Cairn - The Shock of Widowhood on the Eve of Old Age: Male and Female Experiences

CAIRN ÉVOLUE… Apportez votre pierre !

(enquête en ligne)

Mon compte |

Mon panier d'achats |

Ma bibliographie

Contacts |

Aide

· Accès distant · Mot de passe oublié

· Recherche avancée

Population (english edition)
I.N.E.D

I.S.B.N.sans
200 pages

p. 885 à 914
doi: en cours

Veille sur la revue
Veille sur l'auteur

Acquérir ce numéro

Vous consultez

Volume 57 2002/6

Discipline > Revue > Sommaire > Article

Télécharger cet article en PDF

The Shock of Widowhood on the Eve of Old Age: Male and Female Experiences

Christiane Delbès []* Christiane Delbès, Institut National d’Études Démographiques, 133 bd Davout, 75980 Paris Cedex 20, tel: 33 0(1) 56 06 20 40, fax: 33 0(1) 56 06 21 99, Joëlle Gaymu *[**]*

Résumé de l'article

In France today nearly 4 million people are widowed. This marital status is a characteristic of the elderly but also of women: 84% of the widowed are widows.
The present longitudinal study describes the context of widowhood between ages 62 and 75 and the subsequent reorganization of life, on the basis of a comparison between widowed and married respondents. It shows how the adaptation to widowhood can be more or less difficult for men and women. The main areas of retired life (family and leisure activities) are examined. A substantial part of the study focuses on the psychological consequences of widowhood.
All the indicators suggest that the widows’ life is more difficult. They have a more negative view of life and of retirement, suffer more frequently from loneliness, and are more subject to depressive tendencies. Their lesser involvement in leisure activities and their greater social isolation show how much they are disadvantaged.
Although excess mortality following the loss of a spouse is much higher for men than for women, those who avoid death appear to adapt better than women to the loss of their spouse. Actualmente, en Francia, hay casi 4 millones de personas viudas. Esta situación conyugal caracteriza a las personas de más edad y también a las mujeres: el 84% de las personas viudas son mujeres.
Este estudio longitudinal describe el contexto en el que se ha producido el aumento de la viudedad entre los 62 y los 75 años, y la reorganización de la vida que ésta supone, en base a una comparación entre los encuestados que siguen casados y los que se quedaron viudos. El estudio muestra qué aspectos de la adaptación a la viudedad son más y menos difíciles según el sexo. Se examinan las esferas esenciales de la vida de los jubilados, la familia y el ocio. Una parte importante de esta investigación se centra en evaluar las repercusiones psicológicas de este acontecimiento.
Todos los indicadores sugieren que la vida cotidiana de las viudas es más difícil. Tienen una percepción más negativa de la vida y de la jubilación, sufren más de soledad y tienen mayor tendencia a la depresión. Su menor integración al mundo del ocio y su mayor aislamieno social son otra muestra de hasta qué punto están penalizadas.
Si la sobre- mortalidad de los hombres al enviudar es muy superior a la de las mujeres, los que sobreviven parecen adaptarse mejor a la pérdida del cónyuge.

Plan de l'article

The evolution of people’s living conditions after retirement is generally explained by reference to changes in their health, income and marital status, or to the attitudinal changes that occur across generations (see for example the article by Christiane Delbès and Joëlle Gaymu on sexual life after the age of 50 in Population, 6, 1997). Here, Christiane Delbès and Joëlle Gaymu examine the death of a spouse, a typical experience of old age, and its material, social and psychological effects on the surviving spouse. They employ the French “Transition from Working Life to Retirement” survey that is particularly interesting because of its longitudinal design, as the same persons were interviewed at age 62 and again at age 75. They show that the effects and experience of widowhood are different for men and women. When widowhood occurs, men receive more support from their family than women. The attitude of withdrawal to the domestic space with advancing age is more characteristic of widows than of married women, and much more of widows than of widowers.

The population of France currently includes nearly four million widowed people. This marital status, which is very rare among young adults, increases rapidly after 60. At that age, less than one person in ten has lost a spouse, but at 75 the proportion is one in three. This situation is therefore a characteristic of older persons, but also of women: excess male mortality and the age difference at marriage combine to increase the risk for women of losing their spouse. At 75, widowhood affects 42% of women against only 11% of men. In all, 84% of widowed persons are widows.

Widowhood is not an irreversible situation. But the chances of ﬁnding another partner differ greatly for men and women. A woman who becomes widowed at age 45 has the same probability of being remarried ten years later (7%) as a man widowed at 65. Even if they lost their husband only in their early ﬁfties, widows hardly ever start a new life [1]. The very low proportion of women who form a new relationship outside marriage conﬁrms this observation.

These few ﬁgures show that widowhood is primarily a female experience. But if this event is an almost inevitable stage in the life cycle of women, do they necessarily adapt to it more easily? Compared with men, women in the currently retired generations have invested more of themselves in family and home life and less in the professional sphere. What repercussions does this division of roles have for the experience of widowhood by men and women?

Studies on the consequences of widowhood from a gender perspective are rare, especially in France, and are often based on period data. The longitudinal nature of the Transition from Working Life to Retirement Survey (see box) enables us to describe the context in which widowhood occurs and the subsequent reorganization of life by comparing respondents who are still married with those who have been widowed (Paillat et al., 1989). A total of 1,487 individuals, all born in 1922, were interviewed at age 62, in 1984; at age 75, in 1997, 945 of them were interviewed again [2]. During the ﬁrst round of interviews, more than a quarter of the women (25.8%) were already widowed — a situation that was extremely rare among men (2%). Thirteen years later in 1997, close to half of the women and 11% of the men had been widowed for an average of around ﬁve years. Between ages 62 and 75, the death of a spouse is the main cause of change in marital status. Divorce and formation of new unions are rare (1% for each event).

How do these retired persons, the great majority of whom have only ever lived in one union, reorganize their life following the death of their lifetime companion? We ﬁrst examine the effects of widowhood on their health, which is a key indicator of the lifestyle of retired persons and moreover, highly sensitive to changes in personal environment. We then consider the economic consequences of widowhood, and examine whether widowed persons compensate for the recent loss of their partner by investing more in family relations or leisure activities. Finally, we explore the extent to which this event is associated with feelings of loneliness and boredom and more generally with psychological distress.

Longitudinal survey on the Transition from Working Life to Retirement Methodological aspects

This survey was carried out in France, initially between 1981 and 1984, on a representative sample of private sector wage earners, who make up about two-thirds of the labour force. Its aim was to measure the changes that retirement caused in the life of each respondent. A total of 1,487 respondents from the cohort born in 1922 were interviewed face-to-face on three occasions, at ages 59 (in 1981), 60 (in 1982) and 62 (in 1984). In 1997, 945 respondents, corresponding to 78.9% of the survivors, participated in a fourth wave of the survey carried out by postal questionnaire.

In general, non-respondents introduced no significant bias as they did not differ much from the general population. It should be noted, however, that a slightly higher response rate was obtained from male executives or managers (“cadres” [a] who were arried and/or in good health.

