ADOLESCENT GRIEF: RELATIONSHIP CATEGORY AND...

23
OMEGA, Vol. 54(2) 147-167, 2006-2007 ADOLESCENT GRIEF: RELATIONSHIP CATEGORY AND EMOTIONAL CLOSENESS* HEATHER L. SERVATY-SEIB M. CAROLE PISTOLE Purdue University, Indiana ABSTRACT Bereaved adolescents (N = 90) who had experienced relatively common death losses (e.g., grandparent, friend) completed the Texas Revised Inventory of Grief and the Emotional Closeness Scale and Continuum. Results indicated that present grief was significantly higher for friend than for grandparent death loss. A MANOVA revealed that those in the high closeness group reported significantly higher mean scores on past and present grief than those in the low closeness group. Finally, in a hierarchal multiple regression, after demographic variables were entered (e.g., age, present at death), emotional closeness added significant variance to the prediction of past and present grief. This research contributes to the understanding of grief intensity following adolescents’ most common death losses and highlights the importance of counselors’ intentionally and directly assessing bereaved adolescents’ perceived emotional closeness to the deceased as part of grief-related counseling. Because the bulk of the adolescent grief literature has focused on parent or sibling death (e.g., Balk, 1990; Dillon & Brassard, 1999; Fanos & Nickerson, 1991; Martinson & Campos, 1991; Worden & Silverman, 1996), little is known about the grief associated with the death losses of extended family members and peers or *This research was supported in part by a seed grant from Radford University, Radford, Virginia. 147 Ó 2006, Baywood Publishing Co., Inc.

Transcript of ADOLESCENT GRIEF: RELATIONSHIP CATEGORY AND...

OMEGA, Vol. 54(2) 147-167, 2006-2007

ADOLESCENT GRIEF: RELATIONSHIP CATEGORY

AND EMOTIONAL CLOSENESS*

HEATHER L. SERVATY-SEIB

M. CAROLE PISTOLE

Purdue University, Indiana

ABSTRACT

Bereaved adolescents (N = 90) who had experienced relatively common

death losses (e.g., grandparent, friend) completed the Texas Revised

Inventory of Grief and the Emotional Closeness Scale and Continuum.

Results indicated that present grief was significantly higher for friend than

for grandparent death loss. A MANOVA revealed that those in the high

closeness group reported significantly higher mean scores on past and present

grief than those in the low closeness group. Finally, in a hierarchal multiple

regression, after demographic variables were entered (e.g., age, present at

death), emotional closeness added significant variance to the prediction

of past and present grief. This research contributes to the understanding

of grief intensity following adolescents’ most common death losses and

highlights the importance of counselors’ intentionally and directly assessing

bereaved adolescents’ perceived emotional closeness to the deceased as

part of grief-related counseling.

Because the bulk of the adolescent grief literature has focused on parent or sibling

death (e.g., Balk, 1990; Dillon & Brassard, 1999; Fanos & Nickerson, 1991;

Martinson & Campos, 1991; Worden & Silverman, 1996), little is known about

the grief associated with the death losses of extended family members and peers or

*This research was supported in part by a seed grant from Radford University, Radford, Virginia.

147

� 2006, Baywood Publishing Co., Inc.

friends. Nonetheless, grandparent and peer are the more common death losses

during adolescence (Balk & Corr, 1996; Crenshaw, 1990; Oltjenbruns, 1996;

Schachter, 1991-1992). For instance, grandparent death is often the first death

that an adolescent experiences (Glass, 1990); and in one study (Ringler &

Hayden, 2000), 26 adolescents had lost a grandparent, 37 had lost a friend, and

14 had lost a grandparent and a friend; only 3 participants had lost a sibling,

and none had lost a parent.

Adolescent family and friend death losses are important because they are

common and because they are distressing. The responses to grandparent death

loss can be similar to the struggles associated with parent or sibling death loss

(Irizarry, 1992; Wass, 1995), and friend death loss often results in a more

profound grief than might have been predicted (Dyregrov, Gjestad, Bie Wikander,

& Vigerust, 1999; Schachter, 1991-1992; Sklar & Hartley, 1990). Examining

adolescent family and friend death loss, Lurie (1993) concluded that “the loss

of a close friend precipitates a grief reaction similar to that of the loss of a close

family member” (p. 203). Similarly, Ringler and Hayden (2000) concluded that

adolescents are “often deeply affected” (p. 227) by family and friend death

loss. If practitioners are to provide effective, useful services to bereaved adoles-

cents, research on the most common death loss experiences is mandatory.

