of persons contacted and life change were also significant (p < .002 and .004, respectively). Review of the group means in Table 20 118 Table 23 Extent of Contact with Group Members During Treatment Period Treatment No. of New Acquaintances Contacted No. of Contacts Self-help (N=8)* Mean 1.1 5.5 S.D. 0.6 6.7 Confidant (N=8)* Mean 2.0 7.2 S.D. 1.3 5.7 C-R (N=17)* Mean 0.8 2.8 S.D. 1.0 3.4 F df P Multivariate (N=33)* 2.04 4,58 NS No. of new acquaintances contacted 3.62 2,30 .039 No. of contacts 2.47 2,30 NS *Due to an administrative error, data for 10 self-help Ss, 7 confidant Ss, and 3 C-R Ss were not available. 119 120 Table 24 Group Means and Standard Deviations for Contact with Members and Life Change at Follow-up Treatment Leader N No. of Persons Contacted No. of Contacts Group Meetings Life Change Self-help 1 7 Mean 0.43 22.57 0.14 3.14 S.D. 0.54 36.58 0.38 1.07 2 10 Mean 0.80 1.50 0.90 3.40 S.D. 1.14 2.32 1.10 1.43 Confidant 1 8 Mean 4.12 13.00 1.25 2.88 S.D. 2.64 14.20 0.89 1.46 2 6 Mean 0.50 1.00 0.00 4.17 S.D. 0.84 1.67 0.00 0.75 C-R 1 11 Mean 1.82 2.82 0.82 4.09 S.D. 1.33 2.04 1.40 1.22 2 9 Mean 1.33 5.33 0.00 4.44 S.D. 1.00 6.34 0.00 1.01 Total 51 A * Table 25 Multivariate Analysis of Variance on Contact with Group Members and Life Change at Follow-up F df P Test of Treatment Multivariate (N=51) 3.14 8,84 .004 No. of persons contacted 7.03 2,45 .002 No. of contacts 0.84 2,45 NS Group meetings 0.38 2,45 NS Life change 3.36 2,45 .044 Test of Leader Multivariate 3.47 4,42 .016 No. of persons contacted 6.97 1,45 .011 No. of contacts 4.87 1,45 .033 Group meetings 2.66 1,45 NS Life change 2.85 1,45 NS Treatment-Leader Interaction Multivariate 4.45 8,84 .001 No. of persons contacted 8.04 2,45 .001 No. of contacts 2.93 2,45 .063 Group meetings 5.40 2,45 .008 Life change 0.82 2,45 NS 121 indicate that confidant Ss contacted the most group members (an average of 2.3 across leader) followed by the C-R groups (1.6 persons per member) and that self-help Ss con tacted the fewest number of persons (average = .6). The most positive life change occurred in the C-R groups, followed by the confidant groups. The least positive life change occurred in the self-help groups. Univariate tests for number of contacts and group meetings were not signifi cant. The number of individual contacts ranged from a group mean of one to over 22. The grand mean for this variable was 7.09. Informal feedback at the follow-up session revealed that four of the six groups planned to continue to meet on their own. Analysis of variance also yielded a significant main effect for the leader factor. The probability level of the multivariate test was .016; univariate tests for number of persons contacted and number of contacts were also significant (p < .011 and .033, respectively). Review of the group means indicates that subjects in more of the first leader's groups had higher rates of follow-up contact. (These groups also began earlier, and thus had a two-week longer follow-up period.) There was no leader effect for number of group meetings or quality of life change. A significant interaction between treatment and leader also resulted from the analysis. (Multivariate F test yielded a probability level of .001.) Univariate tests for the interaction effect on number of persons contacted and number of group meetings were significant (p < .001 and .008, respectively), and a trend for number of contacts occurred. The group means in Table 18 suggest that this interaction is the result of the two leaders' discrepant outcomes in the confidant treatment. The first leader's highest levels of follow-up contact were reached in this group; the second leader's lowest levels were obtained here. Analysis of Group Evaluation Measures Experimental subjects' ratings of the helpfulness and educational value of their groups were analyzed at posttest and again at follow-up to assess treatment and leader effects. As the multivariate analysis of variance in Table 26 shows, there was a significant main effect for treatment at the posttest, but no effect for leader or for treatment-leader interaction. The treatments were signifi cantly different in both perceived helpfulness and < y i ' Table 26 Subjects’ Evaluation of Treatment Groups at Posttest Treatment Leader N Helpfulness Educational Value Self-help 1 8 Mean 4.88 4.38 S.D. 1.6 2.3 2 10 Mean 4.00 3.90 S.D. 1.6 1.9 Confidant 1 8 Mean 5.38 5.75 S.D. 1.4 1.4 2 7 Mean 5.29 5.29 S.D. 2.4 2.0 C-R 1 11 Mean 6.09 6.09 S.D. 1.0 1.4 2 9 Mean 5.33 5.33 S.D. 0.7 1.0 F df P Test of Treatment Multivariate (N=53) 2.62 4,92 .040 Helpfulness 4.05 2,47 .024 Educational value 5.03 2,47 .010 Test of Leader No significant effects Treatment-Leader Interaction No significant effects 124 educational value with the highest ratings obtained on both variables in the C-R groups, and the lowest in the self- help groups. Table 27 describes the group evaluation data at follow-up. Although the lowest ratings were still given by self-help participants and the highest were still given by C-R members, the treatment main effect was not signifi cant. The univariate test for helpfulness resulted in a trend for significance. Examination of the group evalua tion means and standard deviations suggests that the reduced significance levels at follow-up are attributable to the increased variance in ratings in almost all cells. As in the posttest, the leader effect and leader-treatment interaction were inconsequential. A special evaluation was made for the confidant treatment: each S rated the relationship with her new friend on five variables (others' knowledge of self, liking for the other, ease of discussing problems, present relationship, and predicted future relationship). Means and standard deviations for these variables at posttest and at follow-up are listed in Table 28. The ratings are generally positive. Liking for the other increased somewhat during the follow-up period (from a mean of 5.23 to 5.62 on a Table 27 Subjects' Evaluation of Treatment Groups at Follow-up Treatment Leader N Helpfulness Educational Value Self-help 1 7 Mean 4.57 5.14 S.D. 2.0 2.0 2 10 Mean 3.70 4.40 S.D. 1.9 2.2 Confidant 1 8 Mean 4.50 5.50 S.D. 2.1 1.5 2 6 Mean 5.67 5.67 S.D. 2.2 2.2 C-R 1 11 Mean 6.27 6.45 S.D. 0.9 0.8 2 9 Mean 4.67 5.11 S.D. 1.6 1.0 F df P Test of Treatment Multivariate (N=51) 1.37 4,88 NS Helpfulness 2.62 2,45 .084 Educational value 0.84 2,45 NS Test of Leader No significant effects Treatment-Leader Interaction No significant effects 126 Table 28 Confidant Relationships at Posttest and Follow-up Variable Posttest Follow--up Mean S.D. Mean S.D. Other's knowledge of self 3.54 1.76 3.54 1.56 Liking for other 5.23 1.24 5.62 1.12 Ease of discussing problems 5.08 2.33 4.77 2.39 Present relationship 3.23 1.59 3.15 1.77 Future relationship 4.15 1.77 4.00 1.53 N = 13. Two Ss at posttest and one S at follow-up were excluded due to missing data. seven-point scale). Changes in the other ratings were negligible. At both points in time the predicted future relationship was rated higher than the current relationship. The final group evaluation measures to be described are the experimental subjects' preferences for certain types of group composition expressed at the posttest. As Table 29 specifies, the majority of Ss in the program preferred a widows1 group in which members varied both with respect to duration of widowhood and age. These preferences were congruent with the actual composition of treatment groups in the research. However, almost three- fourths of the Ss at posttest recommended that widowers and widows be included in the same group. Only seven Ss expressed a preference for sex-segregated widowhood groups. 128 Table 29 Group Composition Preferences at Posttest Among Experimental Subjects Type of Preference Subjects Number Percent Diverse duration of widowhood 31 58.5 Uniform duration of widowhood 8 15.1 No duration preference expressed 14 26.4 Totals 53 100.0 Diverse age group 29 54.7 Uniform age group 18 34.0 No age preference expressed 6 11.3 Totals 53 100.0 Sex integrated group 39 73.6 Sex segregated group 7 13.2 No sex preference expressed 7 13.2 Totals 53 100.0 129 CHAPTER V DISCUSSION OF THE RESULTS The Sample The study sample was fairly representative of widows as a national population in terms of age distribu tion, although there were somewhat fewer widows at the upper extreme of the age continuum than would have been expected on purely statistical basis. This may have resulted from greater transportation difficulties in the aged group. Duration of widowhood in the sample was also roughly comparable to the national population pattern. The self-selection of women who had been widowed for long periods of time into a helping program provides unavoidable evidence of the long-term stresses of widowhood, and suggests that programs geared only for the recently widowed may be grossly insufficient. (The question of whether such programs could prevent long-term stress in a crisis intervention model has yet to be tested.) The per centage of Ss widowed less than two years (32.9%) was 130 considerably less than that in the Boston telephone hotline service (Abrahams, 1972). The news release for this study apparently functioned to sanction the needs of more experienced widows. Few women whose husbands had just died enrolled in the program. (Only one S had been widowed less than a month at the pretest.) This suggests that more direct services such as the San Diego home visiting program (Antoniak, 1973) or a hotline service providing greater anonymity and requiring less energy of the participant may be more useful with the recently widowed group. The fact that Ss came from all parts of Los Angeles county, sometimes driving more than two hours to attend the program, speaks again to the vast unmet needs of the urban widowed. The study sample was roughly comparable to the Chicago area sample studied by Lopata (1973) in terms of employment, living arrangement, and housing, although a somewhat higher proportion of this sample owned their own homes and worked. The popular negative stereotype of a widow as a person with nothing but time on her hands who is apt to be living with ungrateful adult children is erroneous at best. The large number of subjects in the study with extremely meager preparation for widowhood (one half had 131 never thought about it; only 16 had specific ideas what to do about it) raises the possibility of a broad-scale educational intervention prior to widowhood as a deterrent to some of the almost inevitable stresses. Programs could tap into the modest goal of preparing the spouses of termi nal illness victims (e.g., Parkes, 1973) or could attempt much more grandiose consciousness-raising in the entire population using grade school and high school curriculum or pre-marital counseling motifs. Although the sample in the current program was generally representative of widows nationally on the variables described above, several sample biases occurred. Ethnic minorities were underrepresented; the Jewish religion was uncharacteristically abundant. Although the current incomes of these widows were moderate (mode: $600 to $900 a month), many subjects had been living with upper- middle class incomes as marital partners (mode: more than $1,500 a month). Finally, the sample as a whole was well- educated. Almost three fourths had had some college, and over 157o held graduate or professional degrees. These factors suggest a general sample bias of higher than aver age socioeconomic status prior to widowhood. Two important questions arise: the reason or reasons for the sample biases which occurred (and as a corollary, the strategies for reaching a more inclusive group), and the relevance of these biases in making predic tions about the response of widows generally to similar therapeutic interventions. With regard to the first question, several hypothe ses can be formulated. The sample biasing may have occurred through differential exposure to the program, differential rates of interest in the program, or differ ential means for taking advantage of it. Location of the program on the University of Southern California campus provided proximity to a large