Between 1984 and 1997, 24.4% of males and 11.4% of females died, a slightly lower umber of deaths than expected given general mortality levels. But the respondents were drawn from the file of employees contributing towards a pension of the National Centre or Old Age Insurance, CNAV. At any given age, persons in the labour force are on average in better health than those who are not. Furthermore, the results of this survey confirm the high excess male mortality, the protective effect of marriage, and the social inequality in mortality. After the respondent’s sex, the state of health declared in 1984 as found to be the best predictor of mortality.

Distribution of the sample by marital status, 1997

IMGIMG Still married Widowed Never married...IMGIMF

For men, the sample’s distribution by marital status is representative of the general population. On the other hand, among women, the married are underrepresented because the sample was drawn from the economically active population aged 59, and single or divorced women are more often in the labour force. Moreover, the lower social strata are over-represented among the widowed, particularly of the female sex. The comparisons between widowed and married persons presented in this study are therefore based on logistic regression methods, which allow us to control for this difference in socio-occupational structure. In addition to socio-occupational category, logistic regression takes into account other variables likely to introduce bias into the comparison (health status, number of children, quality of relationship with children, changes in income, perception of changes in the quality of life, number of family and social gatherings).

The category which in the French classification includes not only managers and executives, but also senior civil servants and members of the liberal and higher intellectual professions. For convenience, it will be translated as “managers”.

I. Health status

1. Reported state of health [3]

The deterioration of the state of health with age, so often observed in previous studies, is obvious: between the ages of 62 and 75, the proportion of respondents who report being in good health decreases from 64% to 38%. At 75, however, few complain strongly about their state of health: only 3% consider it bad and 10% poor. Among women, the loss of a spouse does not affect the pattern of change in the state of health: at 75, equal proportions of widows and married women report being in good health. On the other hand, the health of widowed men tends to deteriorate a little more, and at age 75 they report worse health than married men (36% and 41% respectively reported being in very good or good health, 19% and 12% respectively in poor or bad health). On these two points, however, the differences were not statistically signiﬁcant.

These results are corroborated in the literature. Most researchers have observed widowhood to have either no negative effect on perceived state of health (Heyman and Gianturco, 1973; Lichtenstein, 1996) or an immediate effect that disappears rapidly (Fenwick and Barresy, 1981; Ferraro, 1985-86; Murrel et al., 1988). And yet, these ﬁndings are surprising. Excepting violent deaths, the excess mortality of widowed persons (Desplanques, 1984; Thierry, 1999) should be reﬂected in a deterioration of their health during the period preceding their death. Certainly, perceived state of health is a good predictor of mortality, but this indicator may not be sensitive enough to reveal small differences between groups that often contain few subjects.

Finally, this study produces some very familiar results: among widowed and married persons, women and manual workers more often perceive themselves to be in bad health (Sermet and Grandjean, 1998).

2. Health problems, chronic illness or disability

Another illustration of the decline in health with aging is the rising frequency of reports of health problems, chronic illness or disability. While fewer than half of the respondents mentioned such conditions at 62, six in ten did so at 75.

Loss of a spouse during the period does not increase the probability of observing such loss of ability. At 75, this is equally frequent among widowed and married persons. In general, studies that compare recently widowed and still married persons also conﬁrm the small effect of the loss of a spouse on physical symptoms (Madison and Viola, 1968; Clayton, 1979).

3. Consumption of medicine

Between ages 62 and 75, the number of persons taking medicine regularly increased from ﬁve in ten to eight in ten. The average number of medicines taken also increased. The proportion of persons following a medical treatment did not rise more among persons who became widows than among those who remained married, but the former stand out by a larger increase in their intake of medication. At age 75, widowed persons take 4 medicines daily against 3.5 for married persons, while at 62 both groups were taking the same number of medicines.

This result may at ﬁrst seem paradoxical because married or widowed persons are indistinguishable in terms of their physical health. We discuss later the possible association between this situation and a higher incidence of psychological problems among widowed persons.

Widowhood, therefore, does not have a major inﬂuence on health. What about economic circumstances, another variable with a strong inﬂuence on lifestyle?

II. Income

If retired persons nowadays have a standard of living similar to that of the rest of the population, with young pensioners being even among the most afﬂuent members of French society, we know also that widows, and especially the oldest among them, live sometimes in extremely precarious economic conditions (Coëfﬁc, 2000). Many widows have to make do with a reversion pension, and even when they have acquired their own pension rights, the pension they receive is inevitably small given their shorter working life, less skilled jobs, and their lower wages for the same job.

Between 62 and 75, both married and widowed persons experienced a decline in their income (Table 1). The decline over this period is explained by the retirement of respondents and/or their spouse, by higher taxes on pensions, and by reduced income from capital as a result of legacies and gifts to children [4].

Table 1
Monthly income of respondents’ households, 1984-1997 (in constant 1997 francs)

IMGIMG Men Women 62 75 Change (%) 62 75 ...IMGIMF

Widowed persons experienced a greater decline in resources than married persons. This decline is larger for widows than for widowers. For men, the loss of a spouse produces a relatively small fall in income (- 15.4% against - 9.7% for persons still in union), since less than one in ﬁve men had a spouse who had worked full-time for more than thirty years. Furthermore, in these generations, the woman’s earnings were often no more than a supplementary income. From the woman’s point of view, widowhood is accompanied by a much larger reduction in income (- 46.8% against - 24.2% among women still married).

A result of these differential changes is that widows are far less well off than widowers at 75 (with respectively FF 6,996 per month for the former against FF 10,144 for the latter, in 1997). Unlike widowers, widows all have two sources of income (their own pension and a reversion pension) but they were more often in couples belonging to less favoured social categories. In addition, even if their reversion pensions are higher, they are at a disadvantage because their own pensions are much smaller.

To compare standards of living, we need to take household size into account. Widowers are at an advantage since at 75 they control 86% of a married couple’s income. This proportion is only 57% among widows [5], corresponding to a standard of living just comparable to that of women in a couple [6]. The relatively easy circumstances of widows in this sample are explained by the fact that all of them had been in the labour force.

The low impact of widowhood on the standard of living is conﬁrmed by the respondents’ own assessment. About 40% of men and women, married or widowed, said they were “a bit tighter” ﬁnancially than at the start of retirement. The impression of being more or less materially comfortable depends on resources but also on needs. If widowers do not perceive themselves as different from the others even though they have a better ﬁnancial situation, it is perhaps because, after the death of their spouse, they make much greater use of services.

Thus widowhood is not, on average, synonymous with economic impoverishment. Nevertheless, it occurs in a context of health decline associated with aging and a contraction of the social universe associated with the end of the working life. How far can widowed persons rely on their family for assistance in adjusting to their new life? How much does coping with this frequently traumatic event depend on family support strategies, comparable to those used in the event of unemployment or a child’s divorce?

III. Social ties

1. Seeing the family

At 62, respondents saw another member of their family 15.5 times a month on average, but with age these contacts have become less frequent. At 75, their number is only 12.8 a month. The main reason for the reduction in intensity of family life is the erosion of relations with members of the family other than the children, as a result of the death of some and the maturation of grandchildren into adolescents or adults who are more interested in their friends than in their grandparents. Does the loss of a spouse limit this erosion, or on the contrary, do the relationships built around the parental couple tend to decline when it has been broken by widowhood?

