Resiliency, social support, and coping in rural low-income Appalachian adolescents from two racial groups

Journal of Adolescence 2000, 23, 693–703 doi:10.1006/jado.2000.0353, available online at http://www.idealibrary.com on

Resiliency, social support, and coping in rural low-income Appalachian adolescents from two racial groups CAROL A. MARKSTROM, SHEILA K. MARSHALL AND ROBIN J. TRYON In this study an attempt was made to examine social support and coping in relation to resiliency among a sample of rural, low-income Appalachian adolescents of AfricanAmerican and White races. Fifty-three African-American and 60 White 10th grade students completed the Perceived Social Support Scale for Family and Friends, the Ways of Coping Checklist, and Psychosocial Inventory of Ego Strengths. Utilizing regression analyses, it was shown that social support from family and problem-focused, avoidance, and wishful-thinking forms of coping significantly predicted resiliency. Race and gender distinctions were minimal in respect to the measured variables in the study. Research in the future should include continued examination of rural and diverse samples, the use of various sampling procedures, and longitudinal assessments of resiliency. # 2000 The Association for Professionals in Services for Adolescents

Introduction Resiliency is an adaptive, stress-resistant personal quality that allows the individual to thrive despite unfortunate life experiences. Resilient or invulnerable children and adolescents have greater likelihoods of remaining invincible and developing into competent adults (Werner and Smith, 1982). Terms used to describe the resilient include invulnerables, resistants, thrivers, invincibles, and inoculated. Maladaptive, vulnerable, distressed, and succumbers are labels indicative of those who lack resiliency. Resiliency has been measured according to a variety of constructs, such as, self-esteem, academic performance, physical health, coping and adaptation, and the absence of psychopathology or delinquent behaviors. Given the array of variables used to assess resiliency, confusion surrounding its meaning is not surprising. According to Zimmerman and Arunkumar (1994), at the very least, resiliency ‘‘. . . is a multidimensional phenomenon that is context-specific and involves developmental change’’ (p. 4). Research on resiliency is significant because efforts are made to identify processes of both vulnerability and protection that should help explain why and how resiliency is exhibited (Rutter, 1987). These processes may: (a) diminish or enhance the wellbeing of individuals; (b) diminish or support human strengths; or (c) serve as moderators or mediators that either heighten or reduce the potential of undesirable or negative outcomes in development. Markstrom et al. (1997) describe how, cumulatively, Erikson’s (1964, 1985) eight ego strengths can be construed as an assessment of overall ego strength or resiliency. Such a rationale was given because Erikson’s definition of ego strength bears a strong similarity to Reprint requests and correspondence should be addressed to: Carol A. Markstrom, Division of Family and Consumer Sciences, West Virginia University, Morgantown, WV, 26506-6124, U.S.A. An earlier version of this paper was presented at the Biennial Meetings of the Society for Research in Child Development, Washington, D.C., April 4, 1997. Published with the approval of the Director of the West Virginia Agricultural and Forestry Experiment Station as Scientific Article #2614. Research was supported with funds appropriated under the Hatch Act (Project #373). 0140-1971/00/060693+11$35
00/0