Black population, yet ethnic minorities were underrepresented. The sample bias could be due in part to differential news coverage. There may be different subscription rates and reading patterns for the Los Angeles Times with regard to ethnicity, religion, education, and income. The greater news coverage in the San Fernando Valley, which has a relatively low proportion of ethnic minorities, through a widely distributed commu nity newspaper may have contributed to the sample bias. Location of the program on a university campus may have been much more attractive to university educated potential subjects than those without college backgrounds. (It may also have been a deterrent to potential Ss in the local Black community cautious of university-sponsored research.) Hence, future widowhood programs located in churches or private residences might reach more low S.E.S. groups, as might programs providing free child care and/or transpor tation. The religious bias in the study sample may be due to correlation with the ethnic, income, and educational variables. Jewish women are overrepresented in the private practice of psychotherapy. The helping nature of this program, as well as the research and university components, may have been more congruent with Jewish subcultural norms. An alternative to the hypotheses expressed above is the possibility that the widespread publication of the program led to a sample which accurately reflected the degree of stress experienced by various groups of widows. Thus, for whatever reason, widowhood may be a more stress ful life stage for Jewish women and for those with formerly high socioeconomic status, and a relatively less stressful experience among ethnic minorities. Consistent with this interpretation is Lopata's (1973) discovery that Black women were twice as likely as their white sisters to find widowhood entirely lacking in problems (13% compared to 134 67o.) Lopata (1972) has postulated that the degree of stress in widowhood is related to the extent and intensity of change in social roles. This may be a more important variable than the absolute value of coping resources (such as funds) available to the widow. The "status shock" of widowhood for middle and upper class women may have led to their increased participation in the program. Religion as a variable in widowhood will be investigated in a subse quent paper. The issue of research generalizability when the study sample is a specific subgroup of the population of widows can only be addressed at this point with a general caution: the conclusions reached in this dissertation research were based on data reported largely by widows who are Caucasian, well educated, and formerly financially secure, but who in most other respects are representative of widows as a national group. Extensions of the research with ethnic minorities and lower class groups are highly recommended. Assignment of Subjects The fact that there were no significant differences between treatment groups at the pretest on twelve different 135 dependent variables and on only one of the four demographic variables tested suggests that the procedure of assigning Ss to groups primarily on the basis of subject availability and geographic location approached randomization. The differential rates of long-term illnesses and sudden deaths among the three treatments and control group was probably a random error due to chance. Given the non-random distribution of the cause of death variable, the question of whether it is statistically legitimate to attempt analyses of covariance on this variable can be raised. A necessary assumption in analysis of covariance is homogeneity of regression; that is, the relationship between the covariate and the chosen dependent variable is assumed to be the same in all treatments. A possible violation of this assumption becomes irrelevant when considering that there were no treatment effects for any of the dependent variables on which covariate analyses were undertaken. Response to the Program: Treatment Versus Control Group The results of the comparisons between the control group and the experimental subjects at posttest are 136 characterized by a paucity of significant results. On 12 variables designed to assess psychological functioning, reactions to widowhood, changes in life style, and atti tudes toward women, the treated Ss responded differently from controls on only two. They demonstrated more positive changes in their predictions of future health, and they became significantly less other-oriented in their attitudes toward the roles of women. That the differential assessment of future health by the control and experimental Ss is a real effect in response to the treatment program is implied by the finding that this result persisted throughout analyses of covariance for age, duration of widowhood, cause of death, and widowhood preparation. It is an important finding in view of the poorer health of widows with respect to same- age married persons. It is also intriguing in that the intervention was a psychological and social one--not a physical one. The significant difference in change in health prediction is even more striking when considering that it represents pre- to posttest change on only a four-point scale from poor health to excellent health. The mean change in health prediction by follow-up indicates some retrenchment in the positive gains made by two of the 137 treatments, but a persistent overall pattern of improved outlook in physical health for Ss who attended the widows' groups. This finding merits further study. Parkes (1970) found that among 22 unselected London widows, self- estimates of "poor general health" were closely correlated with assessments of irritability and anger during the interview, but not with the actual number of physical symptoms. The interpretation that the group interventions did in fact lead to a reduction in irritability and anger is supported by the group leaders' experience of the sub jects at the orientation meetings, when frequent hostile remarks and questions occurred, in contrast to the final sessions, when the overall effect was much more positive. The experimental-control difference in change in other orientation at the posttest was maintained in three analyses of covariance, but disappeared when covarying for cause of death. This suggests that although experimental subjects did become less other-oriented, this change in attitudes toward women can be attributed more to the non- random pattern of sudden and long-term causes of deaths among the groups than to the therapeutic intervention per se. Since more husbands of control group Ss died of 138 long-term illnesses, we can infer that sudden deaths are related to a reduction in other orientation. Perhaps it is more difficult for the woman who has cared for a dying husband for years to change her role commitment to others' needs than for the woman jolted into awareness of her own needs by the sudden death of her spouse. The lack of significant control-experimental differences on the other variables deserves attention. Between treatment differences were also rare, and hence could not have been camouflaging improvements made by individual treatment groups relative to the controls. And the presence of significant change across all subjects on several dependent measures argues against the possibility that unusual pretest levels in the sample precluded significant experimental group change. The firmly positive feedback about the program with regard to helpfulness and educational value is contradic tory to the conclusion that the program was ineffective. The substantial rates of contact in all the treatments subsequent to the intervention (four out of six groups con tinued to meet; the average subject had 7 individual contacts with other participants during the follow-up period) is prima facie evidence that the program for widows 139 provided an alternative to loneliness, the number one problem in the population under study. It appears either that anticipation of help by the controls (who were just beginning their small group experience at the posttest) masked experimental-control differences, or that some of the dependent measures were insensitive indicators. Several of the measures consisted of summations of a series of more discrete variables: intensity of grief, social role engagement, and physical symptoms. The I-E scale contains many items irrelevant to the current experience of elderly widows, e.g., items about school tests. There was a tendency for items to be left blank on this scale, despite encouragement from the administrators. The life satisfaction measure may have been too expansive in scope to reflect change occurring in an eight to nine week period. Unfortunately, no measure of changes in specific stresses of widowhood was available at the posttest, and parallel data on the control group Ss was not available at the follow-up, since they had been assigned to a widows1 group in the interim. Nor were there any behavioral measures on the control group during the treatment period. 140 Response to the Program: Comparison of Three Treatments There were no treatment effects at posttest or at follow-up on any of the variables measuring psychological functioning, reactions to widowhood, changes in life style or attitudes toward women, and no treatment effects emerged in the analyses of covariance. However, the behavioral and group evaluation measures did yield significant treatment differences. The confidant subjects as a whole had the most contact with members outside their groups, both during the follow-up and treatment periods. This finding may be interpreted as a demonstration that the confidant treatment in general was successful in generating friendship pairs who continued to relate to each other several months after the intervention; they also predicted an increase in intimacy in the future relative to their current relation ship both at posttest and again at follow-up. The significant leader effect and leader-treatment interaction on variables measuring contact between members suggests a note of caution with regard to the consistency of outcome using the confidant treatment approach. Although the treatment effect was significant, one of the leader's lowest rates of between-member contacts occurred in this group. Both leaders expressed concern over the difficulty in keeping their confidant groups focused on planned tasks. It frequently became necessary to deviate from the couple-oriented interpersonal tasks to deal with the participants' objections to the format. Both leaders had the impression that the demand for intimacy in this treatment was a stressful experience for some group members. The poorest attendance occurred in these groups. The significant treatment effect for extent of outside contacts during the treatment, and the almost significant chi-square analysis of members' perceptions of their leaders' role indicate that the treatment groups were in fact functioning differently, despite the lack of differential response in many of the dependent variables. The group evaluation measures and the rating of positive versus negative life change from pretest to follow-up provide the key assessment of the relative success of the three treatment strategies. All of these measures consistently favored the consciousness-raising groups, both at posttest and at follow-up. The results were consistent for both leaders, despite their differential degrees of comfort with this treatment approach prior to the pretest. C-R members rated 142 their experiences as most helpful and of most educative value. Blind ratings of the personal descriptions of how their lives had changed in the preceding five months were the most positive. Some examples of their responses follow: After living in an apartment for 22 years, I moved to another apartment. I did not take the initiative, but did take the action when an unpleasant situation arose between the landlord and myself. Through the years I had accepted a lot of inconvenience and harassment, but this time I finally rebelled, even though I dreaded the work entailed. I'm very happy now that I made the change. I realized that others had many of the same feel ings and problems as I did. I heard some "stories" that were much more tragic than mine. I learned that life can be hard and is not easy. I have learned more about myself (strengths and weaknesses), the others around me and to appreciate everything and every day of my life. "My Widow's Group" had more impact on me than any other thing in my whole life! Change of job. A new friend out of the group. Taking a workshop at American Institute of Family Relations for "Betterment of Communication." I'm ready to start a square dancing group in March. I participated in a consciousness-raising experi ence and found it very supportive. I met with a group of women from all walks of life with whom I identified very closely--very warmly--looked forward to our weekly sessions--found strength in the realization that women in general and widows in particular had many common needs. It was a sharing experience that was a revela tion to me--since I had never participated on such a "gut" level with people of varied socioeconomic levels. 