Between 62 and 75, regular family relations were better maintained among respondents who became widowed than among those still living as couples [7]. Among widows, the ties weakened less, and widowers, in contrast to almost all other retired persons, experienced a resumption of con-tacts [8]. Since the family environment did not change in markedly different ways for married and widowed persons, the reason must be sought in the willingness of family members to help them cope with the ordeal.

Consequently, at age 75, persons widowed during the last thirteen years see their family more often than respondents who are still married [9]. Widowers see their family more often than widows. In the sub-population of grandparents aged 75, for example, widowers meet their family 17.8 times per month and widows 16.1 times, whereas for married men and women only 14.8 visits are reported (Table 2). The rare longitudinal studies of the effects of widowhood on relations with children have produced contradictory results. For Ferraro and Barresi (1982), widowed persons have more difﬁculty than married persons in maintaining their social contacts as they grow old, whereas others (Wan and Odell, 1983; Roan and Kelly Raley, 1996; Cavalli et al., 2001) observe the opposite, at least in the short run.

At 75, widowed persons visit their children and grandchildren more often. For visits to children, the contrast with married persons is greater for men (for women the difference by marital status is not signiﬁcant). Even if widowers are in better health than widows, the traditional domestic division of labour in these generations leaves them ill-prepared to cope with daily existence after the death of their spouse.

Table 2
Number of monthly encounters for grandparents, by marital status

IMGIMG Men Women 62 75 62 75 Still marr...IMGIMF

Conversely, as regards relations with grandchildren, the differences are greatest among women. Having more spare time than married women and still enjoying relatively good health, widows become much more involved with their young grandchildren. Their own children are delighted to leave them in their care as a way of breaking down their isolation. In addition, when grandchildren become adolescents, they continue to see their grandmothers more often as a result.

The effect of widowhood on relations with other members of the family varies by sex. It stimulates contacts for men and reduces them for women. Since women are known to be responsible for maintaining family sociability (Héran, 1988), these results are quite surprising. No doubt male widowhood is sufﬁciently uncommon that it prompts special attention from the family circle.

Finally, when the death of a spouse occurred long before [10], there is no difference between widows and married women. Whereas women who lost their spouse before 62 enjoy a high level of family support (20.1 visits against 18.6 for married women), they later experience a more rapid decrease in their number of contacts. Over time, their situation ceases to be exceptional. At 75, married and widowed women have the same number of visits (around 14.5). However, with the death of their husbands and the consequent reduction in social contact, should visits not have been less frequent than for married couples? One may therefore conclude that the similarity between the situations reﬂects efforts by offspring to limit the isolation of their widowed mother.

2. Mutual help

One may also ask if mutual help, which is another illustration of the bonds between parents and children, is affected by the death of a father or mother.

Help provided to children

In general, one would expect married life to stimulate the help provided to the family circle. By freeing the other from certain tasks, each spouse provides more time for helping their children [11]. And it is a fact that widowed persons provide less assistance to their children [12], as numerous surveys have conﬁrmed (Eggebeen, 1992; Aquilino, 1994; Lee et al., 1998). But the involvement declines much more when the surviving parent is the father. At 75, widowers are almost twice as likely as married men to provide no assistance, whereas there is no signiﬁcant difference in attitude between widows and married women (Table 3).

The negative effects of being a surviving partner are most marked in the matter of having the children and grandchildren during the holidays, and looking after the grandchildren. This is also where the behaviour of married and widowed men differs most (Table 4), reﬂecting the key role played by women in family sociability. Moreover, given the difﬁculty men often have in coping with their own daily upkeep, how could they also help their children? Consequently, if the proportion of respondents not giving any help to their children rose from less than 9% at 62 to more than 18% at 75, the falling off among widowed men is on a completely different scale: 4.6% of them were not giving any assistance at 62, and 30.8% thirteen years later. By contrast, among women who became widowed between 62 and 75 the decline in help for their children is no greater than among those still married.

Table 3
Proportion of respondents reporting neither helping nor being helped, at ages 62 and 75, by marital status (%)

IMGIMG Not helping their children Not bein...IMGIMF

Help received by the parents

With age, more respondents look to their children for help in daily life. At 62, about one in three men and one in four women was receiving no assistance. Thirteen years later the proportions are only 26.4% and 16% respectively.

But persons who became widowed, whatever their sex, beneﬁt more than others from this strengthening of ﬁlial support. At 62, retired persons about to lose their spouse do not differ in any respect from those who will remain married, while at 75, after the death of the spouse, only about half as many are receiving no help. Women widowed before 62 were already receiving close attention from their children. The latter’s support continued at a high level, because at age 75 these women are no different from recent widows. Most studies conﬁrm the higher support given by children to a surviving parent (Stoller and Lorna, 1983; Eggebeen, 1992; Lee et al., 1998). A consequence of this intensiﬁed concern of children towards their widowed parents is that at 75, widowers receive more assistance than married men and widows slightly more than widowers.

If, overall, the kind of support received is the same for widowed and married persons, marital isolation elicits particular attention from children in certain areas, especially towards their mother. Being driven in a car, for leisure activities or shopping, is the most frequent form of help, but is much more widespread for women who have recently lost a spouse [13] (Table 4). Few women in those generations hold a driving licence and they become dependent on their children once they are widows. In addition, more widows spend their spare time with their children. At 75, they are two to three times more likely than married women to go to live with their children temporarily or for a holiday, and one and a half to two times more likely than widowers. Finally, their children help out more frequently when they are ill. As for widowers, they call on their children for help with household tasks more often than women.

Table 4
Respondents being helped by their children, or helping them, at ages 62 and 75, by type of help (in %)

IMGIMG Help from children Help to children...IMGIMF

The balance of exchanges

The study shows the importance of mutual help within the family. At 75, only 8% of men and 4% of women are not part of any exchange. Similarly, 65% of the former and 71% of the latter reported participating in a network of reciprocal solidarity.

At 62, respondents giving help clearly outnumbered those receiving it. Thirteen years later, the swing in the other direction is marked. If at 75 married persons are still more often givers than receivers of help, among widowed persons the balance of exchange is reversed. They are roughly twice as likely to provide no assistance as to receive no assistance. Compared with their married counterparts, widowers are more likely to stop giving help while widows more often receive help.

Strong links exist between parents and their adult children. Of course, with the passage of time since they lived under the same roof, visits become slightly less frequent, but simultaneously the children help their parents more. The practice of solidarity is particularly intensive when their father or mother has recently been widowed, by increasing their visits and their assistance. As the initial grief fades, the frequency of visits returns to the same level as for persons still married, but the children continue to assist their widowed parent more. Nonetheless, family commitment takes different forms for widowers and widows. Men receive more visits and domestic assistance; women are helped to maintain outside contacts. This result clearly illustrates the specialization of roles when the spouses were alive. For women, widowhood leads to difﬁculties in maintaining outside contacts, and for men, to problems with running a household.

