Feb 26, 2014 - Health inJackson Heights, New York. Dr. Robins is ... tion in Bridgewater, New Jersey. deaf and in hearing ... mous achievement in deaf individuals may differ, on the average, from ..... Berkeley: University of Cali- fornia Press.
Toward Greater Understanding of Depression in Deaf Individuals Irene W. Leigh, Clive J. Robins, Joan Welkowitz, Ronald N. Bond
American Annals of the Deaf, Volume 134, Number 4, October 1989, pp. 249-254 (Article) Published by Gallaudet University Press DOI: 10.1353/aad.2012.0662
For additional information about this article http://muse.jhu.edu/journals/aad/summary/v134/134.4.leigh.html
Access provided by Duke University Libraries (26 Feb 2014 12:47 GMT)
Toward Greater Understanding of Depression in Deaf Individuals Irene W. Leigh, Ph.D., CliveJ. Robins, Ph.D.,
Joan Welkowitz, Ph.D., and Ronald N. Bond, Ph.D. We compared the prevalence of depressive symptoms among deaf and
hearing college students and examined the relationships among depressive
symptoms, personality characteristics, and perceived parental attitudes and
behaviors in these two groups. Measures were revised to meet the language needs of the deaf subjects. Mild levels of depressive symptoms were more prevalent in the deaf than in the hearing students, but more severe depression was not In both groups, depressive symptoms were associated with
perceptions of lower maternal care and higher maternal over-
protection. Deaf and hearing subjects did not differ on these perceived maternal characteristics. Depressive symptoms were associated with so-
ciallydependent personality characteristics in the hearing sample only. We discuss the implications of the findings for the role of personality develop-
ment in depression in deaf individuals.
Unquestionably, many deaf adults have the capacity to show admirable resil-
ience in the face of adversity and can adapt well to frustrating
situations, as exemplified by numerous success stories. Nonetheless, continuing confrontation with the various repercussions of deafness is bound to have some impact on emotional adjustment. Chough (1979) reported that 46 percent of 229 deaf college students rated themselves as "unhappy." Little prevalence data regarding clinical levels of
depression, however, is available. The present study was designed to provide an adequate self-report measure of depressive symptoms for deaf individuals, to assess the
prevalence of depressive symptoms in deaf young people and to determine whether such symptoms are related to the same personality characteristics and developmental factors in Dr. Leigh is a psychologist urith the Lexington Center for Mental Health inJackson Heights, New York. Dr. Robins is assistantprofessor of medical psychology at Duke University in Durham, North Carolina. Dr. Welkowitzis professorof psychology at New York University in New York City. Dr. Bond is a research analyst with A T&T Corporation in Bridgewater, New Jersey. AAD / October 1989
deaf and in hearing persons. Clearly, the diagnosis of deafness in young children has myriad ramifications for parent-child interactions and their possible developmental sequelae. Stein and Jabaley (1981) observed that if normally-bonding parents move through a mourning period subsequent to the diagnosis of deafness in their offspring and work towards acceptance, parent-child bonding continues without excessive alteration. When inadequacies exist in the bonding process, particularly when parents experience protracted difficulties in the necessary adjustment to deafness, however, the deaf child is likely to experience emotional difficulties that may persist into or manifest at adulthood.
Theorists have proposed that certain aspects of the parentchild relationship could predispose a child to later depression (e.g., Arieti&Bemporad, 1978; Bowlby, 1978). According to factor-analytic studies, two principal source variables largely account for differences in parent-child relationships (Parker, Tupling, & Brown, 1979). The first covers affection and acceptance as contrasted with coldness and rejection. The second dimension reflects psychological control or overprotection versus encouragement of autonomy. Assocations between depressive symptomatology in individuals and their 249
perceptions of their parents as insensitive, uncaring and controlling have been noted in several studies (Munro, 1966; Abrahams & Whitlock, 1969; Jacobsen, Fasman & DiMascio, 1975; Blatt, Wein, Chevron, & Quinlan, 1979; Parker, 1979; Crook, Raskin, & Eliot, 1981; McCranie & Bass, 1984). No studies have examined such relations in deaf persons. How do early parent-child relations influence depression later in life? It is likely that this influence is mediated by relatively stable differences in critical personality characteristics. Clinical and research evidence suggest that two personality dimensions may create vulnerability to depression and correlate with different types of depressive experiences, thereby having implications for developing therapeutic strategies tailored to deal with each type of depression (Blatt, 1974; Arieti & Bemporad, 1978; Blatt, Quinlan, Chevron, McDonald & Zuroff, 1982; Beck, 1983).