143 Somewhat calmer in dealing with important issues. Have accepted, to a degree, my bereavement. Am able to handle it better but I am hopeful to do better in time. Have taken new interest in life and in what's going on around me. Wish to get into service or business as I now feel I have some talent to offer. I'm not sure what it is but I'm still searching. More confident currently in going places and doing things alone--gaining back security of ability to be involved, and enjoyment of being involved in social activities, planning of same. Series of incidents of meeting new people (members of opposite sex) and realizing that I am attractive and interesting to them. Am pleased with ability to feel I can be and am selective with the friendships and relationships with people--don't feel a desperate need to hold on to people because of fears of aloneness. Rather be alone than with "anybody." Frustrated at inability to "make" it on job--recognize it's not me, but the "establish ment" set-up of the particular business world I'm in. Have attitude that I have much to offer and it's their loss not taking recognition of same, or maybe they are uncomfortable with my forthrightness. The self-help groups gave or got the lowest ratings. The confidant groups were intermediate in their evaluation of the program. The presence of significant treatment effects on the behavioral and group evaluation measures in the arid context of non-results with the other dependent variables underscores the importance of outcome criteria which pertain closely to demonstrable behavior change and specific experiences, rather than global judgments and 144 attitudes. The superior ratings achieved by the C-R groups raises the obvious question of "Why?" Many explanations might be formulated but this may be premature given the fairly circumscribed differences between treatment out comes. Two possibly important factors can be derived from the leaders' clinical experience with the various treatment modalities. The first is the degree of structure in the C-R group format with suggested topics, a firm limit on confrontive relating, and a process enabling each member to share equally in the discussion. There was considerable structure in the confidant groups as well, but the emo tional sequelae to the basic purpose of the group often overrode it. (This group intervention might have been more effective if limited to those Ss who explicitly lacked a confidant.) In contrast to the C-R groups, the self-help groups received a minimum of structure from the leader. Topics for conversation and group process were both more in the hands of the participants. It was the leaders' experience that this more ambiguous situation was uncom fortable for some of the self-help participants even late in the treatment, since bids for more active leadership persisted. It may be that the self-help format would function more optimally with a leader who is herself 145 widowed (Silverman, 1970). Research is currently underway to test the relative importance of structured group process as opposed to the content differences between the C-R and self-help widows' discussion groups. The second factor which appeared important was the function of the C-R group topics in legitimizing the anger and resentment experienced by widows. The role of anger in combating depression has long been observed clinically. The C-R group experience provided a well-deserved external target for angry feelings by focusing on the sex role oppression of women in this society. It is somewhat surprising in view of the higher ratings of the women's consciousness-raising groups (which by definition are uniform in sex) that so few subjects indicated a preference for all women's groups at the post test. Although it is helpful for women of all ages and varying length of experience as widows to get together to discuss their mutual concerns, many widows deliberately want an opportunity to share their experiences with widowed men. 146 The Nature of Change Over Time Although the program for widows led to few differ ences in measurable psychological change between the wait ing list subjects and the participants or between the treatment groups per se by posttest, significant change did occur across all subjects. The changes that were still reliable at follow-up several months later (in at least the experimental Ss) were increased self-esteem, increased intensity of grief, more negative attitudes toward remarriage, and reduction in other-oriented attitudes toward the roles of women. Of these four types of change, only one was consistently significant in the follow-up analyses of covariance for age, duration, cause of death, and preparation: this was the significant increase in self-esteem. (This effect was substantial in all post test covariance analyses as well, with the exception of duration of widowhood.) Subject attributes accounted for more of the change in grief reactions and attitude toward remarriage than time per se. The significant reduction in other orientation disappeared in the posttest when covary- ing for duration; the same thing happened at follow-up. However, a new effect for increased self orientation across subjects occurred with duration as a covariate in the posttest analysis, and was almost significant in the same analysis at follow-up. Thus the most reliable changes that occurred in this widowed sample regardless of which condition they were assigned to during the five-month observation period were increased self-esteem and increasingly self-oriented attitudes toward the roles of women. Participation in the research project led to a more positive view of oneself as a person and a greater subscription to attitudes reflecting women's needs for individual achievement and autonomy and the belief that traditional family roles are inadequate in meeting their needs. The Importance of Subject Variables Each of the four demographic variables used in the analysis of response to treatment was significantly correlated with change in at least one dependent variable, as indicated by analysis of covariance. However, since the pattern of significant regression effects for age, duration of widowhood, cause of death and preparation for widowhood was inconsistent from posttest to follow-up, the results should be interpreted cautiously. (This inconsistency may 148 result in part from the omission of controls in the follow-up study.) The level of pretest stress recalled at posttest was dependent on the age of the widow, with the greatest stress endured by younger subjects. This finding is con sistent with the conclusion of Heyman and Gianturco (1972) that the crisis of widowhood is greater for younger persons. A variety of factors may account for this phenomenon, such as the greater disruption in life cycle expectations, the reduced likelihood of widowed companions, and the greater likelihood of parenting responsibilities. Reduction of stress level during the study was not correlated with any of the four subject variables selected as covariates. At the posttest, change in self-orientation depended on how long the subject had been widowed, with the greatest increase shown by the most recent widows. Post test change in other orientation was related to preparation for widowhood, with the greatest reduction shown by the least prepared widows. At follow-up, age was significantly correlated with change in self-esteem, attitude toward widowhood, and other orientation. The younger the subject, the greater the increase in self-esteem, the more positive one's attitude 149 toward widowhood became, and the more one deviated from previous other-oriented attitudes toward women. Change in self-esteem by follow-up was also related to preparation for widowhood and cause of death. The greatest increments in self-esteem during the study period were experienced by ill-prepared widows whose husbands had died suddenly. These results suggest the tentative conclusion that those subjects who benefited most from the program were relatively young, rather recently widowed women who had minimal preparation for life roles ensuing after a frequently sudden death of the spouse. Further research is planned to determine whether it is this very group of widows who have the most to gain from any therapeutic intervention, or whether the group modalities offered by this research program were particularly responsive to their needs. 150 CHAPTER VI CONCLUSION Seventy urban widows ranging in age from 32 to 74 and in duration of widowhood from one month to 22 years participated in one of three small group interventions for a two-month period--a self-help group, a "confidant" group, or a women's consciousness-raising group--or a waiting list control group (with subsequent group intervention). Data on a wide variety of dependent variables was obtained prior to treatment, at the final group session, and at follow-up three and one-half months posttreatment. At the pretest assessment of 16 variables, the only difference among the treatment groups was in cause of death, with husbands of self-help and C-R participants experiencing more sudden deaths than husbands of other Ss. There was a high propor tion of Jewish women in the sample (40%,) , an over representation of women from formerly high socioeconomic levels, and few ethnic minority members. Of the sample, 53% were employed. On all other variables assessed, the Ss resembled the population of widows nationally. 