Contact with the family represents the main element of social life for retired persons, with three out of four persons they meet being family members. Friends make up the second pillar. We now look at this second pole of their sociability.

3. Friendships

At 75, just over one in ten male and female respondents reported having no friends or never seeing any, with negligible difference by marital status.

It is generally accepted that friends are made mainly during youth and working life, and that with age the friendship network contracts through the death of some members. If this is indeed the case for married persons, especially for men, it would seem that widowed persons succeed in forming new friendships which more than compensate for the losses (Table 5).

Table 5
Relations between friends at 62 and 75, by marital status

IMGIMG Men Women 62 75 62 75 Proportion ...IMGIMF

In addition, if the number of encounters with friends increased for everyone between 62 and 75 years — in contrast to family contacts — widowers and especially widows have turned more than others to their friends. For example, women who have lost their husbands since the previous survey visit their friends 2.4 times more often; among married women this coefﬁcient is only 1.8.

At 75, respondents see their families less often than at 62, but they see their friends more often. However, this greater contact with friends does not compensate for the weakening of family ties, with the result that retired persons become more isolated. At 62, they were seeing a close relative 19.2 times per month on average against 17.8 at present, with women suffering the larger decline in social life. In this general trend towards an erosion of social relations, the recently widowed are the only ones whose sociability increases. Among men, the whole circle of close relatives is more mobilized, whereas among women the reduced contact with the extended family is compensated by an increased presence of friends. Widowers thus ﬁnd themselves, at 75, with more social relations than couples. And although women, throughout their lives, have more than men maintained relations with family and friends, they appear not to reap the beneﬁts in terms of visits. When they have been widowed, they receive fewer visits than men (21.5 and 18.9 respectively).

IV. Leisure activities

In the Transition from Working Life to Retirement Survey, individuals were questioned about a wide range of leisure activities. Given the predominant position of television, radio and reading in the lives of retired people, the media are analysed separately. So too are holidays, which even in the absence of the constraints of working life remain a privileged means of escape, and a break from daily life.

1. Manual, artistic, cultural or social activities and sports

Over a period of thirteen years, the respondents have largely withdrawn from the world of leisure activities. At 75, 10% of them (slightly more women than men) have no activities of this kind, a situation that was almost non-existent at 62 for men and very rare for women. Even if the decline in the proportion of those participating in leisure activities varies in a ratio of one to two depending on the type of leisure activity, the order of preference remains the same. The most popular activities at 75 are the same as at 62, and are those that have least decreased in popularity. Top of the list are manual activities, followed by cultural activities or sports, social activities, and far behind by artistic activities (Table 6).

Table 6
Proportion of respondents involved at ages 62 and 75, by type of leisure activity and marital status (in %)

IMGIMG Men Women Activity 62 75 62 75 M...IMGIMF

There is no straightforward relationship between changes in marital status and involvement in the various leisure activities. Widowhood causes men and women to abandon manual activities (especially among retired managers and persons suffering from ill-health) and artistic activities but is favourable to a continued participation in sports [14]. Furthermore, marital isolation increases the likelihood that women will abandon cultural activities and decreases the likelihood that men will abandon social activities [15].

As a consequence of these contrasting trends, the range and frequency of the activities undertaken by 75-year-olds vary by marital status. Widowed and married men aged 75 participate in sports in equal numbers. On the other hand, signiﬁcantly more married men aged 75 practice manual activities. Life with a spouse therefore seems to favour an investment in the home, though with this survey data it is not possible to know whether more married individuals live in individual dwellings, as is the case in the general population.

The effect of widowhood on women is a little different. At 75, whether married or widowed, manual activities are their favourite pastime. No doubt they are continuing to knit and sew for their children and grandchildren. Widowed women are more likely than married women to participate in sports, but they also have fewer cultural activities [16].

Finally, widowers and widows are much less likely than married people to have an artistic hobby [17].

Irrespective of marital status, participation rates are strongly correlated with health. While this would appear self-evident in the case of activities requiring a physical effort, it is also true, albeit to a lesser degree [18], for cultural, artistic and social activities.

Beyond the simple participation rates, the frequency with which the various activities are practised conﬁrms that pensioners reduce their involvement as age progresses. The younger retirees take part in a leisure activity (excluding the media) 17 times a month (Table 7), compared with only 12.2 times a month for 75-year-olds. The trend is the same for men and women. Advancing age seems to have little effect on the participation in cultural activities, but sporting, and artistic, manual and social activities are abandoned as people get older.

In addition, the behaviour following the loss of a spouse is radically different among men and women. Men widowed between 62 and 75 remain more involved in their leisure activities than their married counterparts, and at 75, they are more assiduous, and participate in a leisure activity 14.3 times a month compared with 12.8 times for married men. In contrast to all the other respondents, they increase their social activities and have given up on sport less often. Among the women, on the other hand, widows reduce their involvement in leisure activities. They have lost interest more than married women in manual activities and sports and have not kept up the practice of cultural outings as married women do. In these generations women rarely participated in any leisure activities without their husbands, who were probably the main driving force.

Table 7
Monthly participation in leisure activities at 62 and 75, by marital status

IMGIMGStill married Men Women 62 75 62 7...IMGIMF

These contrasting trends indicate reactions and modes of adaptation that are strongly inﬂuenced by gender - a culture of leisure activities in response to loneliness among men, withdrawal to the domestic sphere among women. Furthermore, this withdrawal by women following widowhood is probably even greater in the general population. In fact, all the women in the survey sample had been economically active, and participating in paid employment is a factor in the development of an independent outlook.

2. Television, radio, newspapers and books

Television accounts for a large proportion of the free time of 75 year-old retired persons. This group spends on average 3 hours and 20 minutes per day in front of the screen (Table 8). Men and persons widowed between the ages of 62 and 75 watch a little less television, while women widowed before age 62 spend by far the most time watching television. In contrast with most other leisure activities, time spent watching television increases with old age. Although this change is part of a broader trend - for the population of France as a whole, time spent watching television rose by two hours a week between 1989 and 1997 (Donnat, 1998) - it is ampliﬁed by old age. The increase in time spent watching television is slightly longer among women and married individuals.

Half as much time, or slightly over 90 minutes per day, is spent listening to the radio. As in the case of television but with more pronounced contrasts, men and persons widowed between 62 and 75 devote the least time to the radio. This medium is most popular among women who have been widowed a long time. Moreover, as respondents get older they reduce slightly the time spent listening, with the exception of those who became widows before 62 and still married men.

Table 8
Time spent (minutes per day) watching television, listening to the radio and reading newspapers and magazines

IMGIMG Men Women 62 75 62 75 Time spent...IMGIMF

Reading newspapers or magazines takes up around an hour per day for men and women alike. Among men, this activity is not affected by marital status. Widows, on the other hand, spend longer on this activity than their married counterparts, especially if they lost their husband between the ages of 62 and 75. As with television, the time spent reading newspapers or magazines keeps increasing during retirement. Women and widowers are among those who most increase the time spent reading.