# 2000 The Association for Professionals in Services for Adolescents

694

C. A. Markstrom et al.

the construct of resiliency. For instance, ego strengths are internal qualities indicative of psychosocial health and well-being, as well as the absence of psychopathology. Later development is enhanced from successful incorporation of each strength into the ego. Ego strengths provide the individual with an added edge for approaching life and its difficulties. The present study investigates the relationship between ego strengths and factors that are likely to prove salient to resilient outcomes. In his categorization of invulnerables, Anthony (1987) identified the ‘‘true invulnerables’’ as individuals who had been exposed to high risk, yet seemed to persist and grow from each subsequent adversity. Summarizing the writings of others (e.g. Bornstein et al., 1973), Anthony (1987) stated that some stressful events may ‘‘inoculate’’ against later stressors and enhance resiliency. Two factors that place adolescents at greater risk were considered in this study: gender and/or racial minority status. Gender differences have been observed in reports of resiliency with girls appearing less susceptible and more resilient than boys (Werner and Smith, 1982; Rutter, 1985). Rutter (1987) reported males to be more vulnerable to a variety of risks and specifically noted that boys exhibited higher rates of disorders than girls when exposed to family discord. Interestingly, Seiffge-Krenke (1995) found that adolescent females reported more stressors than males and reported the same events to be more stressful than did males. If one argues that females perceive more stress than males, then they should be more resilient according to the Anthony hypothesis (1987). Taking this line of reasoning, the first prediction of the study was that: (a) the female gender would be more strongly associated with resiliency than the male gender. Those of racial minority status are placed at greater risk because of stress associated with experiences of prejudice and discrimination. Winfield (1995) argued that structural factors related to inequality and restricted access to resources contributes to vulnerability of AfricanAmericans. Although no significant racial differences between the resilient children and two comparison groups were found in Werner and Smith’s longitudinal study (1982), there was evidence to support that racial differences did exist in coping skills and social support. Based on the notion that resiliency is more likely to be observed in those who have been challenged and stressed, African-American should demonstrate greater resiliency. It was hypothesized, therefore, that: (b) the African-American racial status would be more strongly associated with resiliency than the White racial status. In addition to risk factors, it is necessary to explore protective and vulnerability mechanisms–processes that operate on ends of a continuum and either protect against risk or contribute to acquiescence to a risk (Rutter, 1987). Such processes comprise influences from adolescent temperament, positive family characteristics (e.g. warmth and cohesion), and external sources of social support (e.g. a teacher or church) (Garmezy, 1991). If factors are identified as contributors toward resiliency, then potentially they can be made accessible to all children. Prior research on resiliency (e.g. Werner and Smith, 1982; Garmezy, 1983) and SeiffgeKrenke’s (1995) theoretical model of adaptation point to both social support and coping styles as important in the development of resiliency. Adolescents’ usage of these processes may affect how they approach and adjust to stress and, therefore, have further impact on the outcome of resiliency. The presence of social support maximizes the probability that an individual will utilize problem-solving techniques to resolve conflict and demonstrate resiliency (Licitra-Kleckler and Waas, 1993). Social scientists agree that social support is closely linked to coping and both are highly relevant to resiliency. Seiffge-Krenke (1995) conceived social support from parents and peers

Resiliency in Appalachian adolescents

695

as predictive of coping. In this report, consistent with Folkman and Lazarus (1985), social support was regarded as one form of coping. It is the ability to seek and receive social support that makes it a coping skill. However, some coping behaviours are conceived as not particularly healthy or useful, as in the case of wishful-thinking and avoidance (Folkman and Lazarus, 1985). Based on the discussion of social support and coping, two hypotheses were generated. It was predicted that: (c) resiliency would be positively associated with social support from family and from friends and problem-focused coping skills; and (d) resiliency would be negatively associated with wishful-thinking and avoidance forms of coping. Gender and race differences have been reported in relation to coping and social support. Halstead, Johnson, and Cunningham (1993) found that females used social support and wishful-thinking and males used avoidance strategies to cope. Seiffge-Krenke (1995) summarized findings from several studies and concluded that ‘‘female adolescents show a greater willingness to resort to social contacts when coping with problems. . . whereas males are more likely to play down the stressful nature of a problem situation’’ (p. 14). With respect to race, African-American children were more likely than White children to recognize that stressful situations could be changed by using more coping responses (Halstead et al., 1993). Munsch and Wampler (1993) reported that African-American and Mexican-American adolescents viewed adults and young relatives as sources of social support significantly more often than did White adolescents and African-Americans and Mexican-Americans indicated that they received more problem-solving assistance in coping. Prior research provided enough evidence to warrant examination of gender and race distinctions in relation to social support and coping. Similarly, prior findings suggest the importance of examining the role of gender and race in analyses of relations between social support, coping and resiliency. However, formal hypotheses were not generated for this portion of the study. The participants in this study were rural, low-income Appalachian adolescents. Rural adolescents are under-investigated, and it is unknown whether the relationships between race and gender, social support, coping and resiliency are similar under conditions of rural-related risk. Rural living poses many barriers because of, among other influences, inadequate delivery of mental health services, changes in economic structures and economic instability in rural communities, the invisibility of rural poverty, declining close-knit family structures, and loneliness due to geographical isolation (Cutrona, Halvorson and Russell, 1996). In respect to low-income status, participants lived in West Virginia, a state located entirely in the Appalachian region (a highland area in the eastern United States commonly characterized as rural and poor). The undesirable outcomes of poverty are well-documented in the literature (e.g. Huston, 1991; Routh, 1994; Dadds, 1995; McLoyd, 1998).