151 The comparison of controls and subjects in the discussion groups at the posttest yielded a significant difference in change in predicted future health, which favored the experimental Ss. This difference was consis tent for Ss of varying ages, duration of widowhood, cause of husband's death, and preparation for widowhood, and speaks for the success of the interventions. Experimental Ss also showed a greater decrease in other-oriented atti tudes toward women than control Ss, but the difference can be attributed primarily to the changing attitudes of women whose husbands had died suddenly. Experimental control group posttest differences did not occur on ten other variables, which can in part be explained by the insensitive quality of the measures, and by possible positive anticipation of the waiting list subjects who were beginning their small group experience on the day the posttest measures were obtained. Subjects in all groups, regardless of condition, demonstrated increases in self-esteem and self-oriented attitudes toward women. Although two different group leaders were each responsible for one of each of the three group interven tions, remarkably few differences in the response to treatment occurred as a result of the leader variable. The 152 only reliable differences between the treatment groups occurred in their evaluation of the program, and in the behavioral measures. Consciousness-raising participants gave the program the highest ratings of helpfulness and educational value, although all responses were generally positive. They also reported the most positive kinds of life changes during the five to six month evaluation period in an open-ended description. The confidant groups, as hypothesized, had significantly more contact with group members both during and following the group intervention, but the two confidant groups varied on this score. Rates cf contact were substantial in all treatments, with four out of six groups electing to continue meeting as a group even after the follow-up period. The results suggest that those women who benefited most from the program were the younger widows, the more recently widowed women (although no extremely recent widows participated), and those who had the least preparation for widowhood, both in terms of their own report, and the length of the husband's last illness. The results indicate that with this sample of widows, the most effective group intervention was a women's consciousness-raising experience, although there were no 153 differences between the interventions on a number of dependent measures. The structure of this group modality and the opportunity for the sanctioned expression of anger may have contributed to its success. Future research should ascertain whether C-R groups for widows are as help ful when the membership includes more women whose husbands died of long-term illness, more ethnic minority members, more women with minimal education and prior income, and more non-Jewish women. The high degree of continued contact following the group interventions is both an indication of their success in reducing the dreaded loneliness of widowhood, and in the extent of psycho-social needs among widows. The opportuni ties for creative service programs with this group of individuals are immense. The program demonstrates that combining women of varying ages and experience as widows can be a good strategy. Preventive intervention should be carried on at the level of the general public as well. This society needs broad-scale education and attitudinal change which will prepare women for the life cycle stage of widowhood, and alert us all to ways of facilitating the strengths of the widow. 154 REFERENCES REFERENCES Abrahams, R. B. Mutual help for the widowed. Social Work, 1972, U(5), 54-61. Adams, B. N. The middle-class adult and his widowed or still-married mother. Social Problems, 1968, 16(1), 50-59. Antoniak, H. The widowed to widowed program. Unpublished pamphlet, San Diego County Funeral Directors Associa tion, 1973. Bart, P. B. Depression in middle-aged women--some socio logical factors. Unpublished Ph.D. dissertation, University of California, Los Angeles, 1968. Bell, R. Marriage and family interaction. Homewood, 111.: Dorsey, 1963. Beilin, S. S., & Hardt, R. H. Marital status and mental disorders among the aged. American Sociological Review, 1958, 28, 155-162. Berardo, F. M. Survivorship and social isolation: The case of the aged widower. Family Coordinator, 1970, 19(1), 11-25. Bernard, J. Remarriage: A study of marriage. New York: Dryden, 1956. Blau, Z. S. Old age--a study of change in status. Unpub lished Ph.D. thesis, Columbia University, 1957. Bock, E. W. , 6c Webber, I. L. Suicide among the elderly: Isolating widowhood and mitigating alternatives. Journal of Marriage and the Family, 1972, 34(1), 24-31. 156 California Department of Public Health. California Health Survey, Health in California. Sacramento: State Printing Office, 1958. Carter, H., & Glick, P. C. Marriage and divorce: A social and economic study. Cambridge, Mass.: Harvard University Press, 1970. Champagne, M. M. Facing life alone, what widows and divorcees should know. Indianapolis: Bobbs-Merrill, 1964. Clark, M. M. Cultural values and dependency in later life. In R. Kalish (Ed.), The dependencies of old people. Occasional Papers in Gerontology, No. 6 ~, 1969, 59-72. Confrey, E. A., & Goldstein, M. S. The health status of aging people. In C. Tibbitts (Ed.), Handbook of Social Gerontology. Chicago: University of Chicago Press, 1959. Cosneck, B. J. A study of the social problems of aged widowed persons in an urban area. Unpublished Ph.D. thesis, Florida University, 1966. Cramer, E. M. Revision of MANOVA part of multivariate statistical programs. Chapel Hill, N. C.: L. L. Thurstone Psychometric Laboratory, University of North Carolina, 1970. Decker, B., & Kooiman, G. After the flowers have gone. Grand Rapids, Mich.: Zondervan, 1973. Dubey, B. R. Widow remarriage in madhya pradesh. Man in India, 1965, 45(1), 50-56. Egelson, J. , 6c Egelson, J. Parents without partners; a guide for divorced, widowed or separated parents. New York: Dutton, 1961. Folan, W. J. , 6c Weigand, P. C. Fictive widowhood in rural and urban Mexico. Anthropologica, 1968, 10(1), 119-128. 157 Frank, J. D. Persuasion and healing: A comparative study : of psychotherapy. New York: Schoken, 1963. Freeman, V. Personal communication, March, 1973. Glick, P. C. American families. New York: Wiley, 1957. Gorer, G. Death, grief and mourning. New York: Doubleday, 1965. Gump, J. P. Sex-role attitudes and psychological well being. Journal of Social Issues, 1972 28(2), 79-92. Harcourb, A. Poverty and the widow. Australian Journal of Social Issues, 1969, 4(2), 49-59. Heyman, D. K., & Gianturco, D. T. Paper presented at the American Psychological Association Annual Convention in Honolulu, Hawaii, August 1972, and summarized in the article, Reactions to widowhood in old age. Geriatric Focus, November, 1972, 2-3. Holmes, T. H., & Masuda, M. Life change and illness susceptibility. Paper presented as part of a "Sympo sium on separation and depression: Clinical and research aspects." Annual Meeting of the American Association for the Advancement of Science, Chicago, Illinois, December 26-30, 1970. Hurvitz, N. Peer self-help psychotherapy groups and their implications for psychotherapy. Psychotherapy: Theory, Research and Practice, 1970, 7^(1) , 41-49. Komarovsky, M. Functional analysis of sex roles. American Sociological Review, 1950, L5, 508-516. Kraus, A. A., & Lilienfeld, A. M. Some epidemiologic aspects of the high mortality rate in the young widowed group. Journal of Chronic Disease, 1959, 1JD, 207-217. Langer, M. Learning to live as a widow. New York: J. Messner, 1957. 158 Liberman, R. P., & Raskin, E. E. Depression: A behavioral formulation. Archives of General Psychiatry, 1971, 24, 515-523. Lindemann, E. Symptomology and management of acute grief. In R. Fulton (Ed.), Death and identity. New York: Wiley, 1965. Lopata, H. Z. The life cycle of the social role of house wife. Sociology and Social Research, 1966, 51(1), 5-22. Lopata, H. Z. Loneliness: Forms and components. Social Problems, 1969, 17(2), 248-261. Lopata, H. Z. The social involvement of American widows. American Behavioral Scientist, 1970, 14(1), 41-58. Lopata, H. Z. Living arrangements of American urban widows. Washington, D. C.: U. S. Administration on Aging, 1971. (a) Lopata, H. Z. Occupation: Housewife. New York: Oxford University Press, 1971. (b) Lopata, H. Z. Role changes in widowhood: A world perspec tive. In D. Cowgill and L. Homes (Eds.), Aging and modernization. New York: Appleton-Century-Crofts, 1972. Lopata, H. Z. Widowhood in an American City. Cambridge: Schenkman, 1973. Lopata, H. Z., Philbald, T., Adams, D., et al. Widowhood in selected areas of the U. S., Poland and Yugoslavia. England, 1970. Lowenthal, M. F. Social isolation and mental illness in old age. American Sociological Review, 1964, 29, 54-70. Lowenthal, M. F. Antecedents of isolation and mental ill ness in old age. Archives of General Psychiatry, 1965, 12, 245-254. Lowenthal, M. F., & Boler, D. Voluntary vs. involuntary social withdrawal. Journal of Gerontology, 1965, 20, 363-371. 159 Lowenthal, M. F., & Haven, C. Interaction and adaptation: Intimacy as a critical variable. American Sociological Review, 1968, 33, 20-30. Maddison, D., & Viola, A. The health of widows in the year following bereavement. Journal of Psychosomatic Research, 1968, 12, 297-306. Marris, P. Widows and their families. London: Routledge and Paul, 1958. Marsden, D. Mothers alone: Their way of life. New Society, 1969, 13, 705-707. Metropolitan Life Insurance Company. The American Widow. Statistical Bulletin, 1962, 43, 1-4. Neugarten, B. L., Havighurst, R. J., & Tobin, S. S. The measure of life satisfaction. Journal of Gerontology, 1961, 16, 134-143. Nie, N., Bent, D., & Hull, C. Statistical Package for the Social Sciences. New York: McGraw-Hill, 1970. Nuckols, R. C. Widows Study. Abstracted in the Journal Supplement Abstract Service, Catalog of Selected Documents in Psychology, 1973, 3, 9. Owen, J. Z. Widows can be happy. New York: Greenberg, 1950. Palmore, E. B., Stanley, G. L., & Cormier, R. H. Widows with children under social security, the 1963 national survey of widows with children under OASDHI. Washing ton, D. C.: Department of Health, Education, and Welfare, Social Security Administration, Research Report #16, 1963. Parkes, C. M. Effects of bereavement on physical and mental health: A study of the medical records of widows. British Medical Journal, 1964, 2, 274-279. Parkes, C. M. The first year of bereavement. Psychiatry, 1970, 33(4), 444-467. 160 Parkes, C. M. Lecture on bereavement research. Los Angeles County-U.S.C. Medical Center Department of Psychiatry, October, 1973. Parkes, C. M., & Brown, R. J. Health after bereavement. Psychosomatic Medicine, 1972, 34(5), 449-461. Rainwater, L. Family Design. Chicago: Aldine, 1965. Richford, E. Mothers on their own. New York: Harper, 1953. Rosenberg, M. The association between self-esteem and anxiety. Journal of Psychiatric Research, 1962, 1(2), 135-152. Rosenfeld, L. S., et al. Medical care needs and services in the Boston Metropolitan area. Boston: United Community Services of Metropolitan Boston, 1947. Rosenfeld, L. S., et al. Health services for the aging in Saskatchewan. In Illness and health services in an aging population, four papers. St. Louis: Second International Gerontological Congress, 1951. Rosenthal, E. Divorce and religious intermarriage: The effect of previous marital status upon subsequent marital behavior. Journal of Marriage and the Family, 1970, 32(3), 435-440. Rotter, J. B. Generalized expectancies for internal versus external control of reinforcement. Psychologi cal Monographs: General and Applied, 1966, 80Q), 1-28. Safilios-Rothschild, C. The study of family power struc ture: A review 1960-1969. Journal of Marriage and the Family, 1970, 32, 539-552. Segal, B. E. Suicide and middle age. Sociological Symposium, 1969, 3, 131-140. Sheldon, H. D. The older population of the United States. Glencoe, 111.: The Free Press, 1959. 