At 75, widows spend signiﬁcantly more time reading the press than married women. However, fewer of those widowed after age 62 read books. Among men, there is no difference between the widowers and the married men; both groups read less than women (Table 9). Moreover, although married men, widows and widowers have not changed their habits between the ages of 62 and 75, some married women or women who have been widowed a long time have begun to spend more time reading. This new interest is all the more surprising for occurring at a time when reading in general was declining in popularity (Donnat, 1998).

Table 9
Proportion of respondents who have not read a book during the last 12 months, at ages 62 and 75, by marital status (%)

It is not surprising that retired managers and administrators, both widowed and married, are the most keen on reading, but widowhood reinforces social disparities. Following the death of their spouse, some members of this group doubtless returned to reading.

3. Associations

Women of 75 who are still married are no more likely to be members of an association than 13 years before, in contrast with those who have become widows [19]. This attitude seems to be limited in time, however, since women who have been widowed for a long time are indistinguishable from married women in this respect. In contrast, both widowed and married men as they get older abandon their associations, the latter being mostly war veterans’ associations. Despite this withdrawal, the latter account for the higher degree of participation of men. At 75, ﬁve men out of ten, as against four women out of ten, are members of an association, in more than 50% of the cases a pensioners’ club.

Regardless of marital status, the likelihood of belonging to an association increases with social status for men, but no such pattern is visible for women.

4. Holidays

At 75 nearly four respondents out of ten have not gone on holiday [20] during the last twelve months. This proportion reaches a maximum of one respondent out of two for widowed men (Table 10); the same level is found among blue collar workers and persons in bad health.

At 62, an equal number of men and women were taking holidays, but 13 years later they have more often abandoned the idea of going away. At 75 and with identical socio-demographic characteristics, the proportion of men who go on holiday is always smaller, and taxing conditions (widowhood, poor health, or lower social class) accentuate the contrasts between the sexes. Thus, 39% of married men and 33% of married women have not taken a holiday in the last twelve months. In widowhood, the proportions rise to 48.9% and 34.3% respectively.

Table 10
Frequency and length of holidays at 62 and 75, by sex and marital status

The fact that widows take more holidays could be considered surprising [21], since they are less involved in leisure activities and the loss of their spouse causes a larger drop in their standard of living. However, they are more likely than widowers to go on holiday with their children. The same is true of women who are married to older spouses. This is no doubt due in part to a selection effect. Their previous participation in the labour force has certainly made them more dynamic, developed their sense of initiative and broadened their outlook.

Holiday-makers were away on average for 38 days, with the men taking longer holidays than the women. At 75, they took 42 and 34 days respectively. This difference is entirely due to the behaviour of people living alone. Widowed men are reluctant to leave home in the course of a year, but when they do, it is for longer periods: 36 days compared with only 27 days for widows (Table 10).

The difference between the behaviour of married women and that of widows is signiﬁcant. As they grow older, many married women stop taking holidays, but for those who continue the duration is the same. For married couples, taking a holiday depends on the health of both parties. The opposite pattern is observed for women widowed between the ages of 62 and 75: slightly more of them take holidays but they signiﬁcantly reduce the length of their stay (11 days shorter). These women, many of whom were until recently tied to their homes by their husbands’ ill health, are once again in a position to get away and are perhaps encouraged to stay with their families. Moreover, since widowhood is a common situation at these ages among women, they can take a holiday with friends or go on an organized tour. Their less secure economic situation is the most likely reason for the reduction in the length of time they spend away.

Widowers differ from widows in two ways. First, as they grow older, many give up on holidays [22]. (At 62, before the loss of their spouse, the proportion not taking a holiday was 27.7% as against 48.9% when 75 and widowed). Widowers, it is true, are less often invited to stay with their children, and they are perhaps less willing to participate in organized tours, where women often predominate. On the other hand, when the opportunity to take a holiday arises, widowers increase the length of their stay. This can be attributed to their greater ﬁnancial ease and their better health but also to their greater social freedom (an independent role for women in these generations being still often viewed unfavourably).

In the end, the divergent trends in the frequency and length of holidays between widowers and widows cancel out, so that at age 75 the average number of days spent away is more or less identical. Here again, widowhood has fewer negative implications for men. They are much closer to their married counterparts than are the widows (respectively 18.4 and 24.8 days of holiday as against 17.4 and 31.7).

In general, between 62 and 75 the respondents have reorganized their pastimes in favour of leisure activities that are done at home, but the changes that follow the loss of a spouse differ greatly by sex. Widowhood has very few consequences for men’s leisure activities at home and is moreover favourable to keeping on with activities that get them out of the home. In contrast, after the loss of their spouse, women have strongly reduced their leisure activities of all kinds. With time, though, widows again accord a large amount of time to the media. They are keenest on watching television and listening to the radio.

Despite this general decline in their leisure activities, seven out of ten respondents consider that they have fulﬁlled the plans they made before retiring for the use of their free time. On the whole, women, are less independent, have older spouses, and are left ﬁnancially less secure by widowhood, and for these reasons they are more likely than men to have had to abandon their goals.

Analysis of the reorganization of life following widowhood through the changes in family relationships and leisure activities shows that men are less isolated than women and are more likely to maintain social contacts outside the home. Do these different behaviours indicate that they cope better psychologically with the situation? What perception do those involved have of their own lives?

V. Psychological health

1. Perception of life and of self

The widowed have a less positive attitude towards life than married people, which is not an unexpected result [23]. However, despite having had to cope with two major shocks — leaving active life, followed by the loss of their mate — two-thirds of them (80% of married people) claim to be “happy” or “cheerful” most of the time, and three-quarters (nearly 90% of married people) claim to be “basically happy with life”. If in the last statement some are referring to their present situation, others may be looking at the whole of their lives, thus including the period that they spent living with their spouse. Moreover, with the approach of death this attitude may also simply reﬂect their attachment to life.

Behind this optimistic broad view there are also cases of distress. Certainly the extreme helplessness of people reporting their situation as hopeless is very rare among both married and widowed people (less than 5%). But it is nevertheless true that a sizeable number of respondents, especially widows, are genuinely unhappy: 42% of widows feel that their lives are empty, 33% often feel distraught (compared with 12% and 21% respectively of married women and 16% and 18% of widowers). These gloomy attitudes are less developed among men, for whom the loss of a spouse less often leads to pessimistic feelings. Women in these generations made family life their priority and with the death of the partner an essential part of their life is doubtless lost. It is also possible that women are more willing to report their unhappiness.

Ill health causes both married people and widowers to take a more negative view of life. For both groups a privileged social position often, but not always, has a protective effect. Retired managers are less likely to consider their lives to be empty, and are less subject to feelings of distress or hopelessness. Lastly, a fall in living standards depresses morale more among married people than among widowers.

2. Feeling lonely

The respondents are increasingly prone to loneliness as they become older. Widowed persons, however, suffer most [24] (Table 11). At 62 women who were going to lose their spouses already felt this way (32% felt lonely often or sometimes, compared with 20.5% of women who would remain married). At this stage in life, death is often preceded by a long period of disability that forces a couple to reduce its contacts and activities, so it is not surprising that wives feel lonely. Nothing like this is observed, however, for the men. At 75, after being widowed, 85.1% of women and 66% of men report suffering from loneliness. The death of the spouse represents the loss not only of the chief source of conversation but also, especially in cases of ill health, of a precious link with the outside world. This feeling is lessened slightly when the person has been widowed for a longer period of time [25].