Methods Sample selection Several screening criteria were used to obtain the purposefully selected sample in this study. Participants needed to be: (a) African-American or White with both biological parents of their race; (b) living in rural areas or small towns in countries that did not contain metropolitan statistical areas; (c) from families whose household incomes were lower than the West Virginia median of $20,795; and (d) in the 10th grade as of Fall 1994. This study was conducted in five

696

C. A. Markstrom et al.

southern countries of West Virginia, all identified because poverty was recognized as especially problematic (Office of Health Services Research, 1994). Additionally, these five countries were selected because they had relatively high numbers of rural African-Americans. Students at 11 different schools in the five target counties were contacted at the end of their 9th grade. Research assistants described the research study to group of students and requested names, telephone numbers, addresses, and parental background information from interested students. A total of 892 students from the 11 rural schools completed background information forms. Two-thirds of that number were screened out of the study due to incomes above the West Virginia median. The remaining one-third were pursued for potential participation in the study. Of that number, the various reasons for non-participation were: (a) the family did not have a telephone and did not return a mailed questionnaire requesting a home visit; (b) the adolescent was biracial (i.e. both parents were not African-American or White; (c) the family had made a geographical move; (d) the adolescent and/or his or her parents did not want to participate in the study; or (e) the family agreed to be in the study, but later changed their mind. The final sample consisted of 60 white (42 females and 18 males) and 53 AfricanAmerican (34 females and 19 males) 10th grade students. For both races, the range of ages were 14 through 17 and the mean age was 15.2.

Instrumentation Although participants and their parents completed an assortment of questionnaires and interviews in their homes, only those completed by the adolescents and relevant to the present investigation are described.

Demographics.

Gender and race were assessed through respondents’ self-identifications.

Psychosocial Inventory of Ego Strengths (PIES). The PIES (Markstrom et al., 1997) was used as an assessment of resiliency. The original measure consisted of 64 items designed to measure Erikson’s (1964, 1985) eight ego strengths (i.e., hope, will, purpose, competence, fidelity, love, care, and wisdom). A reduced version of 32 items was used in the present study. Responses were based on a Likert-type scale ranging from a score of 5, dose not describe me well, to a score of 1, describes me very well. The total ego strength score has been shown to be positively related to self-esteem, internal locus of control, and purpose in life, and negatively related to feelings of hopelessness (Markstrom et al., 1997). In the present study, internal consistency was shown with Cronbach’ s alphas of 0?81 for Whites and 0?78 for African-Americans. Perceived Social Support (PSS) from family and friends. Adolescents reported their perceptions of social support from family and from friends using 20 items to measure family support and 20 to measure friend support. Response options were yes, no, or don’ t know. Examples of questions from these instruments are ‘‘My family (friends) give me the moral support I need’’ and ‘‘I rely on my family (friends) for emotional support’’. Procidano and Heller (1983), the authors of these measures, found that both PSS-Family and PSSFriends were inversely related to symptoms of distress, with PSS-Family having a stronger relationship. Scale scores were derived according to Procidano and Heller’s protocol by coding no or don’t know responses as ‘‘0’’ and yes responses as ‘‘1’’ and then summing item scores within each scale. Internal consistency for White subjects in the present study were 0?84 for PSS-Family and 0?82 for PSS-Friends. Among the African-Americans, Cronbach’s alphas were 0?67 for PSS-Family 0?78 and for PSS-Friends.