161 Silverman, P. R. Services for the Widowed during the period of bereavement. In Social Work Practice, 1966. New York: Columbia University Press, 1966. Silverman, P. R. Services to the widowed: First steps in a program of preventive intervention. Community Mental Health Journal, 1967, 3(1), 37-44. Silverman, P. R. The widow-to-widow program. Mental Hygiene, 1969, 53(3), 333-337. Silverman, P. R. How you can help the newly-widowed. The Magazine of the American Funeral Director, May 1970. Silverman, P. R. Widowhood and preventive intervention. Family Coordinator, 1972, 21^(1), 95-102. Simon, A. Physical and sociopsychologic stress in the geriatric mentally ill. Comprehensive Psychiatry, 1970, 11, 242-247. Singh, T. R. Widow remarriage among Brahmans: A socio logical study. Eastern Anthropologist, 1969, 22(1), 75-85. Smith, T. L. A demographic study of widows. Sociologia, 1962, 24(2), 95-106. Spence, J. T., & Helmreich, R. The Attitudes Toward Women Scale: An objective instrument to measure attitudes toward the rights and roles of women in contemporary society. Catalog of Selected Documents in Psychology, 1972, 2, 66-67. Sumner, W. G., & Keller, A. G. The science of society. Yale, 1929. Thompson, E. J. Suttee; a historical and philosophical enquiry into the Hindu rite of widow-burning. Boston: Houghton Mifflin, 1928. Townsend, P. The family life of old people. London: Routledge and Kegan Paul, 1957. Westermarck, E. A history of human marriage. Allerton, 1921. Wood, V. I. Patterns of role change and life styles of middle-aged women. Unpublished Ph.D. dissertation, University of Chicago, 1963. Woolsey, T. D. Estimates of disabling illness prevalence in the U. S., based on the current population survey of February 1949 and September 1950. Public Health Service Publication 181. Washington, D. C.: Govern ment Printing Office, 1952. APPENDIX A NEWS RELEASE FOR OBTAINING SUBJECTS NEWS RELEASE FOR OBTAINING SUBJECTS* Widowhood can mean a lonely, difficult time for the woman trying to adjust to a new lifestyle because society pays so little attention to her needs. To respond to those needs, the University of Southern Califor nia's Department of Psychology has established a research project in which widows are asked to enroll. Small group discussions with widows of all ages, regardless of how long they have been widowed, are planned. Ninety women are being sought for free discussion groups, to be held on the USC campus. The project will be directed by Ms. Carol Barrett of Alhambra, who is working on her doctoral degree in clinical psychology. Six groups will meet for two-hour sessions in the late- afternoon on weekdays or on Saturdays for an eight-week period, beginning this month. "The project was conceived because widowhood is a stressful period to most women and there are very few resources, personal or social, for coping with this stress," explains Ms. Barrett. She notes that the needs of the 10 million widows in the U.S. have been relatively ignored. The project is being conducted under the guidance of Dr. Albert Marston of Beverly Hills, professor of psychology; Dr. Judith Friedman of Los Angeles (90066), assistant professor of sociology; and Dr. Steven Frankel of Palos Verdes Estates, associate professor and director of clinical training in psychology. "The program will compare three different types of group discussion, each of which has shown considerable promise as an approach to personal growth in other research," says Ms. Barrett. *News release prepared by Lynne Jewell of the University of Southern California News Bureau. 165 Each group is designed to facilitate group discussion and sharing of common experiences, and problems. One group will focus on the development of friendships, a second on possible roles of women in the society and a third, on sharing solutions to specific problems of widowhood. Anyone interested should send a card or letter as soon as possible to Ms. Carol Barrett, P. R. S. C., 734 W. Adams Blvd., Los Angeles, CA. 90007, indicating name, number of months or years widowed, age, address and telephone number. Originally from Longview, Wash., Ms. Barrett did her under graduate work in psychology at the University of Washington. She earned her master's degree in psychology from USC. She currently is an intern in clinical psychology at the Los Angeles County-USC Medical Center. Her husband, Elwin, is working on his doctoral degree in social work at USC. August 8, 1973 166 APPENDIX B INITIAL LETTER TO PARTICIPANTS UNIVERSITY OF SOUTHERN CALIFORNIA Psychological Research and Service Center 734 West Adams Boulevard Los Angeles, California 90007 Department of Psychology (213) 746-2287 Dear Widowed Friend: Thank you very much for your interest in the Widowhood Project at U.S.C. I am looking forward to meeting with you in one of our small discussion groups. I believe that this kind of combination research and service project is very important, in that the participant gains from the group experience and the researchers obtain important data for planning future programs. The Widows' Groups will meet for two hours each week for eight weeks on weekday afternoons or on Saturday. The first meeting will be an Orientation Meeting where you will find out more about the Project, meet the group leaders, and fill out some research forms. Your Orientation Meeting is scheduled for __________________________ from ________ to ________ . It will be held in Room 101, Kerchkoff Hall, 734 West Adams Blvd., Los Angeles. (See the enclosed map for direc tions.) Parking may be available in the parking lot behind the build ing on 27th St. Because so many widows would like to participate in the Project, and only six groups are currently available, we have decided to have another group later this fall for those who cannot be immediately placed in a discussion group. But it is important that all those interested attend the Orientation Meeting. At that time we will compare schedules and find out which groups you would be able to attend. After the Orientation Meeting we will write and let you know which group you have been assigned to. I look forward to seeing you, and hope that together we can examine some of the difficulties you face in your widowhood and find ways of helping other widows with similar situations. If for any reason you cannot attend the Orientation Meeting as scheduled, please call me at home in the evening as soon as possible (_________ ) . Or, you may leave a message for me by calling ________ during the day (on or after Wed., Aug. 29). Sincerely, /s/ Carol Barrett 168 APPENDIX C GENERAL CONSENT FORM GENERAL CONSENT FORM I understand that as a participant in the U.S.C. Widowhood Research Project I will be attending a weekly two-hour small group discussion concerned with the experiences of widows, for seven weeks following the initial orientation meeting, and that all information I provide will be for research purposes only and will be kept confidential. I further understand that a $15.00 registration deposit is required, and that it will be returned to me in full upon completion of the final research forms approximately three months after the last group discussion meeting, provided I do not miss more than one group meeting, unless I am ill. Signature: ___________________________________ Date: 170 APPENDIX D LETTER TO WAITING LIST CONTROLS UNIVERSITY OF SOUTHERN CALIFORNIA Psychological Research and Service Center 734 West Adams Boulevard Los Angeles, California 90007 Department of Psychology (213) 746-2287 Dear Thank you for your participation in the orientation meeting for the widowhood research project at U.S.C. last week. Because so many women would like to attend a widows' discussion group, we have been unable to accommodate everyone at this time, but a group will be available to you later this fall. Your first group meeting will be on Saturday, October 27 from 12:30 to 2:30 P.M. in Room 101, Kerchkoff Hall, 734 W. Adams Blvd. in Los Angeles. I look forward to seeing you then! Sincerely, /s/ Carol Barrett Carol Barrett 172 APPENDIX E SUGGESTED WIDOWS' CONSCIOUSNESS-RAISING TOPICS SUGGESTED WIDOWS' CONSCIOUSNESS-RAISING TOPICS 1. Are women a minority group? 2. Are you still a wife? 3. Masculine/feminine: What do they mean? 4. Does widowhood oppress women? (How are you limited legally, politically, economically, socially, psychologically, physically?) 5. Sexuality among widows. 6. Stereotypes of widows (What is a widow?) 7. Should (can?) widows remarry? (Is remarriage a solution?) 8. Famous widows. 9. Do women become "obsolete"? (What difference does your age make?) 10. The widow as mother/father. 11. The potential of widowhood. 174 APPENDIX F GENERAL INFORMATION FORM GENERAL INFORMATION FORM Some of the following questions are personal, but we need this informa tion in order to study widowhood. Your answers will be kept strictly confidential. SUBJECT NUMBER DATE 1. Marital status: Widowed Divorced___Single Married 2. The widows' discussion groups will meet at the following times. Please put an "X" beside any group it will be impossible for you to attend. Put a "1" beside the time which is your first choice; and a "2" beside your second choice. Wednesday, 1:00-3:00 PM Saturday, 10:00-12:00 PM Friday, 1:00-3:00 PM Saturday, 2:00-4:00 PM Friday, 4:00-6:00 PM 3. If there is anyone whom you would like to be with in a discussion group, please indicate her name: __________________________________ 4. Do you have transportation available for the discussion groups? Yes No____ Some women may have difficulty with transportation to the USC campus. May I give your phone number to a widow in your geograph ical area so that she can talk to you about the possibility of sharing car rides and transportation costs? Yes No____ 5. What is your age? _____ 6. How many children do you have? _____ For each of your children, please check whether they are sons or daughters, and indicate where they live and how often you see them. 176 Daughter Son He or she lives: A. In same home with me A. Nearly every day B. In my neighborhood B. About once a week C. In the same city C. About once or twice a D. In Southern Calif. week E. Elsewhere D. Several times a year __E. Once a year or less A. A. B. B. C. (see above) C. (see above) D. D. E. E. A. A. B. B. C. C. D. D. E. E. A. A. B. B. C. C. D. D. E. E. A. A. B. B. C. C. D. D. E. E. A. A. B. B. C. C. D. D. E. E. 7. How many grandchildren do you have? 177 8. On the average, how often do you see your grandchildren? Nearly every day ___ Several times a year About once a week ___ Once a year or less About once or twice a month 9. List the relatives whom you see once a month or more, and indicate by circling a number on the scale how well you get along with them. Relationship (e.g., your sister, I Get Along With Her/Him: son, niece, mother, etc.) Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well Not at Extremely all well 1 2 3 4 5 6 7 well 178 10. Do you live in: A house that I own A house that I rent The home of one of my chiliren An apartment A retirement apartment or home A hotel A rented room in a house Other (Please describe): 11. Do you live: Alone With my children With married children With my sister or brother With a roommate With my parents Other (Please describe): 12. What city do you live in? 13. With which ethnic group do you identify yourself? White Black (Negro) Chicana (Mexican American) Oriental (Asian American) Other (Please specify): 14. Your religion: 15. How important is religion in your life? Not at all important 1 2 3 4 5 6 7 Extremely important 179 16. How often do you attend a church or synagogue? Hardly ever Special occasions only About once a month About once a week or more 17. How often did you attend a church or synagogue before you were, widowed? Hardly ever Special occasions only About once a month About once a week or more 18. How often do you listen to religious services on the radio or television? Hardly ever Special occasions only About once a month About once a week or more 19. Do you belong to any activity groups through your church or synagogue? Yes No____ 20. In what ways is your religion important to you? (Check those which apply to you.) It gives me hope and courage It provides a group of people I can be comfortable with It gives me a chance to meet new friends It makes me feel part of something permanent It helps me like myself better It gives me something to do It gives me a chance to help others and to feel useful It helps me stay out of trouble It gives me something to think about Religion is not very important to me It is important to me in other ways (Please describe) : 21. Are you very active in political affairs? Yes No 180 22. Did you vote in the presidential election last fall? Yes No 23. Have you worked for any political candidates in the last two years? Yes No____ 24. My political opinions tend to be: Conservative Moderate Liberal Radical 25. How many groups composed only of women have you belonged to in the last five years? None Several Many 26. Have you participated in a women's liberation or consciousness- raising group? Yes No____ 27. How has your health been in the last five years? Excellent health Good health Fair health Poor health 28. Has your health limited your activities? Yes No____ (If yes, please describe: ) 29. Have you passed the menopause or "change-of-life" period? Yes No Currently experiencing this change____ 30. Is there anyone in particular you confide in or talk to about yourself or your problems? Yes No____ 181 31. In general, when you have personal problems, who do you usually talk to? (Put a "1" by the person you are most likely to talk to, a "2" by the person you are next most likely to talk to, etc. Leave the items blank that do not apply to you.) No one in particular Minister, rabbi or priest Women friends Men friends Neighbors Relatives (Who? ______________________________________________ ) Psychotherapist Social worker or welfare worker Other (Who? ___ ) 32. If your husband were living how likely is it that you would talk to him about your personal problems? Extremely likely 1 2 3 4 5 6 7 Not at all likely 33. Have you ever been hospitalized for psychiatric reasons? Yes No____ 34. Do you have a current driver's license? Yes No____ If not, did you ever have one? Yes No____ 35. Do you own or have access to a car? Yes_____ No___ 36. How long were you married to your previous husband? years months 37. How did you feel about your marriage as you remember it? (Circle a number on the scale which represents your feelings.) Extremely happy 1 2 3 4 5 6 7 Extremely unhappy 38. Was this your only marriage? Yes____ No____ 182 If you have been married more than once, please indicate the following: Approximate Approximate Date Begun Date Ended Ended by Death of husband Divorce Death of husband Divorce Death of husband Divorce 39. In general, what do you think about marriage? First marriage Second marriage Third marriage 1 2 3 4 5 Being married is the best thing that can happen to a person 40. How long have you been widowed? 