Table 11
Feelings of loneliness and boredom at ages 62 and 75, by marital status (%)

IMGIMG Men Women 62 75 62 75 Often or o...IMGIMF

These results, which are supported by the ﬁndings from numerous other surveys (Croutte, 1993; de Jong Gierveld, 1998; Lichtenstein, 1996; Pan Ké Shon, 1999), show that despite an increased presence of family members, the emotional void left by the death of a spouse is only partially ﬁlled.

Women feel lonely more frequently than men and the loss of the spouse accentuates these differences [26]. In addition to the possible effect of a speciﬁcally female sensibility, it may be the consequence of their deeper commitment to family life. More affected than the men by the emptiness of the home when their children have moved out, they are left feeling more helpless by the loss of their spouses.

For widowed and married persons, being in good health allows them to combat isolation, though a high socio-economic status has no protective effect. Poor social integration (having very few social relations and no leisure activities besides listening to the radio and watching television) has no consequences for married people but increases the sense of loneliness among the widowed. Clearly the spouse makes up for a lack of social contacts; but this points to the fragility of a life focused on the couple, since the death of one partner increases the distress of the survivor. This impression of emptiness is exacerbated in widowed persons who were very satisﬁed with their emotional life. Lastly, being less comfortable ﬁnancially increases the feeling of loneliness [27]. The withdrawal into oneself associated with a fall in income is implicit here.

3. Boredom

The death of the partner means an end to daily conversations and probably also to many shared activities in which the spouse was the driving force. This being so, a growing sense of boredom could be expected to accompany the feeling of loneliness. Paradoxically, this is not observed among men. The proportion of widowers who do not know how to ﬁll their free time has even decreased slightly between 62 and 75 while nearly tripling for widows [28]. At 75, widowers are no more numerous than married men to report that they “often do not know what to do” (12% of both groups). The situation for women is the complete opposite: 32% of those who have lost their spouse between 62 and 75 often have this feeling compared with 15% of women who are still married [29]. With the passage of time, however, there is adjustment to the new situation and a reorganization of life, and the feeling of boredom is weaker among women already widowed at the age of 62 (25%) than among women widowed after that age.

For the most part, the women of these generations moved straight from the parental home to live with their husband and had few independent activities. Their leisure activities were generally carried out as a couple. In widowhood, as we have seen, they give up a number of activities - for a given socio-demographic environment there is a strong link between the increase in feelings of boredom and the abandonment of leisure activities — of which the spouse was no doubt the initiator. Added to this, many cannot drive and are reluctant to go out alone in the evening. Moreover, the death of a spouse results in fewer household tasks for women but a signiﬁcant increase for men (Umberson et al., 1992). Widowers ﬁnd that they have less spare time and they have less opportunity to feel bored. In addition, this study has shown that for the men, widowhood does not cause a reduction in leisure activities.

Even for couples, however, the feeling of boredom has practically doubled for women and remained stable for men. Women are more involved with the family and spend more time looking after their grandchildren. When the grandchildren grow up their grandmothers perhaps experience a sense of being redundant. Furthermore, women of these generations usually go out accompanied by their husbands. We know that “the older partner makes his mark on the sociability of his spouse” (Blanpain and Pan Ké Shon, 1998). Married to older men and so probably less dynamic than themselves, the women may well have to accept a certain lag between their aspirations and those of their partner.

For both widowed and married people good health protects against feelings of boredom. Even if their education, the interest of their occupation, the density of their social networks and their high income do not help retired managers to overcome feelings of loneliness, these factors do allow them to ﬁll their free time, whether they are married or widowed. The same is true to a lesser extent for white collar workers. Moreover, if leisure activities are a means to combat boredom, this attitude assumes a particular importance for widowers: having very few social contacts, not reading and not taking holidays increase the probability of experiencing boredom [30].

4. Depressive tendencies [31]

If the attention shifts from the particular psychological consequences of widowhood to general mental health, the conclusion is inevitable: retired persons who are widowed between 62 and 75 are psychologically more fragile. At 75, 42.5% of widowed men show depressive tendencies compared with 27.9% of married men. The proportions for women and married women are 55.9% and 36.7% respectively. All research evidence agrees on this point: women have a greater tendency to feel depressed than men, and widows and widowers tend to feel more depressed than married people (Clayton, 1979; Madison and Viola, 1968; Thuen et al., 1997; Forette et al., 1999).

On the other hand, concerning the question of whether the loss of a spouse is more damaging for men than for women, the results are contradictory. Some researchers have found no difference between the sexes (Feinson, 1986; Stevens, 1995). Others suggest that men are the principal victims (Lee et al., 1998; Van Grootheest et al., 1999), while others yet argue that women suffer most (Farnsworth et al., 1989; Gallagher et al., 1983). By controlling for health status, socio-occupational category, and perceived changes in income, this survey ﬁnds the differences by marital status at 75 to be more pronounced for women.

As in other studies (Bennett, 1996), we note that with time the grieving process runs its course. The women who have been widows longest are less depressed than those who have lost their spouse within the last thirteen years (51.1% and 55.9% respectively) though the married women still have the best psychological health. Moreover, being depressed is relatively rare among married people, but is greatly worsened by bad health or a drop in income, factors that have much less effect on widows and widowers suffering from the loss of a spouse.

Even if at 75 the respondents keep a generally positive outlook on life, the preceding thirteen years have been marked by the spread of an insidious gloom, especially among the women. More and more often, retired people suffer from loneliness or empty free time. Widowers, and widows even more, stand out clearly by their dark view of life and their more fragile psychological state.

The same conclusion is valid for the perception of retirement, which remains positive on the whole. Only 10% think that this stage in life is associated uniquely with drawbacks, and more than four times as many see only the advantages. But opinions are more qualiﬁed at 75 than at 62, and a tendency to take a pessimistic view of retirement is more frequent among those who have lost a spouse. Here again, widows stand out unfavourably: 17% of them see nothing but drawbacks in their lives, and less than double that proportion see only advantages.

Conclusion

Widowhood is essentially a woman’s lot. But the frequency of widowhood does not seem to make the experience less painful. The greater unhappiness of widows is the clearest illustration of this. Widows have a much more negative view of life and retirement, suffer more frequently from boredom and loneliness, and are more susceptible to depression. Their diminished involvement in leisure activities and their increased social isolation indicate the extent to which they are disadvantaged. All these indicators point to the particular difﬁculties of daily life for widows.

However, it should be pointed out that excess mortality is much higher for widowers than widows, particularly during the ﬁrst few years following bereavement. Faced with an apparently insurmountable emotional trauma, men are more likely than women to adopt dangerous types of behaviour (suicide, alcoholism, smoking) which are the main causes of this excess mortality (Thierry, 2000). Those who manage to avoid this outcome appear to adapt better than women to the loss of their spouse. Having been less focused during their lives on their marriage and on family life in general, they certainly have less difﬁculty in building a new life for themselves.