Resiliency in Appalachian adolescents

697

Ways for coping checklist. The original measure consisted of 66 items (Folkman and Lazarus, 1980). Eight subscales were defined by Folkman et al. (1986), and were later delineated by Halstead et al. (1993) according to four-factors of problem-focused, seeks social support, wishful-thinking, and avoidance. Coping items are responded to in terms of frequency of use on a four-point Likert-type scale ranging from is not used to used a great deal. In the present study, 35 of the original items with the highest item-total correlations from the Halstead et al. (1993) study were utilized to devise a shortened measure. Cronbach’s alphas for Whites and African-Americans were, respectively, 0?83 and 0?92 for problem-focused, 0?71 and 0?93 for wishful-thinking, and 0?53 and 0?57 for avoidance. The seeking support subscale was not used in the present study. The PSS-Family and PSS-Friends were preferred measures of social support because they allowed for the delineation of two separate sources of support pertinent to adolescents.

Results Preliminary analyses examined the roles of gender and race in resiliency, social support, and coping using a series of univariate 262 ANOVAs. The only significant findings were an effect for gender6race in respect to resiliency, F (1,109) = 4?16, p50?05, and a main effect for race in relation to social support from friends F (1,109) = 4?30, p50?05. Both White males and African-American females scored highest in resiliency (M = 118?3, S.D. = 19?0 and M = 117?2, S.D. = 14?8, respectively). African-American males and White females scored lowest in resiliency (M = 110?4, S.D. = 15?4 and M = 112?1, S.D. = 15?3, respectively). In relation to social support from friends, Whites scores higher (M = 53?6, S.D. = 5?5) than African-Americans (M = 51?1, S.D. = 6?6). Hypothesis (a) stated that the female gender would be more strongly associated with resiliency than the male gender, and hypothesis (b) stated the African-American racial status would be more strongly associated with resiliency than the White racial status. The finding of an interaction between gender and race reveals a more complex relationship than anticipated. In the other measured variables, no other gender differences emerged, and there was only one race difference. Next, zero-order correlations were performed between social support, coping, and resiliency with the whole sample (see Table 1) and then conducted separately for each gender and race (see Tables 2 and 3). Fisher’s z-scores were calculated to test whether the correlations between the variables were significantly different between males and females. No Table 1

Intercorrelations between social support, coping and resiliency (total sample, n = 113)

Subscale

Social Support-Friends Problem Focused Wishful Thinking Avoidance Resiliency *p50?05. **p50?01.