41. Did your husband pass away: 6 7 Being married is the worst thing that can happen to a person years months Very suddenly Somewhat unexpectedly After a long illness 42. Had you ever thought about the possibility of becoming a widow? Yes No 43. Did you ever expect to be a widow? Yes No____ 44. Did you have any ideas about what you would do if you were widowed? Yes No (If so, please describe below.) 45. Did you make any plans for your widowhood, such as helping to prepare a will, or finding out where records were kept, or any thing else? Yes No (If so, please describe below.) 183 46. How many widowed friends did you have prior to your husband's death? None Several Many 47. How many widowed friends do you have now? None Several Many 48. How much education have you conpleted? Graduate or professional school Some graduate school Completed college Some college Completed high school Some high school Grade school 49. Have you taken any additional courses not leading to a degree Yes No____ If so, were they Academic courses Vocational courses Recreational or "hobby" courses 50. How much education had your husband completed? Graduate or professional school Some graduate school Completed college Some college Completed high school Some high school Grade school 184 51. Your current employment: Employed full time Employed part time Retired Not employed by choice Unemployed but would like a paying job 52. If you are employed, approximately how much do you earn at your job each month? _____________ 53. Your current occupation: ________________________________________ 54. During how many of your married years were you employed? All or almost all Some None 55. What was your major occupation while married? 56. Altogether, approximately how many years have you been employed either part time or full time? __________________________________ 57. What was your husband's occupation? 58. Check the kinds of income you currently receive: Salary or wages Social security Life insurance Private pension Investment returns Welfare (e.g., old age assistance, disability or general assistance, aid to dependent children) Financial support from relatives Other (Please describe): ______________ 59. Approximately how much total income do you currently receive each month? (Include all sources of income) Under $150 a month_______________ $ 600-$ 900 $150-$300 ___ $ 900-$1200 $300-$450 ___ $1200-$1500 $450-$600 ___ Over $1500 a month 185 60. State the year your husband passed away and the total monthly Income your household received prior to his death as nearly as you can recall. Year: Under $150 a month $150-$300 $300-$450 $450-$600 $ 600-$ 900 $ 900-$1200 $1200-$1500 Over $1500 a THANK YOU! month 186 APPENDIX G REACTIONS TO WIDOWHOOD REACTIONS TO WIDOWHOOD Many of the following questions are personal, but we need this information in order to study widowhood. Your answers will be kept strictly confidential. SUBJECT NUMBER DATE: 1. How do you think of yourself as far as age goes: young, middle- aged, elderly, old, or what? _____________________________________ 2. How interesting has your life been? Circle the number on the scale which best describes your life. Not at all interesting 1 2 3 4 5 6 7 Extremely interesting 3. What do you think your health will be like in the next five years? Excellent health Good health Fair health Poor health 4. Please indicate how often you experience the following: Headaches: Frequently___ Sometimes___ Hardly ever___ Digestive disturbances: Frequently Sometimes Hardly ever Aching muscles or limbs: Frequently Sometimes Hardly ever Inability to sleep well: Frequently Sometimes Hardly ever Oversleeping:__Frequently___ Sometimes___ Hardly ever___ Poor appetite: Frequently___ Sometimes___ Hardly ever___ 5. How independent are you? Extremely dependent 1 2 3 4 5 6 7 Extremely independent 188 6. When you think about your life, what do you think about most often? The past The present The future 7. When you think about your life, what do you think about least often? The past The present The future 8. Many people are sometimes concerned about when they will die. Can you tell us how you feel about this? If so, circle the number on the scale which best describes your feelings. 1 2 3 4 5 6 7 I am not at all concerned I am extremely concerned about this about this Many people are also sometimes concerned about how they will die. How do you feel about this? 1 2 3 4 5 6 7 I am not at all concerned I am extremely concerned about this about this We would like to know how your life now compares to what it was like when you were recently widowed. The following questions describe a variety of feelings very common in widowhood. Please answer the questions twice. The first time respond the way you would have during your first month of widowhood. Then respond as you are now. (If you have been widowed less than a month, you need only respond once.) 189 190 Recently Widowed 11. 12. 1 2 3 4 5 6 7 Extremely lonely Not at all lonely 10. 1 2 3 4 5 6 7 Frequently wanted Seldom wanted help help from friends from friends or or family family 13. 14. 1 2 3 4 5 6 7 Frequently felt Seldom felt in in a "crisis" a "crisis" 1 2 3 4 5 6 7 Felt extremely Felt no relief relieved 1 2 3 4 5 6 7 Frequently felt Seldom felt guilty guilty 1 2 3 4 5 6 7 Frequently felt Seldom felt angry angry toward my toward my husband husband 1 2 3 4 5 6 7 Extremely lonely Not at all lonely 1 2 3 4 5 6 7 Frequently want Seldom want help help from friends from friends or or family family 1 2 3 4 5 6 7 Frequently feel Seldom feel in in a "crisis" a "crisis" 1 2 3 4 5 6 7 Feel extremely Feel no relief relieved 1 2 3 4 5 6 7 Frequently feel Seldom feel guilty guilty 1 2 3 4 5 6 7 Frequently feel Seldom feel angry angry toward my toward my husband husband 191 Recently Widowed 15. 1 2 3 4 5 6 7 Frequently felt Seldom felt angry angry with my with my friends friends or family or family 16. 1 2 3 4 5 6 7 Frequently felt Seldom felt angry angry toward God toward God 17. 1 2 3 4 5 6 7 Frequently felt Seldom felt angry angry at my at my husband's husband's doctor doctor 18. 1 2 3 4 5 6 7 Felt extremely Felt extremely capable incapable 19. 1 2 3 4 C 6 7 Felt intense Felt no admiratioi admiration for for my husband my husband Now 1 2 3 4 5 6 7 Frequently feel Seldom feel angry angry with my with my friends friends or family or family 1 2 3 4 5 6 7 Frequently feel Seldom feel angry angry toward God toward God 1 2 3 4 5 6 7 Frequently feel Seldom feel angry angry at my at my husband's husband's doctor doctor 1 2 3 4 5 6 7 Feel extremely Feel extremely capable incapable 1 2 3 4 5 6 7 Feel intense Feel no admiration admiration for for my husband my husband 192 Recently Widowed 20. 1 2 Frequently felt overcome with grief 3 4 5 6 7 Seldom felt over come with grief 21. 1 2 Kept very busy 3 4 5 6 7 Stayed very inactive 22. 1 2 Frequently tried to avoid thinking about my husband 3 4 5 6 7 Seldom tried to avoid thinking about my husband 23. 1 2 3 4 Frequently felt depressed 5 6 7 Seldom felt depressed Now 1 2 3 4 5 6 7 Frequently feel Seldom feel overcome overcome with with grief grief 1 2 3 4 5 6 7 Keep very busy Stay very inactive 1 2 3 4 5 6 7 Frequently try Seldom try to to avoid thinking avoid thinking about my husband about my husband 1 2 3 4 5 6 7 Frequently feel Seldom feel depressed depressed 24. How do you personally feel about being a widow? 1 2 Being a widow is a terrible experience 3 4 5 25. How do you think most widows feel? 6 7 Being a widow is a wonderful opportunity 1 2 Being a widow is a terrible experience 3 4 5 6 7 Being a widow is a wonderful opportunity 26. How much has your life changed since the time you became a widow? 1 2 3 4 5 6 7 My life has changed My life has hardly a great deal changed at all 27. How has it changed? (Please describe.) 28. Have you changed your residence or living arrangement since the time you became a widow? (Please describe.) 29. Have you changed your hobbies or recreational interests since the time you became a widow? (Please describe.) 30. Have you changed your television or reading preferences since the time you became a widow? (Please describe.) 31. How do you feel about the possibility of getting married again? 1 2 I would definitely like to remarry 3 4 5 6 7 I would definitely not like to remarry 32. 1 2 I expect to remarry eventually 3 4 5 6 7 I expect to stay a widow 193 33. How actively involved are you at this time As a mother: 1 2 3 4 5 6 7 Very involved Not at all involved As a grandmother: 1 2 3 4 5 6 7 Very involved Not at all involved As a daughter 1 2 3 4 5 6 7 Very involved Not at all involved With my friends: 1 2 3 4 5 6 7 Very involved Not at all involved With dates: 1 2 3 4 5 6 7 Very involved Not at all involved On the job: 1 2 3 4 5 6 7 Very involved Not at all involved In volunteer work: 1 2 3 4 5 6 7 Very involved Not at all involved With my religion: 1 2 3 4 5 6 7 Very involved Not at all involved In social life 1 2 3 4 5 6 7 genera y. Very involved Not at all involved 34. Is there anything you think women should know or do before they are widowed? (Please describe.) 194 35. All of us want certain things out of life. Think for a moment about what really matters in your life, about your wishes and hopes for the future. In other words, imagine your future in the best possible light. Now take the other side of the picture. Think for a moment about your fears and worries of the future. In other words, imagine your future in the worst possible light. 100 Here is a picture of a ladder. Suppose we say that the top of the ladder represents the best possible life for 90 you and the bottom represents the worst possible life for you. 80 Where on the ladder do you feel you personally stand 70 at the present time? Step number ______ (Use any number from 0 to 100.) 60 Where on the ladder would you say you stood when you 50 were first widowed? Step number 40 Where do you think you will be on the ladder five years from now? Step number 30 20 10 0 THANK YOU 195 APPENDIX H WIDOWHOOD PROJECT EVALUATION FORM WIDOWHOOD PROJECT EVALUATION FORM SUBJECT NUMBER________________ DATE____________________ Indicate the degree to which the following common stresses of widowhood were a problem to you two months ago and the degree to which your situation has improved during the last two months. 1. Loneliness 1 2 no problem 3 4 5 6 7 a big problem 1 2 3 4 no improvement 5 6 7 much improvement 2. Grief 1 2 no 3 4 5 6 7 a big 1 2 3 4 no 5 6 7 much problem problem improvement improvement 3. Finances 1 2 3 4 5 6 7 no a big 1 2 3 4 no 5 6 7 much problem problem improvement improvement 4. Legal issues 1 2 no problem 3 4 5 6 7 a big problem 1 2 3 4 no improvement 5 6 7 much improvement 5. Family 1 2 3 4 5 6 7 1 2 3 4 5 6 7 difficulties no a big no much problem problem improvement improvement 6. Health problems 1 2 no problem 3 4 5 6 7 a big problem 1 2 3 4 no improvement 5 6 7 much improvement 7. Work related 1 2 3 4 5 6 7 1 2 3 4 5 6 7 stress no a big no much problem problem improvement improvement 197 Other stresses: 8. 