In the future, because of the succession of new generations, an improvement in women’s experience of widowhood may be expected. They will certainly be better equipped psychologically to deal with loneliness after the death of their spouse, as they will have more contact with the outside world through their greater participation in paid employment. They will have lived in more egalitarian relationships and a growing number of them will have contracted successive unions interspersed with periods of living alone (O’Bryant and Straw, 1991).

Moreover, by being more independent ﬁnancially (though also, for example, with regard to their mobility: at present only 30% of women aged 65 and over have a driving licence!), tomorrow’s retired women will certainly be better integrated in society and better equipped to cope with widowhood. In particular, they can be expected to have a greater participation in leisure activities and in consumer spending.

Furthermore, even though the decline of mortality will continue to push back the onset of widowhood, it is quite clear that this marital status will continue for a long time to be the lot of women. Although recent trends in mortality seem to suggest a narrowing of the gap between the sexes, the time when men and women will live out their old age together is still a long way off.

Even if the risk of being widowed at each age continues to decline at the same pace, the onset of old age for the baby boom generations spells an inﬂation of this population. This trend will result in an increase in the demand for support services.

Today, the burden of caring for a widowed parent generally falls on daughters. In the future, however, these women, more still than their mothers, will have discovered their identity in other spheres than the family role. All the factors of change in the female condition that favour a better adaptation of mothers to widowhood are also likely to deter daughters from accepting the constraints imposed by a widowed parent’s dependency.

Notwithstanding a genuine policy of support for professional help in recent decades, it is possible that collective solidarity will begin to crumble under the burden of ﬁnancing social protection. One result would be a far-reaching transformation in the experience of widowhood.

BIBLIOGRAPHIE

· Aquilino W., 1994, “Later life parental divorce and widowhood: Impact on young adults’ assessment of parent-child relations”, Journal of Marriage and the Family, 56, pp. 908-922.

· Bennett K., 1996, “A longitudinal study of well-being in widowed women”, International Journal of Geriatric Psychiatry, 11(11), pp. 1005-1010.

· Blanpain N., Pan Ké Shon J.-L., 1998, “Vieillir, c’est discourir… un peu”, Gérontologie et Société, 86.

· Cavalli S., Lalive d’épinay C., Spini D., 2001, “Le décès de proches: son impact sur la santé et sur la vie relationnelle des vieillards. Un suivi sur cinq ans d’une cohorte d’octogénaires”, Gérontologie et Société, 98, pp. 141-158.

· Clayton P.J., 1979, “The sequelae and non sequelae of conjugal bereavement”, American Journal of Psychiatry, 136, pp. 1530-1534.

· Coëffic N., 2000, “Faibles retraites et minimum vieillesse”, Études et résultats, Dress, 82.

· Croutte P., 1993, Le sentiment de solitude chez les personnes âgées de 60 ans et plus, Credoc, 54 p.

· De Jong GierveldJ., 1998, “A review of loneliness: concept and deﬁnitions, determinants and onsequences”, Reviews in Clinical Gerontology, 8, pp. 73-80.

· Delbès C., Gaymu J., 1997, “Convoler après 50 ans”, Gérontologie et Société, 82, pp. 95-105.

· Desplanques G., 1984, “L’inégalité sociale devant la mort”, Économie et Statistique, 162, pp. 29-50.

· Donnat O., 1998, Les pratiques culturelles des Français, La Documentation française, 357 p.

· Eggebeen D., 1992, “Family structure and intergenerational exchanges”, Research on Aging, 14, pp. 427-447.

· Farnsworth J., Pett M.A., Lund D.A., 1989, “Predictors of loss management and well-being in later life widowhood and divorce”,Journal of Family Issues, 10, pp. 102-121.

· Feinson M.C., 1986, “Aging widows and widowers: are there mental health differences ?”, International Journal of Aging and Human Development, 23, pp. 241-255.

· Fenwick R., Barresi C.M., 1981, “Health consequences of marital status change among the elderly: a comparison of cross- sectional and longitudinal analysis”, Journal of Health and Social Behavior, 22, pp. 106-116.

· Ferraro K.F., Barresi C.M., 1982, “The impact of widowhood on the social relations of older persons”, Research on Aging, 4(2), pp. 227-247.

· Ferraro K.F., 1985-1986, “The effect of widowhood on the health status of older persons”,Aging and Human Development, 21(1), pp. 9-25.

· Forette F. (dir.), Thierry X., Delbès C., Nizard A., 1999, Les répercussions du veuvage sur la morbidité et la mortalité, restricted diffusion, FNG, 140 p.

· Gallagher D.E., Breckenridge J.N., Thompson L.W., Peterson J.A., 1983, “Effects of bereavement on indicators of mental health in elderly widows and widowers”, Journal of Gerontology, 38, pp. 565-571.

· Héran F., 1988, “La sociabilité, une pratique culturelle”, Économie et Statistique, 216, INSEE, pp. 3-22.

· Heyman D.K., Gianturco D.T., 1973, “Long term adaptation by the elderly to bereavement”, Journal of Gerontology, 28, pp. 259-353.

· La Rue A., Bank L., Jarvik L., Heltland M., 1979, “Health in old age: how do physicians ratings and self-ratings compare ?”, Journal of Gerontology, 34, pp. 687-691.

· Lee G., Willetts M., Seccombe K., 1998, “Widowhood and depression, gender differences”, Research on Aging, 20(5), pp. 611-630.

· Lichtenstein P., 1996, “A co-twin- control study of response to widowhood”, Journal of Gerontology: Psychological Sciences, 51B (5), pp. 279-289.

· Madison D., Viola A., 1968, “The health of widows in the year following bereavement”, Journal of Psychosomatic Research, 12, pp. 297-306.

· Maddox G.L., Doulass E.B., 1973, “Self-assessment of health: a longitudinal study of elderly subjects”, Journal of Health and Social Behavior, 14, pp. 87-93.

· Murrel S.A., Himmelfarb S., Phifer J., 1988, “Effects of bereavement/loss and prevent status on subsequent physical health in older adults”, Aging and Human Development, 27(22), pp. 89-107.

· O’Bryant S., Straw L., 1991, “Relationship of previous divorce and previous widowhood to older women’s adjustment to recent widowhood”, Journal of Divorce & Remarriage, 15(3-4).

· Paillat P. (dir.), Attias-Donfut C., Clément F., Delbès C., Renaut S., Rozenkier A., 1989, Passages de la vie active à la retraite, PUF (Coll. Politique d’aujourd’hui), 267 p. PAN KÉ SHON J.-L., 1999, “Vivre seul, sentiment de solitude et isolement relationnel”, INSEE première, 678, 4 p.

· Roan C., Kelly Raley R., 1996, “Intergenerational coresidence and contact: a longitudinal analysis of adult children’s response to their mother’s widowhood”, Journal of Marriage and the Family, 58(3), pp. 708-717.

· Sermet C., Grandjean E., 1998, Évolution de l’état de santé des personnes âgées en France: 1970-1996 (Coll. Questions de santé), INSERM, 334 p.

· Stevens N., 1995, “Gender and adaptation to widowhood in later life”, Ageing and Society, 15, pp. 37-58.

· Stoller E., Lorna L.E., 1983, “Help with activities of everyday life: sources of support for non-institutionalized elderly”, Gerontologist, 23, pp. 64-70.

· Stroebe M.S., Stroebe W., 1983, “Who suffers more? Sex differences in health risks of the widowed”, Psychological Bulletin, 93(2), pp. 279-301.

· Thierry X., 1999, “Risques de mortalité et de surmortalité au cours des dix premières années de veuvage”, Population, 54(2), pp. 177-204.

· Thierry X., 2000, “Mortel veuvage: risques de mortalité et causes médicales des décès aux divers moments du veuvage”, Gérontologie et Société, 95, pp. 27-46.

· Thuen F., Reime M., Skrautvoll K., 1997, “The effect of widowhood on psychological wellbeing and social support in the oldest groups of the elderly”, Journal of Mental Health, 6(3), pp. 265-274.

· Umberson D., Wortman C.B., Kessler R.C., 1992, “Widowhood and depression: explaining long term gender differences in vulnerability”, Journal of Health and Social Behavior, 22, pp. 10-24.

· Van Grootheest D.S., Beekman A.T.F., Broese Van Groenou M.I., Deeg D.J.H., 1999, “Sex differences in depression after widowhood. Do men suffer more ?”, Soc. Psychiatry Epidemiol., 34, pp. 391-398.

· Wan T., Odell B., 1983, “Major role losses and social participation of older males”, Research on Aging, 5(2), pp. 173-196.

· Yvesavage J.A., 1992, “Depression in the elderly: how to recognize masked symptoms and choose appropriate therapy”, Postgrad Med., 9, pp. 255-258, 261.

NOTES

[*]

Fondation Nationale de Gérontologie and Institut National d’Études Démographiques, Paris.

[**]

Institut National d’Études Démographiques.Translated by O. Chimere-Dan

[1]

The structure of the marriage market is unfavourable for women. At ages 60-64 twice as many women as men are unmarried, and four times more over 75. Moreover, men who marry late tend to have spouses who are younger - sometimes much younger - than themselves (Delbès and Gaymu, 1997).

[2]

Thanks to ﬁnancial support provided mainly by the National Centre for Old Age Insurance (CNAV) and the Association for Complementary Pension Schemes (ARRCO).

[3]

The state of health corresponds to the individual’s perception and has not been validated by a medical examination. Numerous studies (La Rue et al., 1979; Maddox and Doulass, 1973) have, however, conﬁrmed the subjective assessment: “perceived” and “diagnosed” states of health are strongly correlated.

[4]

When only the last two factors operate, i.e. in households where both spouses were already retired thirteen years ago, income fell by 6%.

[5]

If the observed differences between widowed and married persons at 75 are greater than expected in view of the loss of income following the death of a spouse, this could be because persons still in union at that age belong on average to better off social groups.

[6]

Whether we allocate 1.5 or 1.7 consumption units per couple.

[7]

Result corroborated by a logistic regression controlling for socio-occupational category, number of children, and state of health.

[8]

It must be noted, however, that at age 62, men who were subsequently widowed saw much less of their families than those who remained married (16.9 against 18.6 visits per month). Some of them must have had wives suffering from ill health, which is a cause for neglecting family relations for women. On the other hand, the women who became widows saw their family as often as women who were still married.

[9]

At 75, family visits are more frequent among blue collar workers. Widowers are somewhat more often lower on the social scale, though this characteristic only partly explains their greater family sociability.

[10]

The effect of time on the family sociability of widowers cannot be examined as there are very few widowers in their early 60s.

[11]

The level of helping behaviour of the married respondents is slightly overestimated in our survey, since some reported help that was given by their spouses. Thus, a large number of married men claimed to knit.

[12]

The lesser involvement of widowers is conﬁrmed by an analysis that assumes all things being otherwise equal, and accounts for health status, number of children and socio-occupational category.

[13]

Result conﬁrmed by a logistic regression ﬁrst integrating socio-occupational category and health status, and then number of children, quality of relationship with the children and perceived change in the quality of life.

[14]

Results conﬁrmed by logistic regression controlling for socio-occupational category and health status.

[15]

See footnote 14.

[16]

The socio-occupational distribution explains only part of this result.

[17]

A pastime often limited to taking a few photographs.

[18]

For these activities, the inﬂuence of health is only signiﬁcant for married people; the relation is the same for widows and widowers but not signiﬁcant.

[19]

Differences in participation rates between 62 and 75 are, however, not signiﬁcant.

[20]

More than four days away from home.

[21]

Logistic regression taking into account occupational category, health status and changes in income shows that widows take more holidays, but the difference between the sexes is not statistically signiﬁcant.

[22]

The only behavioural pattern that distinguishes them from their married counterparts who are much less likely to give up their holiday. A slight increase in the number of days spent away is observed for both groups.

[23]

Result conﬁrmed by analyses that hold all other things equal, controlling ﬁrst for the effects of socio-occupational category and health status, then introducing perceived changes in income and changes in the number of family and social visits.

See footnote 23.

See footnote 23.

See footnote 23.

The association is statistically signiﬁcant only for married people.

[28]

In a given socio-demographic environment, taking into account the initial frequency of the feeling of boredom, at 75 widowers are no different from married men while widows are much more frequently bored than their married counterparts. Furthermore, the difference between widows, widowers and married people is accentuated by bad health. Losing a spouse is more harmful to those retired from managerial positions than to those retired from other professions.

[29]

These contrasts are conﬁrmed by a logistic regression that ﬁrst controls for the effects of occupational category and health, before introducing the perception of changes in income and in social contacts. The effect of sex is found only among the widowed respondents.

[30]

These variables are not signiﬁcant for married people.

[31]

To measure the depressive tendencies affecting the retired at 75, we used the scale for evaluating depression in the elderly proposed by J.A. Yvesavage (1992). It takes the form of 15 questions, with points given for each. A score of between 5 and 9 corresponds to a strong likelihood of depression, a higher score means that the person is almost certainly depressed. Respondents with a score above 5 were considered to exhibit depressive tendencies.

Volume 57 2002/6

The Shock of Widowhood on the Eve of Old Age: Male and Female Experiences

Christiane Delbès [*] Christiane Delbès, Institut National d’Études Démographiques, 133 bd Davout, 75980 Paris Cedex 20, tel: 33 0(1) 56 06 20 40, fax: 33 0(1) 56 06 21 99, Joëlle Gaymu [**]

I. Health status

II. Income

III. Social ties

IV. Leisure activities

V. Psychological health

Conclusion

BIBLIOGRAPHIE

NOTES

Christiane Delbès []* Christiane Delbès, Institut National d’Études Démographiques, 133 bd Davout, 75980 Paris Cedex 20, tel: 33 0(1) 56 06 20 40, fax: 33 0(1) 56 06 21 99, Joëlle Gaymu *[**]*