Social support Family

Friends

0?34* 0?19* 70?20* 70?14 0?41**

0?31** 70?11 70?18 0?41**

Problem focused

Wishful thinking

0?31** 0?26** 0?36**

0?30** 70?25**

Avoidance

70?26**

C. A. Markstrom et al.

698

significant differences were found. Next, differences in correlations between variables were tested between African-American and White adolescents. Only the relationship between wishful thinking and resiliency was found to be significantly different (z = 71?97, p50?05); a significant negative correlation (r = 70?44, p50?01) was found for African-American respondents and a non-significant relationship was revealed for White adolescents (r = 70?09, n.s.). It should be noted that this significant relationship may have been a chance finding since a total of 30 comparisons were made (including both gender and race comparisons). Standard multiple regression analyses were performed to examine the roles of social support and coping in the prediction of resiliency. Hierarchical regressions were conducted with each predictor entered separately. These regressions were repeated with each predictor, in turn, entered last in order to obtain the unique (non-shared) variance explained by that variable. Sr2 values were obtained to examine the unique contribution to R2 from each independent variable. Sr2, which is obtained by squaring the partial correlation, is thought to be the best indicator of the importance of an independent variable in standard multiple regression (Tabachnick and Fidell, 1996). It can be observed in Table 2 that problem-focused coping had the single largest unique variance contributing to the prediction of resiliency in the overall sample (Sr2 = 0?12). For the overall sample, hypothesis (c) was supported to the extent that problem-focused coping contributed the most unique variance in the prediction of resiliency. Next, we explored the relationships between the predictors and resiliency when the sample was split by gender. The strategies described above were used to determine the unique variance of each independent variable in predicting resiliency for males and females separately. Results displayed in Table 3 indicate gender differences in the patterns of relationships between social support and coping and resiliency. For males, the predictors with the largest unique variance contributing to the prediction of resiliency were problemfocused coping (Sr2 = 0?09) and wishful thinking (Sr2 = 0?09). Amongst females, the predictors with the largest Sr2 were problem-focused coping (Sr2 = 0?12) and avoidance (Sr2 = 0?10). Similarly, separate hierarchical regressions were conducted for African-Americans and Whites to explore the relationships between the predictors and resiliency. Results (see Table 4) indicate different patterns for each race. Amongst African-American respondents, wishful thinking had the largest unique variance contributing to the prediction of resiliency (Sr2 = 0?13) while problem-focused coping accounted for the greatest amount of unique variance (Sr2 = 0?12) amongst White participants. Table 2

Regression analysis for the prediction of resiliency, (total sample n = 113)

Variable Social support–Family Social support–Friends Problem-focused coping Wishful-thinking coping Avoidance coping Overall R2 = 0?41. *p50?05. **p50?01.

B

SE B

0?43 0?32 0?81 70?76 71.06

0?18 0?22 0?17 0?25 0?36

b 0?20 0?14 0?42 70?25 70?24

t-value

Sr2

2?41* 1?64 4?75** 73?01** 72?91**

0?03* 0?02 0?12** 0?05** 0?05**

Resiliency in Appalachian adolescents

699

Table 3

Regression analysis for the prediction of resiliency by gender

Variable

Females (n = 76) B

SE B

b

Males (n = 37)

t-value

Sr

Social support–family 0?44 0?19 0?22 2?31* Social support–friends 0?29 0?26 0?11 1?10 Problem-focused 0?78 0?20 0?39 3?89** Wishful-thinking 70?56 0?30 70?18 71?82 Avoidance 71?45 0?43 70?34 73?38**

2

B

SE B

b

Sr2

t-value

0?04* 0?33 0?44 0?12 0?76 0?01 0?48 0?45 0?19 1?07 0?12** 0?80 0?36 0?43 2?23* 0?03 71?11 0?48 70?36 72?29* 0?10** 70?45 0?75 70?09 70?60

0?01 0?02 0?09* 0?09* 0?01

Overall R2 for males = 0?45 and females = 0?42. *p50?05. **p50?01.

Table 4

Regression analysis for the prediction of resiliency by race

Variable

African-American (n = 53) B

SE B

b

t-value

Social support–family 0?30 0?29 0?12 1?00 Social support–friends 0?52 0?30 0?23 1?76 Problem-focused 0?69 0?26 0?34 2?68** Wishful-thinking 71?06 0?31 70?38 73?45** Avoidance 71?14 0?49 70?27 72?32

White (n = 60) Sr

2

B

SE B

b

t-value

0?01 0?56 0?24 0?28 2?31* 0?03 0?19 0?37 0?06 0?52 0?08* 0?82 0?25 0?44 3?32** 0?13** 70?33 0?43 70?10 70?77 0?06* 71?01 0?55 70?22 71?83

Sr2 0?06* 0?00 0?12** 0?01 0?04

Overall R2 for African-American = 0?49 and White = 0?39. *p50?05. **p50?01.

Discussion The primary focus of this study was the prediction of resiliency according to perceived social support and coping skills. This research was conducted with an understudied population, Appalachian youth. These youth are not only understudied, but also considered at-risk because of their rural residence and low-income status. Risks posed by gender and race also were considered in the examination of resiliency. We suggested, according to the Anthony (1987) hypothesis, that if females and AfricanAmericans perceive more stress, they are placed at higher risk and may therefore report greater resiliency. This hypothesis was supported for African-American females only. The finding that African-American males and White females scored the lowest in resiliency was contrary to the hypothesis. Perhaps it is the combination of being both African-American and female and associated stressors that activates mechanisms that promote resiliency. What is less easily explained is that White males also scored high on the measure of resiliency. It may be that White males’ higher scores were reflective of what Anthony called ‘‘pseudoresilients’’ – those who appear resilient, but had not yet been put to the test. These conjectures are highly speculative, however, because assessments of ‘‘true resilients’’ as opposed to ‘‘pseudo-resilients’’ have not yet been developed. Gender and race differences in the subscales of social support and coping were minimal. The only significant difference was that Whites scored higher than African-Americans

700

C. A. Markstrom et al.

in social support from friends. The minimal differences in significance may be due to the fact that the sample shared backgrounds of rural lower-income households and this defined them as more similar than different, at least in respect to the variables in the study. Perceived social support from friends was not predictive of resiliency. These findings are of interest because it has been consistently stated in the literature that perceived social support, regardless of the source, is associated with greater resiliency (Sarason et al., 1983; Werner and Smith, 1982). Perhaps the fact that participants lived in rural locations limited the amount of contact they had with friends. Indeed, geographical isolation from friends has been raised as an issue for rural youth (Cutrona et al., 1996). For White adolescents and females, familial social support may then have had even greater meaning because of geographical distance from peers. Interestingly, our findings indicated that social support was not significantly associated with resiliency for African-Americans and males when entered into the same regression equation as the other coping strategies. For these youth, neither sources of social support (friends or family) were associated with resiliency. However, it has been suggested that social support maximizes the probability that an individual will use problem-solving techniques to overcome difficulties (Licitra-Kleckler and Waas, 1993). Social support and problemsolving shared variability in the explanation of resiliency, indicative of the possibility that the use of social support enhances problem-focused coping in African-Americans and males. Problem-focused coping was found to be a strong predictor of resiliency and accounted for the most unique variability. Those who cope in direct, problem-solving modes enhance the possibility that life’s difficulties will be resolved successfully. Perhaps these individuals reflect greater ego strength, as discussed by Erikson (1964, 1985), which would be reflective of greater inner resolve. To engage in problem-focused coping, the individual must employ strategies that involve facing problems and considering how to resolve issues. As such, individuals have the opportunity to gain confidence in facing difficulties and to accumulate successful skills in coping with issues. Wishful-thinking and avoidance, negative styles of coping, were significantly negatively associated with resiliency. (Although the low Cronbach’s alphas of the avoidance subscale reduce confidence in the findings). Negative forms of coping do not eliminate problems and offer only temporary escape from difficulties. The individual may be continually faced with unresolved issues that can weaken inner resolve and self-confidence. It is not surprising, then, that there is limited presence of resiliency in such cases. These two negative forms of coping did not contribute the same degree of unique variability and consistency across gender and race in predicting resiliency as did problemfocused coping. These two coping strategies were predictive of lower resiliency only amongst African-American but not White adolescents. Wishful thinking was predictive of lower resiliency for males and not females, while avoidance was negatively associated with resiliency amongst females but not males. Wishful-thinking and avoidance strategies may be employed for types of problems that are unresolvable or that may disappear with time. Therefore, it is possible that individuals may successfully utilize wishful-thinking or avoidance for certain types of situations. As such, wishful-thinking and avoidance may then be weaker predictors of resiliency in contrast to problem-focused coping. Further, the relationships between wishful thinking and avoidance and resiliency may differ by gender and race because of the possible differences in the types of problems each face.

Resiliency in Appalachian adolescents

701

The fact that participants were similar in respect to potentially confounding subject variables of socio-economic status and rural residence strengthened the internal validity of the study. Further, this study employed a measure of resiliency rather than a proxy for resilience. In contrast, some other studies have assessed resiliency according to a variety of constructs such as self-esteem, academic performance, physical health, coping and adaptation, and the absence of psychopathology or delinquent behaviors. However, a limitation of this study is evidenced in the use of purposeful sampling, a strategy common to many other studies on this topic. The levels of social support, coping, and resiliency that operated among those who did not engage in the study are unknown. The fact that potential participants with telephones were more likely to be scheduled for home-visits also may have targeted a certain portion of the population. Purposeful sampling of rural African-American adolescents dictated where school and home visits were to occur. These factors decreased the generalizability of the findings, and the results may not describe all adolescents living in low income families in rural West Virginia or Appalachia. Certainly, continued studies on the topic or resiliency, using various sampling strategies, are needed on rural adolescents. The study was not longitudinal or developmental, posing another limitation related to generalizability. It can only be suggested that social support and coping contributed to resiliency for adolescents at a particular time in their lives. Rutter (1988) identified several advantages of longitudinal research that apply to the study of resiliency, for instance: (a) to delineate more carefully the hypothesized risk factors (in prospective as opposed to retrospective research designs); (b) to understand how some individuals ‘‘escape’’ the risk process as well as to examine the heterogeneity of abnormal outcomes to specified risk factors; (c) to specify the ages of onset in respect to the correlates of disorders; (d) to examine intra-individual change; and (e) to examine the ascendance and descendance of the dependent variables associated with risk. In short, causality could only be suggested as one possible explanation of the findings emerging in the present investigation. Extant research findings suggest a number of relationships between resilience and various constructs, but the role of resilience as a cause or effect is unclear. For instance, a pertinent question for this and related investigations is whether social support and coping promote the outcome of resiliency or whether the presence of resiliency creates more positive coping styles and the ability to draw on social support?

References Anthony, E. J. (1987). Risk, vulnerability and resilience. In The invulnerable child, Anthony, E. J. and Cohler, B. J. (Eds). New York: Guilford, pp. 3–48. Bornstein, P. E., Clayton, P. J., Halikas, J. A., Maurice, W. L. and Robins, E. (1973). The depression of widowhood after thirteen months. British Journal of Psychiatry, 122, 561–566. Cutrona, C. E., Halvorson, M. B. J. and Russell, D. W. (1996). Mental health services for rural children, youth, and their families. In Families and the mental health system for children and adolescensts, Heflinger, C. A. and Nixon, C. T. (Eds). Thousand Oaks, CA: Sage, pp. 217–237. Dadds, M. R. (1995). Families, children, and the development of dysfunction. Developmental Clinical Psychology and Psychiatry, 32, 42–47. Erikson, E. H. (1964). Insight and responsibility. New York: Norton. Erikson, E. H. (1985). The life cycle competed. New York: Norton.

702

C. A. Markstrom et al.

Folkman, S. and Lazarus, R. S. (1980). Manual for the Ways of Coping (Revised) Questionnaire. California: Consulting Psychological Press. Folkman, S. and Lazarus, R. S. (1985). If is changes it must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48, 150–170. Folkman, S., Lazarus, R. S., Dunkel-Schetter, C., DeLongis, A. and Gruen, R. J. (1986). Dynamics of a stressful encounter: Cognitive appraisal, coping, and encounter outcomes. Journal of Personality and Social Psychology, 50, 992–1003. Garmezy, N. (1983). Stressors of childhood. In Stress, coping and development in children, Garmezy, N. and Rutter, M. (Eds). New York: McGraw Hill, pp. 43–84. Garmezy, N. (1991). Resiliency and vulnerability to adverse developmental outcomes associated with poverty. American Behavioral Scientist, 34, 416–430. Ge, X., Conger, R. D., Lorenz, F. O., Elder, G. H., Montague, R. B. and Simons, R. L. (1992). Linking family economic hardship to adolescent distress. Journal of Research on Adolescence, 2, 351–378. Halstead, M., Johnson, S. B. and Cunningham, W. (1993). Measuring coping in adolescents: an application of the Ways of Coping Checklist. Journal of Clinical Psychology, 22, 337–344. Hutson, A. C. (1991). Antecedents, consequences, and possible solutions for poverty among children. In Children in poverty, Hutson, A. C. (Ed.). New York: Cambridge University Press, pp. 282–315. Licitra-Kleckler, D. M. and Waas, G. A. (1993). Perceived social support among high-stress adolescents: The role of peers and family. Journal of Adolescent Research, 8, 381–402. Markstrom, C. A., Sabino, V. M., Turner, B. and Berman, B. C. (1997). The psychosocial Inventory of Ego Strengths: Development and validation of a new Eriksonian measure. Journal of Youth and Adolescence, 26, 705–732. McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 52, 185–204. Munsch, J. and Wampler, R. S. (1993). Ethnic differences in early adolescents’ coping with school stress. American Journal of Orthopsychiatry, 63, 633–646. Office of Health Services Research. (1994). 1992 Population estimate. West Virginia State Census Data Center Newsletter. Morgantown, WV: Author. Procidano, M. E. and Heller, K. (1983). Measure of perceived social support from friends and from family: Three validation studies. American Journal of Community Psychology, 11, 1–24. Reaves, C. C. (1992). Quantitative research for the behavioural sciences. New York: John Wiley & Sons. Routh, D. K. (Ed.). (1994). The impact of poverty on children, youth, and families {Special issue}. Journal of Clinical Child Psychology, 23. Rutter, M. (1987). Resilience in the face of adversity: Protective factors and resistance to psychiatric disorder. British Journal of Psychiatry, 147, 598–611. Rutter, M. (1985). Psychosocial resilience and protective mechanisms. American Journal of Orthopsychiatry, 57, 316–331. Rutter, M. (1988). Longitudinal data in the study of causal processes: Some uses and some pitfalls. In Studies of psycho-socio risks: the power of longitudinal data, Rutter, M. (Ed.). New York: Cambridge University Press, pp. 1–28. Sarason, I. G., Levine, H. M., Basham, R. B. and Sarason, B. R. (1983). Assessing social support: The Social Support Questionnaire. Journal of Personality and Social Psychology, 44, 127–139. Seiffge-Krenke, I. (1995). Stress, coping and relationships in adolescence. Mahwah, NJ: Lawrence Erlbaum. Simons, R. L., Whitbeck, L. B. and Wu, C. I. (1994). Resilient and vulnerable adolescents. In Families in troubled times: Adapting to change in rural America, Conger, R. D. and Elder, G. H. Jr. (Eds). New York: Aldine De Gruyter, pp. 223–234. Skinner, M. L., Elder, G. H. and Conger, R. D. (1992). Linking economic hardship to adolescent aggression. Journal of Youth and Adolescence, 21, 259–276. Tabachnick, B. G. and Fidell, L. S. (1996). Using multivariate statistics. 3rd Edn. NY: Harper Collins. Werner, E. E. and Smith, R. S. (1982) Vulnerable but invincible: A longitudinal study of resilient children and youth. New York: McGraw Hill.

Resiliency in Appalachian adolescents

703

Winfield, L. F. (1995). The knowledge base on resilience in African-American adolescents. In Pathways through adolescence: Relation to social contexts, Crockett, L. J. and Crouter, A. C. (Eds). Mahwah, NJ: Lawrence Erlbaum, pp. 87–118. Zimmerman, M. A. and Arunkumar, R. (1994). Resiliency research: implications for schools and policy. Society for Research in Child Development, 8, 1–19.

View publication stats