1 2 3 4 5 6 7 no a big 1 2 3 4 5 6 7 no much problem problem improvement improvement 9. 1 2 3 4 no problem 5 6 7 a big 1 2 3 no 4 5 6 7 much problem improvment improvement 10. Please describe any contact you may have had with members of the group outside of our regular meeting times. Indicate the first name of the other person, the nature of the activity, and the number of times this occurred. Name Nature of the Activity (Check) No. of Times Shared transportation Talked after the group Talked on the phone Visited at home Other (please describe Shared transportation Talked after the group Talked on the phone Visited at home Other (please describe Shared transportation Talked after the group Talked on the phone Visited at home Other (please describe 198 11. List the first name of anyone in your group you were acquainted with before the Widowhood Project began: ____ ____________________ 12. How much contact do you think you will have with other group members in the future? None ____________________ A Great Deal 1 2 3 4 5 6 7 13. How satisfied were you with the general focus of your widows discussion group? Extremely Dissatisfied ____________________ Extremely Satisfied 1 2 3 4 5 6 7 14. How would you describe the major role of your group leader? (Circle the letter of the statement which best describes her role.) a. Helping us to help each other with problems b. Helping us to get better acquainted c. Direct sharing of ideas d. Other (please describe): 15. How would you describe your major role in the group? (Circle the most appropriate letter.) a. Sharing my experiences as a woman b. Giving and receiving help c. Developing friendships d. Other (please describe): 16. At this point, if you could have chosen the focus of your discus sion group, what would be your personal choice? a. Development of new friendships b. Specific problems of widows c. The roles of women in society d. Other (please describe): 17. In general, how helpful has the widows' discussion group been to you? Not at all helpful _____________________ Extremely helpful 1 2 3 4 5 6 7 199 18. In general, how interesting has the discussion group been to you? Not at all interesting ' Extremely interesting 1 2 3 4 5 6 7 19. In general, how much did you like the group experience? Not at all ____________________ A great deal 1 2 3 4 5 6 7 20. In general, how much did you learn by participating in the group? Nothing ____________________ A great deal 1 2 3 4 5 6 7 21. How helpful has the group leader been to you personally? Not at all helpful _____________________ Extremely helpful 1 2 3 4 5 6 7 22. In general, how helpful have the other group members been to you personally? Not at all helpful _____________________ Extremely helpful 1 2 3 4 5 6 7 23. How well do you like the other group members in general? Not at all _____________________ Extremely well 1 2 3 4 5 6 7 24. Would you recommend a widows' discussion group like this one to a friend of yours if it were available? Yes No Not sure____ 25. If it were possible to develop an on-going organization to extend the services to the widowed in the Los Angeles area, would you like to be involved? Yes No Unsure 200 26. How would you like to be involved? (Check those which apply.) Getting additional help with the stresses of widowhood Meeting to plan and develop such an organization Helping other widowed persons on an individual basis Leading a discussion group for widowed individuals Being an administrator for such an organization Other (Please describe): I would not like to be further involved 27. If you are interested in working to develop such an organization, what in particular would you like to contribute? (e.g., what skills, resources, personal or work experience would you be will ing to share?) 28. What do you think would be the best way to arrange future discus sion groups for the widowed? (Circle "1" or "2" depending on your choice.) A. 1. Form different groups for widows of different ages 2. Combine widows of different ages in the same group B. 1. Form different groups depending on how long the person has been widowed 2. Combine recently widowed persons with those who have been widowed much longer C. 1. Form separate groups for widows and widowers 2. Combine widows and widowers in the same group 29. Do you have any additional suggestions for how to improve the program for widows? If so, please describe: 201 APPENDIX I ROLES OF WOMEN ROLES OF WOMEN SUBJECT NO._____________________ Do Not Write Here 1 2 3 4 5 6 7 0 S A DATE: The statements listed below describe attitudes toward the roles of women in society which different people have. There are no right or wrong answers, only opinions. You are asked to express your feelings about each statement by indicating whether you (a) agree strongly, (b) agree mildly, (c) disagree mildly, or (d) disagree strongly. Please indicate your opinion by writing to the left of the number the alternative which best describes your personal attitude. Please respond to every item. a = Agree strongly b = Agree mildly c = Disagree mildly d = Disagree strongly 1. I would like to marry a man I could really look up to. 2. Having children is important to me but I must have some area of work apart from my family in which I can find personal fulfillment. 3. I don't hold with the old-fashioned idea of being submissive to a man. 4. A woman can make no greater contribution to society than the successful rearing of normal, well-adjusted children. 5. A capable and trained woman has an obligation to society to use that training in a job, even if she has a family. 6. A working mother can establish just as strong and secure a relationship with her children as can a stay-at-home mother. 7. If I would have to give up my education and get a job so that my husband would be able to continue his education, I would feel envious and resentful that he should be doing things I've always wanted to do. 203 8. Marriage— more than anything else that can happen to me— will make me certain of who I am, of what direction my life is taking. 9. I would rather not marry than sacrifice some of my essential beliefs and ideals in order to adjust to another person. 10. I am capable of putting myself in the background and working with zest for a person I admire. 11. Obligations incurred in being a wife and mother would give me a sense of importance nothing else could. 12. It is hard for me to give up the satisfactions I get out of achieving things to be in a house all day, cooking and cleaning. 13. A woman should refrain from being too competitive with men and keep her place rather than show a man he is wrong. 14. I sometimes feel that I must do everything myself, that I can accept nothing from others. 15. I would make more concessions to my husband's wishes than I would expect him to make of mine. 16. I am not sure that the joys of motherhood make up for the sacrifices. 17. I am more concerned with my personal development than I am with the approval of other people. 18. I believe there is a conflict between fulfilling oneself as a wife and mother and fulfilling oneself as an individual. 19. No matter how successful a woman may be in utilizing her intelligence and creativity, she can never know true happiness unless she marries and has a family. 20. I become irritated when I must interrupt my activities to do something for someone else. 21. For a woman it is marriage which will give her a sense of identity, a respected place in society. 204 22. Even though women have most of the same privileges as do men, they really aren't as free to pursue their interests and self-development as men are. 23. I believe that a wife's opinion should have exactly the same bearing upon important family decisions as the husband's. 24. Women have an obligation to be faithful to their husbands. 25. Divorced men should help support their children but should not be required to pay alimony if their wives are capable of working. 26. Under ordinary circumstances, men should be expected to pay all the expenses while they're out on a date. 27. The intellectual equality of woman with roan is perfectly obvious. 28. The initiative in dating should come from the man. 29. Under modern economic conditions with women being active out side the home, men should share in household tasks such as washing dishes and doing laundry. 30. There are many jobs in which men should be given preference over women in being hired or promoted. 31. A woman should not expect to go to exactly the same places or have quite the same freedom of action as a man. 32. Sons in a family should be given more encouragement to go to college than daughters. 33. Special attentions like standing up for a woman who comes into a room or giving her a seat on a crowded bus are outmoded and should be discontinued. 34. Women should have full control of their persons and give or withhold sexual intimacy as they choose. 35. The husband has in general no obligation to inform his wife of his financial plans. 36. A woman should be as free as a man to propose marriage. 205 APPENDIX J WIDOWHOOD RESEARCH PROJECT FINAL EVALUATION FORM WIDOWHOOD RESEARCH PROJECT FINAL EVALUATION FORM I.D. # __ DATE How do you feel about your marriage as you remember it? Extremely happy 1 2 3 4 5 6 7 Extremely unhappy Indicate the degree to which the following common stresses of widowhood were a problem to you at the start of this project and at this point in time. Start of Project Now 1. Loneliness 5. Family diffi culties 1 2 3 4 5 6 7 1 2 3 4 5 6 7 No A big No A big problem problem problem problem 7. Work related stress 2. Grief 1 2 3 4 5 6 7 1 2 3 4 5 6 7 3. Finances 1 2 3 4 5 6 7 1 2 3 4 5 6 7 4. Legal issues 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 6. Health problems 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7 207 Other stresses: • ----------------- 1 2 3 4 5 6 7 1 2 3 4 5 6 7 9‘ ----------------- 1 2 3 4 5 6 7 1 2 3 4 5 6 7 10. In general, how helpful was the widows' discussion group to you? Not at all helpful 1 2 3 4 5 6 7 Extremely helpful 11. In general, how interesting was the discussion group to you? Not at all interesting 1 2 3 4 5 6 7 Extremely interesting 12. In general, how much did you like the group experience? Not at all 1 2 3 4 5 6 7 A great deal 13. In general, how much did you learn by participating in the group? Nothing 1 2 3 4 5 6 7 A great deal 14. Please describe any contact you have had with other members of your discussion group since the last formal meeting last fall. (If you have not talked with anyone, check here: ____.) Please indicate the name of the other person, the nature of the activity, and the number of times you did this. Name of Other Person Nature of the Activity No. of Times (Write "Group" for (Phone call, visit at an activity involving her home, dinner at a your whole group) restaurant, etc.) 208 15. How much contact do you think you will have with other group members in the future? None 1 2 3 4 5 6 7 A great deal 16. On this page, please describe the ways in which you or your life have changed since the start of the Widowhood Research Project. Please include changes of all kinds, positive or negative, whether they seem related to the project or not. 209 APPENDIX K SPECIAL EVALUATION FORM FOR CONFIDANT GROUPS SPECIAL EVALUATION FORM FOR CONFIDANT GROUPS SUBJECT NUMBER DATE: 1. How well do you feel you know your "new friend" from the group? Not at all 1 2 3 4 5 6 7 Extremely well 2. How well do you feel she knows you? Not at all 1 2 3 4 5 6 7 Extremely well 3. How much do you personally like your "new friend" from the group? Not at all 1 2 3 4 5 6 7 A great deal 4. How much do you think she likes you? Not at all 1 2 3 4 5 6 7 A great deal 5. How easy would it be to discuss personal problems with your new friend? Extremely hard 1 2 3 4 5 6 7 Extremely easy 6. What kind of a relationship do you have at.present? Casual acquaintance 1 2 3 4 5 6 7 One of my best friends 7. What kind of relationship would you like to have with her in the future? No relationship 1 2 3 4 5 6 7 One of my best friends 211 8. What kind of relationship do you think you will have three months from now? No relationship 1 2 3 4.567 One of my best friends 212