This study examined grief intensity following family kin (e.g., grandparent) and

friend death loss.

Grief Intensity Perspectives—

Evolutionary Salience and Emotional Closeness

Evolutionary salience and emotional closeness are two interesting perspectives

that may account for adolescents’ grief intensity following kinship or friend death

loss. From the perspective of modern evolution theory, in natural selection,

“the genes and behavior that enhance reproductive success are selected for, not

the genes and behavior that promote survival” (Belsky, 1999, p. 141). Littlefield

and Rushton (1986) argued, therefore, that bereaved kin’s grief intensity would

reflect the degree of genetic investment in the deceased and the deceased’s

propagation potential. They found support for this proposal when investigating

grief associated with a child death loss, which is biologically costly because the

genes do not survive to be reproduced. For example, mothers, who have more

genetic investment in each child due to their reproductive capacity being more

limited than men’s, grieved significantly more intensely than fathers. In other

research, kinship has been found to predict grief intensity, and bereaved parents

have displayed more intense grief responses than adults experiencing other kin or

spousal death losses (Burnett, Middleton, Raphael, & Martinek, 1997; Sanders,

1979-1980; Zisook & Lyons, 1988). Cleiren, Diekstra, Kerkhof, and van der Wal

(1994) also found significant grief intensity differences among kin, “with parents

(particularly mothers) . . . of the deceased being more strongly affected than adult

148 / SERVATY-SEIB AND PISTOLE

children, brothers, and widows” (p. 22). And Ringdal, Jordhoy, Ringdal, and

Kaasa (2001) found that grief for close family members was stronger for female

(i.e., more genetically invested) versus male participants and for younger (i.e.,

more reproductive potential) versus older family members.

Another perspective argues that “differences in severity of grief reactions . . . are

likely to follow more accurately the closeness of personal relationships . . . than

anything that directly results from the degree of genetic relatedness” (Archer,

1999, p. 158), with closeness perhaps being confounded with kinship because

family members typically support and sustain daily existence. Consistent with

this thinking, many thanatological researchers (Brent et al., 1992; Cleiren, 1993;

McNeil, Silliman, & Swihart, 1991) have argued that closeness is relevant to

grief intensity and must be assessed (Cleiren et al., 1994; McIntosh, 1993;

Oltjenbruns, 1996; Swanson & Bennett, 1982-1983). In Bugen’s (1977) theory,

centrality, “the closeness of the relationship between the mourner and the

deceased” (p. 197), is a major predictor of grief intensity.

Rationale for the Study

In terms of hypotheses about adolescent grandparent or friend death loss,

a sparse clinical and anecdotal literature suggests that the grief responses to

grandparent death will vary according to the relationship’s closeness (Crenshaw,

1990; Raphel, 1983; Webb, 1993). Similarly, in the few adolescent friend loss

studies (Brent et al., 1992; McNeil et al., 1991) closeness was significantly related

to grief intensity. However, in other research, “the closeness of the student/peer

relationship did not seem to predict the intensity or duration of mourning”

(O’Brien, Goodenow, & Espin, 1991, p. 435). The paucity of research, the

inconsistency in findings, and the lack of kin and friend death loss comparative

studies sheds little light on the contribution of kinship or closeness as a salient

predictor of grief intensity.

At the simplest level of an evolutionary saliency perspective, kin are important

to the adolescent’s inclusive fitness, which in turn is relevant to reproductive

success (Simpson, 1999). Because the likely index of relatedness to grandparents,

aunts, and uncles is 25% (Archer, 1999), adolescents might display more grief

for kin, whose genetic heritage they share and with whom they are linked in

terms of evolutionary survival. In addition, due to grandparents no longer being

at an age where they are reproductively valued, there may be greater grief

intensity for aunts/uncles/cousins, who may still have reproductive value. If so,

adolescents would exhibit less grief for friends, with whom they have no genetic

heritage and no evolutionary survival link.