Beck (1983) described a dimension of sociotropy which encompasses a person's investment in receiving gratification
from positive interpersonal relations involving understanding, intimacy, etc. The highly sociotropic person relies on social relations to validate self-worth. In contrast, the au-
tonomous dimension encompasses a goal-dominated orientation that reflects a person's investment in mastery of independent functioning to enhance self-worth. When either dimension dominates the personality, and the individual experiences circumstances in which these needs are threat-
ened, the potential for depression increases. Robins and Block (1988) have reported that sociotropy does appear to serve as a vulnerability to depressive symptoms in the context of negative events. It is possible that needs for social dependency or autonomous achievement in deaf individuals may differ, on the average, from those of their hearing counterparts as a result of differences in socialization, such as child-rearing patterns that rely on strong parental control and that reinforce passivity rather than active exploration (Schlesinger & Meadow, 1972; Brinich, 1980; Wedell-Monnig & Lumley, 1980). In view of clinical and research evidence that has established relation-
ships between strong needs for dependency or goal achievement and depression, as well as between parent bonding and depression, the interrelation of all these variables needs further scrutiny. In particular, no studies have looked at these interrelations in a sample of deaf individuals. The present study therefore addressed the following questions: 1) Do deaf and hearing populations differ in the extent to which they experience depressive symptoms?; 2) Do these two groups dif-
fer in levels of the personality characteristics of sociotropy and autonomy, in perceived parental characteristics, or in the
relations between these characteristics and level of depres-
2 or earlier (in actuality, 90 out of 102 subjects reported age of onset at birth while the remainder reported becoming deaf anywhere from 2 to 24 months of age); no additional handi-
caps; parents with normal hearing; and age younger than 25
Instruments used in this investigation included versions of
the Beck Depression Inventory, the Sociotropy-Autonomy Scale, and the Parental Bonding Instrument, each of which had been revised for use with a college-aged deaf population, most of whom had limited exposure to American Sign Language (NTID, 1983-1984). The nature of these revisions is fully described elsewhere (Leigh, Robins, & Welkowitz, 1988), but mainly involved rewriting items to a fourth or fifth grade vocabulary level. This simplification was intended to minimize the influence of English language deficiencies on the participants' responses on psychological questionnaires (Bodner & Johns, 1977; Garrison, Tesch & DeCaro, 1978). The Beck Depression Inventory (BDI) is a 21-item selfreport questionnaire that assesses depressive symptomatol-
ogy (Beck, 1967). Bumberry, Oliver, and McClure (1978)
found the BDI to be valid for a university population according to clinical ratings made by a psychiatrist blind to BDI
scores. Split-half reliability of .86 was reported. ScoresoflO18 are considered to represent mild depression, 19-25 moder-
ate and 26 and above severe depression. The Sociotropy-Autonomy Scales (SAS) consist of two 30item self-report scales of Sociotropy and Autonomy (Beck, Epstein, Harrison, & Emery, 1983). Subjects rate on a 0 to 4 scale how much each statement applies to them. Factor analyses have suggested that the Sociotropy scale measures three factors, labeled Concern about Disapproval, Attachment and Pleasing Others, and that the Autonomy scale measures three factors, labeled Achievement, Freedom from Control and
Preference for Solitude. Robins (1985) reported that in a sample of 424 undergraduates, coefficient alpha was .90 for Sociotropy and .80 for Autonomy. Test-retest stabilities over
a four- to six-week period were .75 and .69, respectively.
The Parental Bonding Instrument (PBI) consists of 25 items describing parent behaviors and attitudes (Parker et al., 1979). Subjects rate each statement on a four-point scale for how well it matches their recollection of their parents when they were children. Two scale scores are computed for care and overprotection. Parker et al. (1979) reported split-half reliabilities of .88 and .74, respectively, and test-retest stabilities of .76 and .63, respectively. PBI ratings of maternal attitudes have been validated against the mothers' own reports (Parker, 1981). The present study dealt only with reports of
sive symptoms?; and 3) How are the personality characteris-
recollections of maternal behavior and attitudes.
tics of sociotropy and autonomy related to perceived parental
childrearing practices of deaf and hearing individuals? Method
Subjects A normally-hearing group of 62 female and 50 male undergraduates who participated in partial fulfillment of course requirements and a deaf group of 51 females and 51 males attending the National Technical Institute for the Deaf (NTID) comprised the sample. Deaf subjects met the following criteria: hearing loss greater than 80 dB; onset of deafness at age 250
Of the 112 hearing subjects, 56 completed the original
versions and 56 the revised versions of the measures. The 102
deaf students all completed the revised versions. AU measures were given in counterbalanced order across subjects. Results Instrument Revisions
Table 1 shows the internal consistencies, assessed by Cronbach's alpha, of the revised versions of all the measures for the AAD/October 1989
deaf and hearing samples separately, along with the internal
consistencies of the original versions, both in the present hearing sample thatreceived those versions, andinthe origin-
al validation samples described by the developers of the instruments.
The revised instruments all had acceptable levels of internal consistency with the hearing sample comparable to those of the original instruments. Internal consistencies in the deaf
sample were also good, although the revised BDI had only moderate internal consistencies which were significantly
lower than the hearing group, t (156) = 3.17, Ï• < .01.
Differences Between Deaf and Hearing Samples An analysis of variance (ANOVA) of BDI scores was conducted as a function of gender and hearing status. No effect of gender, nor of the gender Ï‡ deafness interaction was documented. Deafness showed a significant effect, F (1,210) = 6.55, Ï• < .05. Deaf subjects were significantly more depressed, as a group, than hearing subjects. A frequency breakdown of the number of deaf and hearing subjects falling in each range of depressive symptom severity is shown in Table 3. Of the 112 hearing subjects, 37 63 percent) and 52 of the 102 deaf subjects (51 percent) were at least mildly depressed based on a BDI score of 10 or greater. This group difference was significant, x2 (1) = 6.93, Ï• < .01.
Internal Consistencies of Original and Revised Instruments Instrument
Hearing Original (Î• =5Î˜
Deaf Revised (N=102)
______________________Table 3______________________ Classification of Depression Scores According to Beck's Scoring Criteria BDIScore
Parental Bonding Instrument Care
a: Split-half Pearson correlation coefficients. All others are Cronbach's alpha. Means and standard deviations on all the measures for the
three groups are shown in Table 2. The two hearing samples that received either the original or revised versions of the instruments did not differ on any measure except PBI Overprotection, on which the revised measure generated lower scores, /(HO) = 2.79, Ï• < .01. Data analyses involving the hearing subjects therefore combined both groups, except for those analyses involving overprotection, which were conducted for each group separately as well.
______________________Table 2______________________ Mean Scores of Deaf and Hearing Subjects on all Measures Variable
Hearing Original Hearing Revised Deaf Revised M
BDI PBI Care
Overprotection SAS Sociotropy
7.28 6.88 8.06 69.90 18.30
7.22 28.08 8.06 8.84 6.11 66.32 14.41
10.03 5.44 29.45 5.63 9.92 5.54 72.09 14.34
20.13 3389 15.75 74.68 33.02
8.18 8.58 4.64 11.72 5.67
18.61 3304 14.61 73.74 33.11
6.52 6.26 4.63 10.84 5.81
20.89 6.18 35.09 6.77 16.13 3.73
27.93 6.01 13.96 3.58
7.71 28.08 14.95
Pleasing Others SAS Autonomy Achievemnt Freedom from Control Prefer Solitude
50(49) 44(43) 8(8) 0(0)
Beck Depression Inventory
0-9 (Normal range) 10-18 (Mildiy Depressed) 19-25 (Moderately depressed) 26 and above (Severely depressed)
Hearing N(%) 75(67) 30(27) 4(4) 3 6)
Sociotropy and Autonomy scores of the two groups were compared by i-tests. The deaf subjects scored significantly higher on Sociotropy than did the hearing subjects, Ã- (212) = 1.93, Ï• = 05. They did not differ on the overall Autonomy factor, t (212) = 1.52, but deaf subjects scored significantly higher on the subfactor assessing need for Freedom from Control, i(212) = 5.49, Ï• < .001, and significantly lower on Preference for Solitude, t (212) = 2.90, Ï• < .01. On the PBI variables, deaf and hearing subjects did not differ significantly on the perceived Care dimension, but did differ significantly on perceived Overprotection, /(212) = 3.47, Ï• < .001, the hearing group scoring higher than the deaf. However, comparing deaf subjects with just those hearing subjects receiving the revised PBI, the two groups did not differ significantly on either Care or Overprotection.
Relations of BDI to Personality and Perceived Parenting
The zero-order correlations of BDI scores with PBI Care and
Overprotection, Sociotropy and Autonomy are shown for the hearing and deaf samples in Table 4. Gender was not significantly related to any of the above variables. BDI was nega-
tively associated with perceived Maternal Care on the PBI in both groups, more strongly so in the hearing subjects, t (212) = 2.00, Ï• < .05. BDI was positively associated with perceived maternal Overprotection in both groups, and with Sociotropy in the hearing group but not in the deaf group. BDI was not related to Autonomy in either group.
The unique statistical effects of the perceived maternal variables and personality variables on depression in each
group were examined by a hierarchical multiple regression analysis. In this analysis, group (Hearing vs. Deaf) was en-
Care, Maternal Overprotection, Sociotropy and Autonomy were entered as a set, and at the third step were entered the
tered at the first step. At the second step, scores on Maternal
interactions of these variables with the group AAD / October 1989
tiple regression analyses in which group was entered as the first step, followed by PBI Care and Overprotection as a set, and finally by the interaction effects. Results of these analy-
Pearson Correlations betwen Depression, Perceived Maternal Characteristics and Personality, for Deaf and Hearing Samples BDI BDI
0.13 0.39e 0.14
ses are shown in Table 6.