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Weight Reduction As A Function Of The Timing Of Reinforcement In A Covertaversive Conditioning Paradigm

PDF

Death Anxiety In Leukemic Children

PDF

An Examination Of Positive And Negative Reinforcement In Classical And Operant Conditioning Paradigms In The Primary Psychopath

PDF

Modification Of Low Self-Confidence In Elementary-School Children By Reinforcement And Modeling

PDF

Attention, retention, and incentive processes in observational learning

PDF

The Relationship Of Teacher Empathy And Student Personality To Academic Achievement And Course Evaluation

PDF

Self-Sacrifice, Cooperation, And Aggression In Women Of Varying Sex-Role Orientations

PDF

The effects of physical proximity and body boundary size on the self-disclosure interview

PDF

The Effects Of Self Vs. Ideological Advocacy On The Self-Esteem And Endorsement Of Black Power Ideology Of Black College Students

PDF

Strategies Of Marital Communication

PDF

Looking Back After Coming Down: Conformity And Commitment In Campus Protest

PDF

The Effects Of Anxiety And Threat On Self-Disclosure

PDF

Prognostic Expectancy Effects In The Desensitization Of Anxiety Over Invasion Of Body Buffer Zones

PDF

The Effect Of Personalized Emotional Stimuli On Asthmatic Reactions

PDF

The Effects Of Diphenylhydantoin On The Galvanic Skin Responses Of Psychopathic And Normal Prisoners

PDF

A Multiple Investigation Of Child-Rearing Attitudes

PDF

Mental Imagery As A Function Of Muscular Tension And Suggestion

PDF

The Effects Of Repression-Sensitization, Race, And Levels Of Threat On Extensions Of Personal Space

PDF

Factors Of Adaptation And Rehabilitation In Home Hemodialysis

PDF

The Differential Effectiveness Of External Versus Self-Reinforcement On The Acquisition And Performance Of Assertive Responses

Asset Metadata

Creator Barrett, Carol Jeanne Clark (author)

Core Title The Development And Evaluation Of Three Therapeutic Group Interventions For Widows

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,Psychology, clinical

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Marston, Albert R. (committee chair), Frankel, Andrew Steven (committee member), Ragan, Pauline K. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-736675

Unique identifier UC11356658

Identifier 7501044.pdf (filename),usctheses-c18-736675 (legacy record id)

Legacy Identifier 7501044.pdf

Dmrecord 736675

Document Type Dissertation

Rights Barrett, Carol Jeanne Clark

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA