Pet attachment and generativity among young adults. Shaela G. Marks; Jean E. Koepke; Cheryl L. Bradley.

Author's Abstract: COPYRIGHT 1994 Heldref Publications

Attachment to pets and generativity (concern for the next generation) may be positively correlated in young adulthood in that both involve nurturance. To test the hypothesis that pet attachment and generativity are related, we administered two relatively new instruments, the Pet Attachment Survey (PAS; Holcomb, Williams & Richards, 1985) and the Loyola Generativity Scale (LGS; McAdams & de St. Aubin, 1992), in addition to the Pet Attitude Inventory (Wilson, Netting, & New, 1987) and demographic questions to 179 young university students. Students who did not own pets answered 12 questions taken from the Pet Attitude Inventory. Pet owners' attachment scores were significantly and positively correlated with their generativity scores, as we had predicted. These and other findings concerning gender, age, and being primary caregivers of pets contribute to the ongoing validation of both the PAS and the LGS.

Full Text: COPYRIGHT 1994 Heldref Publications

COMPANION ANIMALS provide numerous psychological, emotional, and physiological benefits for their owners, one of which is that they may serve as outlets for nurturance (Savishinsky, 1983). Previous research has shown that people are attached to their pets (Holcomb, Williams, & Richards, 1985) and that attachment to pets involves nurturance (Rynearson, 1978). Generativity, conceptualized as concern for the next generation (Erikson, 1950), encompasses nurturance, and a connection between the two has been supported in recent research (Van De Water & McAdams, 1989). That nurturance is involved in both pet attachment and generativity suggests a relationship between the concepts.

In this study, we investigated the relationship between attachment to pets and generativity with the expectation that attachment to pets would be positively related to generativity in young adults. A second aim of the study was to replicate the initial findings regarding pet attachment (Holcomb et al., 1985) and generativity (McAdams & de St. Aubin, 1992), especially concerning gender, age, and being a primary caregiver of a pet.

Many studies demonstrate the beneficial effects of companion animals. Friedmann, Katcher, Lynch, and Thomas (1980) found that having a pet in the home was the strongest predictor of 1-year survival following discharge from a coronary care unit. These effects were independent of socioeconomic status, social network, geographic mobility, and living situation. Katcher (1981) found that touch-talk dialogue with animals was associated with lower blood pressure than dialogue with people. Mugford and M'Comisky (1975) found that a group of elderly people who were given budgies showed greater improvement in their attitudes toward people and their own psychological health than before they owned the budgies. Hyde, Kurdek, and Larson (1983) reported that college-aged pet owners had higher empathy and interpersonal trust scores than people without pets.

The positive effects of pets may be partly due to the fact that they invite nurturance. Savishinsky (1983) proposed that pets serve as outlets for nurturance, displacement, and projection and can function as child substitutes at various phases of the life cycle. Katcher and Beck (1987) suggested that people in urban societies are often deprived of opportunities to be nurturant and affectionate toward others and that pets can serve as an extension for human nurturing. Caring for a pet also wards off depression and despair, for it gives the owner a feeling that he or she is productive (Beck & Katcher, 1983).

Veterinarians have observed that people are often very attached to their pets (Voith, 1985). Attachment, which initially referred to the bond that develops between a caregiver and his or her infant (Ainsworth & Bell, 1974), is defined in several ways. It can refer to an emotional state or feeling and also to specific behaviors that an individual uses to keep another individual close to him or her (Voith, 1985). Specific patterns of attachment, however, depend on individual differences among species, early experiences, and environmental conditions. A genetic predisposition may even underlie the development of attachments (Voith, 1985).

Activities that build and maintain attachments between people may also occur between individuals and their pets (Voith, 1985). Many of the interactions that take place between people and their pets resemble those that take place between individuals and their children. This is because children and pets share many of the same attributes. Pets are similar to children in that they rely on someone to take care of them. Many pets can be picked up and carried, just like children. In this respect, pets may be viewed as perpetual children (Voith, 1985).

In 1985, Holcomb, Williams, and Richards designed the Pet Attachment Survey (PAS) to measure the degree to which individuals are attached to their pets. The items on the scale were derived from several sources, one of which was Ainsworth's model of human attachment (Ainsworth & Bell, 1974). Examples of items based on Ainsworth's research include "You feel sad when you are separated from your pet" and "You don't like your pet to get too close to you" (Holcomb et al., 1985, 28). Other items were derived from an earlier questionnaire on pet attachment (Katcher, Friedmann, Goodman, & Goodman, 1983), and further items were selected based on the research of Templer, Salter, Dickey, Baldwin, and Veleber (1981), Lago (1985), Sheldon, Levy, and Shott (1985), and Voith (1983).

The PAS is divided into two subscales: (a) a Relationship Maintenance scale that measures behaviors such as interaction, communication, and time or financial involvement with one's pet; and (b) an Intimacy scale that measures behaviors such as proximity and emotional importance of one's pet. Cronbach alphas for the Relationship Maintenance and Intimacy scales were .83 and .74, respectively. No overall reliability scores were reported (Holcomb et al., 1985).

In validating the PAS, Holcomb et al. (1985) used two samples of adult volunteers (mean age, 35 and 40 years) who brought their pets to veterinary facilities. They found that women scored higher than men on the PAS and that primary caregivers scored higher than nonprimary caregivers. Also, dog owners scored higher on the Relationship Maintenance scale than cat owners did.

Generativity, a concept devised by Erik Erikson (1950), becomes salient in the adult years. Erikson defines generativity as the time when the adult individual is engaged in establishing and guiding the next generation. In its broadest application, generativity encompasses procreativity, creativity, and productivity in a mutually beneficial relationship between the caregiver and that which is cared for. Generativity includes nurturing, teaching, and promoting the next generation, while generating life products that benefit society.

McAdams and de St. Aubin (1992) designed the Loyola Generativity Scale (LGS) primarily to assess individual differences in the aspect of generativity that refers to genuine concern for the next generation. To establish reliability and validity, McAdams and de St. Aubin (1992) administered the LGS and two other generativity scales to 165 college students and also to a sample of 149 adults (ages 19-68). The LGS showed high internal consistency. Cronbach's alpha was .83 for the adult sample and .84 for the college sample. The LGS also had low correlations with social desirability. In addition, for the combined sample, McAdams and de St. Aubin reported that LGS scores were significantly higher for women and were modestly correlated with age (r = .17). In subsequent research with young, midlife, and older adults, McAdams, de St. Aubin, and Logan (1993) found no gender differences in LGS scores and significant age differences only in a retest administration of the LGS by telephone. Reliability of the LGS was further established by a test-retest correlation of .55.

Generativity and attachment to pets may be related in young adults in that both involve nurturance. Erikson (1950) and McAdams and de St. Aubin (1992) describe the desire to nurture and be needed as important aspects of generativity. Evidence that generativity and nurturance are related was provided by Van De Water and McAdams (1989) in a study examining the relation between three earlier measures of generativity and four personality traits. They found that nurturance (Jackson, 1974) was the strongest predictor of all three measures of generativity.

Nurturance is also involved in pet attachment, according to Voith (1985), who proposed that one of the activities serving as an attachment mechanism is caring for or nurturing pets. Theoretically, nurturance thus appears to be a common component of both pet attachment and generativity, suggesting that attachment to pets may be positively related to generativity. We hypothesized that pet attachment, as measured by the PAS, would be positively correlated with generativity, as measured by the LGS.

Consistent with the findings of Holcomb et al. (1985), we also hypothesized that female pet owners would score higher on the PAS than male pet owners, that primary caregivers would score higher than nonprimary caregivers, and that dog owners would score higher than cat owners on the Relationship Maintenance scale. In attempting to replicate the initial findings of McAdams and de St. Aubin (1992) we expected that women would score higher than men on the LGS and that age would be positively and modestly correlated with LGS scores.

Method

Subjects

Participants were undergraduate psychology students (N = 179) attending Simon Fraser University. All were volunteers (136 women, 43 men) ranging in age from 18 to 44 years (M = 22.98, SD = 4.6) and of whom 87% were single and most were childless (93%). Subjects were asked to participate in a study concerning attitudes toward pets, work, family, and the community. Fifty seven percent (102) were pet owners, and 43% (77) did not own pets. Because all the subjects did not complete all the questions, the number of subjects in the statistical analyses varied somewhat, with 91 completing all the questions on both scales.

Measures

The main measures included in this study were the 20-item Loyola Generativity Scale (LGS) and the 27-item Pet Attachment Survey (PAS). The LGS uses a 4-point Likert-type scale ranging from very often (3) to never (0). The PAS also uses a 4-point Likert-type scale ranging from almost always (4) to never (1). Thus, high scores on the LGS and the PAS indicate greater expression of generativity and greater pet attachment, respectively. Demographic questions and 12 questions for nonowners from the Pet Attitude Inventory (Wilson, Netting, & New, 1987) were included to obtain descriptive information about the attitudes of nonowners toward pets.

We prepared 2 additional questions concerning whether subjects enjoyed caring for their pets and whether they considered their pets as children; those questions were scored using the same 4-point Likert-type scale as the PAS. The first question was designed to provide a preliminary indication of nurturance, which has been suggested as a component of both pet attachment and generativity. The second was included to determine whether assuming a parental role in relation to a pet might also be related to pet attachment and generativity.

Procedure

The overall questionnaire, entitled Pets, Family, Work, and the Community, included the various measures, organized as follows. First, all participants completed demographic questions indicating their age, gender, marital status, place of residence, and number of children. Second, subjects completed the Loyola Generativity Scale. Next, pet owners completed the Pet Attachment Survey and the 2 additional questions prepared by the researchers, while nonowners completed 12 questions taken from the Pet Attitude Inventory. Questionnaire results were analyzed, using Pearson correlation coefficients, t tests, and two-tailed tests of significance.

Results

A significant positive correlation was found between pet owners' scores on the PAS and their corresponding LGS scores, r(91) = .319, p [is less than] .01. Women scored slightly but not significantly higher than men on both the PAS and LGS, as indicated by t tests. The correlation between age and LGS scores was also not significant, r(169) = .12.

Primary caregivers of pets scored significantly higher on the PAS than nonprimary caregivers, t(92) = 3.45, p [is less than] .001. Mean scores for the 69 primary and 25 nonprimary caregivers were 72.4 (SD = 12.6) and 62.1 (SD = 13.1), respectively. Dog owners scored significantly higher than cat owners on the Relationship Maintenance scale of the PAS, t(85) = 3.07, p [is less than] .005. Mean scores for the 55 dog owners and 32 cat owners were 44.9 (SD = 8.0) and 39.7 (SD = 7.3), respectively.

TABLE 1
 
Mean Pet Attachment Survey (PAS) and Loyola Generativity Scale (LGS) Scores
for Men and Women
 
Scale                  N                M              SD
 
PAS
Men                   19             66.7            14.31
Women                 75             70.4            13.26
Overall               94             69.7            13.48
 
LGS
Men                   39             35.8             7.94
Women                130             37.9             7.47
Overall              169             37.4             7.60

Enjoying taking care of a pet was significantly and positively correlated with pet owners' scores on both the PAS, r(93) = .69, p [is less than] .001, and the LGS, r(98) = .25, p [is less than] .05. In addition, considering one's pet as a child was significantly and positively correlated with PAS scores, r(94) = .50, p [is less than] .001, but not with LGS scores, r(99) = .06.

Of the 77 nonowners, 81% (62) indicated having previously owned a pet. Their main reasons for not currently owning a pet were that they could not keep a pet at their present residence (25%) or they did not have the time or money to care for a pet (42%). Only 11% said they did not enjoy animals. When asked if they would want a pet if they could afford the time and cost, 78% replied yes.

Discussion

Our hypothesis that attachment to pets would be positively related to generativity in young adults was confirmed by a significant positive correlation between PAS and LGS scores. This suggests that young adults who are more attached to their pets also express higher levels of generativity, as measured by the LGS. The finding that both the PAS and LGS scores were significantly correlated with enjoying taking care of pets is a preliminary indication that nurturance may be involved in both, as was expected theoretically. Thus, young adults who enjoyed caring for or nurturing their pets also expressed higher levels of both pet attachment and generativity.

Although assuming a parental role and viewing a pet as a child were positively related to pet attachment, they were not related to generativity, supporting the proposal that it may be nurturance rather than a parent-child type of relationship that links pet attachment and generativity. This result is consistent with Erikson's belief that generativity extends beyond parenting and is not synonymous with it, and also with the fact that the LGS is broadly based and not directly related to parenting.

Keeping pets as outlets for nurturance (Katcher & Beck, 1987; Savishinsky, 1983) may be a major factor in motivating so many people to become pet owners. In this study, most subjects not currently owning pets (62) had previously owned them. Combining these with the current pet owners (102) indicates that 92% (164) of the young adults owned pets at some time. This is comparable to Cain's (1985) report that 93% of her sample of 896 adults had pets while growing up. In addition, of those who could not keep pets while students, most wanted to have them in the future. This popularity enjoyed by pets may well reflect the emotional value that nurturing them has for many individuals.

The hypothesis concerning gender differences in pet attachment was not confirmed, as female pet owners did not score significantly higher than males on the PAS. This is inconsistent with the findings reported by Holcomb et al. (1985) for adults, although the mean scores are in the same direction. It is unlikely that this difference in results is related to the low proportion of men in this sample, because the proportion of men in both studies is comparable, 24% in this study and 28% in the research by Holcomb et al. (1985); the inconsistency could be related to cohort effects.

The lack of gender differences in PAS scores in our current study is, however, consistent with findings reported by Stevens (1990), who administered the PAS to eighth-grade children. She reported no significant differences in attachment to pets between boys and girls. Her findings and those of the present study suggest that gender differences in pet attachment may increase with age, becoming more evident in middle adulthood.

Furthermore, the PAS values obtained in this study are clearly lower than those found by Holcomb et al. (1985) for middle-aged adults but are slightly higher than those found by Stevens (1990). Holcomb et al. (1985) reported mean scores per item of 3.18 for women and 3.02 for men, whereas similar scores in the present study were 2.60 for women and 2.47 for men. Taken together, these findings suggest that pet attachment scores may be higher and may vary by gender more for middle-aged persons taking their pets to veterinary facilities than for young adults and adolescents from a more general population.

The hypothesis concerning gender differences in LGS scores was not confirmed. Women did not score significantly higher than men, although the means were in the hypothesized direction and comparable to those reported by McAdams and de St. Aubin (1992) for college women and men (M = 40.4, 37.2) and by McAdams et al. (1993) for the overall scores of young adults (M = 40.1). Bradley (1992) also found no significant gender differences when she administered the LGS to middle-aged adults. The mean LGS scores in her study were again slightly higher for women than for men (M = 37.74, 36.07) and were very similar to those obtained in this study. Although our results could have been influenced by the low proportion of male participants, they nevertheless suggest only limited support for the gender differences reported by McAdams and de St. Aubin (1992) and are consistent with later research by McAdams et al. (1993) in which no gender differences were found.

There was no significant relation between age and LGS scores. This finding differs from the results initially reported by McAdams and de St. Aubin (1992) and may well be related to the limited age distribution in this sample, as most subjects (82%) were between 18 and 25 years of age. The present findings are consistent, however, with those of a later study by McAdams et al. (1993) in which age differences were not found initially, though a weak relation was found in a re-test administration of the LGS by telephone.

The hypothesis that primary caregivers of pets would score significantly higher on the PAS than nonprimary caregivers was confirmed. This is consistent with the findings reported by Holcomb et al. (1985) and supports the idea that primary caregivers, who interact with and care for their pets, are more attached to pets than nonprimary caregivers.

The hypothesis that dog owners would score significantly higher than cat owners on the Relationship Maintenance scale was also confirmed. This finding is consistent with those reported by Holcomb et al. (1985), suggesting that dog owners may be more involved in some of the caretaking activities included in the Relationship Maintenance Scale, such as grooming and exercising.

Overall, these results contribute to the ongoing validation of both the PAS and the LGS. This study, the first reported in which the PAS was administered to college students, supports and extends the original findings concerning pet attachment (Holcomb et al., 1985). Both young and middle-aged adults who are primary caregivers showed stronger attachment to their pets than nonprimary caregivers, and those who are dog owners were more concerned than cat owners with relationship maintenance. These young adults, however, showed less attachment to their pets than did middle-aged adults taking their pets to veterinary facilities (Holcomb et al., 1985) and were more similar to adolescents in the eighth grade (Stevens, 1990) in this and in the lack of gender differences. This suggests that generalizations about pet attachment may need to be qualified in terms of age and other criteria.

With respect to the LGS, the generativity scores obtained by college students in both the initial (McAdams & de St. Aubin, 1992) and later research (McAdams et al., 1993) were replicated. The significant gender differences initially reported (McAdams & de St. Aubin, 1992), however, were not replicated in this or the later research (McAdams et al., 1993). In addition, the present results, in failing to replicate the modest correlation with age, initially noted by McAdams and de St. Aubin (1992), are consistent with the primary findings later reported by McAdams et al. (1993).

These results do contribute new information showing that young adults who are more attached to their pets, as measured by the PAS, also tend to be more generative, as measured by the LGS. That these young adults also enjoyed caring for their pets is consistent with the suggestion that nurturance may link pet attachment and generativity and indicates the need for further research.

REFERENCES

Ainsworth, M., & Bell, S. (1974). Mother-infant interaction and the development of competence. In K. Connoly & J. Bruner (Eds.), The growth of competence (pp. 97-118). San Diego: Academic Press.

Beck, A., & Katcher, A. (1983). Between pets and people. New York: G. P. Putnam.

Bradley, C. (1992). Generativity vs. stagnation: A study in the validation of a measure. Unpublished master's thesis, Simon Fraser University, Burnaby, B.C., Canada.

Cain, A. (1985). Pets as family members. Marriage & Family Review, 8, 5-10.

Erikson, E. (1950). Childhood and society. New York: Norton.

Friedmann, E., Katcher, A., Lynch, J., & Thomas, S. (1980). Animal companions and one-year survival of patients after discharge from a coronary care unit. Public Health Report, 95, 307-312.

Holcomb, R., Williams, R., & Richards, P. (1985). The elements of attachment: Relationship maintenance and intimacy. Journal of the Delta Society, 2, 28-34.

Hyde, K., Kurdek, L., & Larson, P. (1983). Relationships between pet ownership and self-esteem, social sensitivity, and interpersonal trust. Psychological Reports, 52, 110.

Jackson, D. (1974). Personality research form. Goshen, NY: Research Psychologists Press.

Katcher, A. (1981). Interactions between people and their pets: Form and function. In B. Fogle (Ed.), Interrelations between people and pets (pp. 41-67). Springfield, IL: Charles C. Thomas.

Katcher, A., & Beck, A. (1987). Health and caring for living things. Anthrozoos, 1, 175-183.

Katcher, A., Friedmann, E., Goodman, M., & Goodman, L. (1983). Men, women, and dogs. California Veterinarian, 2, 14-16.

Lago, D. (1985, October). Assessment of favorable attitudes towards pets. Paper presented at the meeting of the Delta Society, St. Louis, MO.

McAdams, D., & de St. Aubin, E. (1992). A theory of generativity and its assessment through self-report, behavioral acts, and narrative themes in autobiography. Journal of Personality and Social Psychology, 62, 1003-1015.

McAdams, D., de St Aubin, E., & Logan, R. (1993). Generativity among young, midlife, and older adults. Psychology and Aging, 8, 221-230.

Mugford, R., & M'Comisky, J. (1975). Some recent work on the psychotherapeutic value of caged birds with old people. In R. S. Anderson (Ed.), Pet animals and society (pp. 54-65). London, England: Bailliere-Tindall.

Rynearson, E. K., (1978). Humans and pets and attachment. British Journal of Psychiatry, 133, 550-555.

Savishinsky, J. (1983). Pet ideas: The domestication of animals, human behavior, and human emotions. In A. Katcher & A. Beck (Eds.), New perspectives on our lives with companion animals (pp. 112-131). Philadelphia, PA: University of Pennsylvania Press.

Sheldon, S., Levy, H., & Shott, S. (1985). The distribution of pets and the pet attachment patterns in a sample of victims of child abuse and neglect. People, Animals, Environment, 3, 14.

Stevens, L. (1990). Attachment to pets among eighth graders. Anthrozoos, 3, 177-183.

Templer, D., Salter, C., Dickey, S., Baldwin, R., & Veleber, D. (1981). The construction of a pet attitude scale. The Psychological Record, 31, 343-348.

Van De Water, D., & McAdams, D. (1989). Generativity and Erikson's "belief in the species." Journal of Research in Personality, 23, 435-449.

Voith, V. (1983). Human/animal relationships. In R. S. Anderson (Ed.), Nutrition and behavior in dogs and cats (pp. 147-156). Oxford, England: Pergamon Press.

Voith, V. (1985). Attachment of people to companion animals. Veterinary Clinics of North America: Small Animal Practice, 15, 289-295.

Wilson, C., Netting, F., & New, J. (1987). The Pet Attitude Inventory. Anthrozoos, 1, 76-84.