In contrast, from a closeness perspective, bereaved adolescents may display

greater grief intensity based on their emotional closeness to the deceased, whether

a grandparent, other kin, or a friend. Regarding closeness and kin death loss, we

reasoned that adolescents may or may not have been close to a family member,

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 149

particularly if the deceased lived geographically far away and if interaction was

more intermittent (e.g., twice a year) than continuous (e.g., daily, weekly). In

addition, for a grandparent, grief may be attenuated because the person is at

an age that seems old to the adolescent and so consistent with dying. More

specifically, the adolescent may not perceive the self as close to or emotionally

linked with someone at such a different phase of the life cycle. Nonetheless, if

kinship is a primary predictor of grief intensity, then adolescent grief intensity

would be higher for grandparents and kin death loss, whether or not closeness

is high; but if closeness is the salient predictor, then grief intensity will be

higher when closeness is higher, regardless of relationship category.

Regarding friend death loss, adolescents may have strong grief reactions

because the friendship is close (Oltjenbruns, 1996; Toray & Oltjenbruns, 1996;

Schacter, 1991-1992) and intense (Crenshaw, 1990; Davies, 1991; Raphael, 1983;

Schachter, 1991-1992). Adolescent friendships have adaptive, developmental

advantages linked to identity (Erikson, 1968; O’Brien et al., 1991; Oltjenbruns,

1996; Podell, 1989; Preto, 1999) as well as attachment (Allen & Land, 1999;

Hazan & Zeifman, 1999) and affiliative functions (e.g., shared daily activities,

caring, intimacy). For example, closeness may accrue in friendships as adoles-

cents’ primary social support shifts from parents to friends in early and middle

adolescence (Collins, 1997). In addition, developmental functions may culminate

in adolescents’ friends being important, influential, and connected to the self

(Berscheid, Snyder, & Omoto, 1989). Further, adolescent friend death may elicit a

comparison with the deceased that, because of perceived similarity, may lead

bereaved peers to confront their own vulnerability and mortality, thereby possibly

increasing the emotional intensity of the loss (Crenshaw, 1990; Podell, 1989;

Silverman, 2000; Smith, Lingle, & Brock, 1978-1979). These possibilities suggest

an implicit closeness in adolescent relationships, which could elicit high grief

intensity in response to friend death loss. If so, then grief may be more intense for

friend death loss, if closeness is higher versus lower.

This study examined the relatedness of grief intensity and closeness in

relation to death loss of family kin (i.e., grandparent and aunt/uncle/cousin) or

friend. For hypothesis one (H1), consistent with Archer’s (1999) contention

that closeness rather than kinship is salient to grief intensity, we expected that

mean scores on grief intensity would be significantly higher for friend death

loss than for grandparent or aunt/uncle/cousin death loss. The second hypothesis

(H2) was that adolescents who reported high levels of emotional closeness

would have significantly higher mean scores on grief intensity than those who

reported lower levels of emotional closeness, regardless of relationship category

(i.e., friend, grandparent, aunt/uncle/cousin). In addition, we asked a research

question (RQ): Does the level of emotional closeness predict grief intensity above

and beyond known biographical (i.e., age, sex, time since death, being with

the deceased at the time of death, and suddenness of the death) predictors of

grief intensity?

150 / SERVATY-SEIB AND PISTOLE

METHOD

Participants

Participants were 90 adolescents aged 14-18 years (M = 15.8, SD = 1.10) who

had experienced a death loss within the previous two-year period. The sample,

which was 77.8% female and 91.1% White, was recruited from one suburban

and two rural high schools in the southeastern region of the United States. The

findings presented here are one segment of a larger investigation. Approximately

half of the participants were from intact families (n = 46, 51.1%), while 14.4%

(n = 13) were from families with divorced parents and 16.7% (n = 15) were from

divorced families with both biological parents remarried. With regard to death

loss, 50% (n = 45) of the sample reported on the death of a grandparent, 25.5%

(n = 23) reported on the death of friend, 11.1% (n = 10) on the death of an aunt,

8.9% (n = 8) on an uncle, and 4.4% (n = 4) on a cousin. Time since death loss

ranged from five days to two years (M = 11.4 months, SD = 7.6). The primary

causes of death identified were cancer (n = 24, 26.7%), old age (n = 11, 12.2%),

car accident (n = 11, 12.2%), and heart attack (n = 10, 11.1%). Data were also

collected on several variables known to be associated with grief intensity includ-

ing participant age and sex, time since death, adolescent presence at the time of

death, and suddenness of the death.

Materials

Texas Revised Inventory of Grief (TRIG)

The TRIG (Fashingbauer, Zisook, & DeVaul, 1987) is a brief questionnaire

designed to quantify the intensity of negative grief reactions. A modified version

of the instrument was used in this research in order to make the items more

appropriate for bereaved adolescents who experienced friend or grandparent

death. Several items were revised to smooth readability and reduce repetition of

phrases (e.g., “the person who died”), and one item (i.e., “No one will ever take the

place in my life of the person who died”) was deleted because this item might

not apply to friend-death loss. Three items were added by being more specific

about the phrase “when I think of the person who died.” For example, “I still cry

when I think of the person who died” became “I cry when I think about how

this person used to be before his/her death” and “I cry when I think about how

he/she could have been if not for his/her death.” In the current research the sample

specific reliabilities on the TRIG subscale scores exceeded the original. Two

experts have verified the content and acceptability of the present version of the

TRIG (S. Zisook, personal communication, June 23, 2005; B. Hayslip, personal

communication, July 10, 2005).

TRIG-Past, which is Part I of the measure, consists of eight items assessing

grief-related feelings/behavior at the time of the death loss; a sample item is “After

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 151

this person died, I found it hard to get along with certain people.” TRIG-Present,

Part II, consists of 15 items, such as “It is painful to recall memories of how

he/she used to be.” which tap present grief-related feelings/ behavior. Participants

rated all items based on a 5-point Likert-type scale ranging from 5 = completely

true to 1 = completely false. Higher scores indicate greater negative grief intensity.

In terms of scores’ reliability, Fashingbauer et al. (1987) reported an alpha

coefficient of .77 and a split-half reliability of .74 for TRIG-Past scores, with a

coefficient alpha of .86 and a split-half reliability of .88 for TRIG-Present scores.

For the current sample, Cronbach’s alphas for the scores were: TRIG-Past = .82

and TRIG-Present = .93.

Emotional Closeness

Although a comprehensive discussion is beyond the scope of this article, the

measurement of closeness has been quite varied and fuzzy (Berscheid et al., 1989).

Bereavement researchers have not agreed on a measure of closeness and have

relied on either a single item measure (Grabowski & Frantz, 1992-1993; Park

& Cohen, 1993; Reed & Greenwald, 1991; Russac, Steighner, & Canto, 2002;

Swanson & Bennett, 1982-1983) or a multidimensional measure that includes

related constructs such as intimacy (Cleiren, 1993). Because of such difficulties, a

two-part measure was designed for this study to provide a concise, synthesized,

single factor focus for emotional closeness. As defined in the present investi-

gation, emotional closeness is the bereaved adolescent’s subjectively reported

level of emotional openness, awareness, and understanding in the relationship

with the deceased. Items were generated by two experts in closeness, as the term

is used in death loss research and practice. Face validity was determined by a

review of the items conducted by three other experts in the topic. The two parts

of the measure were designed to be used both separately and together.

The Scale of Emotional Closeness (SEC), Part I, is a brief seven-item

paper-and-pencil questionnaire (see Appendix A) that assesses the extent of

subjectively perceived emotional closeness in the respondent and deceased’s

relationship. Participants rated all items (e.g., “I kept my distance emotionally

from this person”) on a 7-point scale from 7 = very strongly agree to 1 = very

strongly disagree. Two items are reverse scored. Higher scores indicate greater

emotional closeness. Sample-specific psychometrics calculated on the data

indicated that Cronbach’s alpha reliability of the scores was .87, and the mean

inter-item correlation was .47. The SEC correlated significantly with the TRIG-

Past, r = .43, p < .01, and TRIG-Present, r = .45, p < .01, subscales, as is consistent

with expectations that the scales measure separate, highly related variables.

The Emotional Closeness Continuum (ECC; Appendix B), Part II, asks

respondents to “make a vertical slash mark on the line below to indicate your level

of closeness with the person who died.” Following an example of a neutral

response, the question: “How aware was this person of your most personal

152 / SERVATY-SEIB AND PISTOLE

feelings?” precedes the 100 millimeter long horizontal line. This line is anchored

by “Completely unaware of my most personal feelings (–)” on the left and

“Completely aware of my most personal feelings (+)” on the right, with a zero in

the center. A metric ruler is used to determine scores, which range from 0 to 100.

Higher scores indicate greater emotional closeness. Pearson correlations calcu-

lated on the sample data revealed that the ECC was significantly related to the

TRIG-Past, r = .54, p < .01, and the TRIG-Present, r = .47, p < .01, as was expected

for different but associated constructs.

With regard to the SEC and ECC being used together with a total score, the

ECC correlated with the SEC at r = .80, p < .01, as is consistent with the two

measuring a similar construct. A factor analysis was conducted on the sample’s

data using the seven items of the SEC and the one score from the ECC. The

principal component extraction procedure using varimax rotation resulted in a

one-factor solution that accounted for 60.7% of the variance. Factor loadings

ranged from .60 to .86.

Procedure

The procedures for data collection in particular rural and urban schools varied

because of school district specifications. In school I, after a brief description of

the study, students who met research criteria raised their hands and took a

packet (i.e., parental consent form, youth assent form, two separate self-addressed

stamped envelopes for consent forms, and questionnaires). The response rate

(i.e., packets taken versus packets returned) from this school was 11%. In

school II, all students received packets prior to the project description and then

left the packet at the door after class if they did not meet the participant criteria or

were not interested in completing the packet. The response rate was 16%. In

school III, students were presented with consent forms during the first visit by

the researcher and asked to return them the following school day if they were able

and willing to complete the packet. Students then completed the questionnaires

during class time the next day. The response rate was 26%. An ANOVA indicated

that there were no significant mean grief intensity differences due to the varied

procedure, F (4, 174) = .63, p > .05.

RESULTS

To examine H1 that adolescents who experienced a friend death loss

would report higher levels of grief intensity than those experiencing a family

kinship death loss (i.e., grandparent or aunt/uncle/cousin), a one-way MANOVA

using relationship category (i.e., friend, grandparent, aunt/uncle/cousin death)

as the independent variable was performed on the TRIG-Past and TRIG-

Present subscales (see Table 1). The multivariate main effect was significant,

F(4, 174) = 2.35, p < .05, �2 = .05, as was the univariate effect for TRIG-Present,

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 153

F(2, 87) = 3.49, p < .05, �2 = .07. Scheffe post hoc analyses indicated that

adolescents reported higher present grief intensity (i.e., TRIG-Present) for friend

death loss than for grandparent death loss. Because it could be that adolescent

experience of friend death loss was more recent, a post hoc analysis using time

since death as a covariate was calculated. The omnibus finding was relatively

unchanged, F(4, 172) = 2.45, p < .05, �2 = .05., and the analysis did not signifi-

cantly alter the TRIG-Present univariate effect, F(2, 86) = 3.37, p < .05, �2 = .07.

In both analyses, the univariate effect for TRIG-Past, F(2, 87) = 1.65, p > .05,

�2 = .04, and F(2, 86) = 1.84, p > .05, �2 = .04, was not significant.

Two procedures were used to test H2 that closeness would account for grief

intensity over and above relationship category. In the first procedure, the total

score from the SEC/ECC was used as a covariate, along with time since death, in

the MANOVA with relationship category as the independent variable and grief

intensity as the dependent variable. The results revealed that the omnibus effect

was no longer present, F(4, 168) = 1.38, p > .05, �2 = .03; there were no significant

mean differences in grief intensity.

Because failure to reject the null hypothesis does not constitute support

for the hypothesis, a second procedure was performed. Emotional closeness

was dichotomized into a categorical variable. Participants’ total scores on the

SEC/ECC were divided based on a mean split into high emotional closeness and

low emotional closeness, with two participants excluded because of failing to

complete one or both of the closeness measures. A 2 × 3 MANOVA, using level

of emotional closeness (i.e., high, low) and relationship category (i.e., friend,

grandparent, aunt/uncle/cousin) as the independent variables and time since death

as the covariate, was performed on the TRIG-Past and TRIG-Present subscales

(see Table 2). The main effect for level of emotional closeness was significant,

F(2, 81) = 10.10, p < .001, �2 = .20. Univariate analyses indicated that those

in the high emotional closeness group scored significantly higher on both

the TRIG-Past (M = 26.23; SD = .92), and TRIG-Present (M = 54.91; SD = 1.95)

than those in the low emotional closeness group (M = 20.20; SD = 1.02 for Past;

154 / SERVATY-SEIB AND PISTOLE

Table 1. Grief Intensity by Relationship Category

Friend

(n = 23)

Grandparent

(n = 45)

Aunt/Uncle/

Cousin

(n = 22)

Relationship M SD M SD M SD F(2, 87) �2

TRIG-Past

TRIG-Present

23.78

54.65

6.69

10.95

21.78

45.76

6.64

13.28

24.41

49.68

6.64

15.14

1.65

3.49*

.04

.07

*p < .05

M = 44.77; SD = 1.95 for Present), F(1, 82) = 12.18, p < .001, �2 = .13; and

F(1, 82) = 19.22, p < .001, �2 = .19, respectively. The main effect for relationship

category was not significant, F(4, 164) = 1.45, p > .05, �2 = .03; and there was

no significant interaction effect, F(4, 164) = .44, p > .05,�2 = .01.

To analyze RQ1, whether emotional closeness adds to the prediction of grief

intensity above and beyond the biographical predictor variables, two hierarchal

multiple regressions (n = 84) were computed with age, sex, time since death,

presence at death, and suddenness of death entered together at Step 1 and the

total SEC/ECC score entered at Step 2 (see Tables 3 and 4). For TRIG-Past,

R was significantly different from zero at the end of each step. For step 1, R2 = .24

(Adjusted R2 = .19), F(5, 78) = 4.88, p < .001. Beta weights indicated that par-

ticipant age, time since death, and presence at death were significant contributors

to TRIG-Past. For step 2, with all variables in the equation, the equation was

significant, R = .66, F (6, 77) = 10.03, p < .001, with �R2 = .20 indicting that

emotional closeness explained significant additional variance in TRIG-Past,

R2 = .44 (adjusted R2 = .40), Finc (1, 77) = 27.50, p < .001. After Step 2, time

since death, presence at the time of death, and emotional closeness contributed

significantly to the prediction of TRIG-Past.

For TRIG-Present, R was significantly different from zero at the end of each

step. After step 1, with participant age, sex, time since death, presence at the time

of death, and suddenness of death in the equation, R2 = .21 (Adjusted R2 = .16),

F(5, 78) = 4.24, p < .01. Participant age and time since death contributed

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 155

Table 2. Grief Intensity by Level of Emotional Closeness and

Relationship Category

TRIG-Past TRIG-Present

Relationship category M SD n M SD n

Friend

High emotional closeness

Low emotional closeness

Grandparent

High emotional closeness

Low emotional closeness

Aunt/Uncle/Cousin

High emotional closeness

Low emotional closeness

25.25

20.43

24.80

20.00

28.63

20.18

6.31

6.78

5.67

6.06

5.50

4.79

16

7

15

28

11

11

57.37

48.43

50.53

43.32

56.81

42.55

10.34

10.36

15.51

11.25

8.13

17.41

16

7

15

28

11

11

Note: High and low emotional closeness was determined through a mean split of

participant totaled score on SEC and ECC.

156 / SERVATY-SEIB AND PISTOLE

Table 3. Intercorrelations among Regression Variables

Variable 1 2 3 4 5 6 7 8

1. Age

2. Sexa

3. Time since death

4. Presence at deathb

5. Suddenness of deathc

6. Emotional closeness

7. TRIG-Past

8. TRIG-Present

— –.17

.15

.01

–.09

.00

–.07

–.12

–.08

–.06

.01

–.16

.17

–.01

.26*

.02

–.16

.10

.24*

.21*

.07

.54**

–.27*

.26*

.15

.14

.09

.49**

.67**

a0 = male and 1 = femaleb0 = not present at death and 1 = present at deathc0 = not sudden and 2 = sudden

*p < .01. **p < .01.

Table 4. Summary of Hierarchical Regression Analysis for Variables

Predicting TRIG-Past and TRIG-Present (N =84)

TRIG-Past TRIG-Present

Variable B SE B � sr2 B SE B � sr2

Step 1

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Step 2

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Emotional closeness

–1.38

–.44

.01

7.14

.57

–1.05

–1.48

.01

4.54

.49

.09

.62

1.57

.00

2.19

1.31

.54

1.37

.00

1.96

1.13

.02

–.23

–.03

.32

.32

.04

–.18

–.10

.32

.21

.04

.48

.05*

.00

.10**

.10**

.00

.03

.00

.10***

.04*

.00

.20***

–3.81

5.50

.01

7.98

2.09

–3.17

3.52

.01

3.00

1.94

.18

1.32

3.35

.01

4.68

2.80

1.19

3.03

.01

4.33

2.50

.04

–.30

.17

.23

.17

.08

–.25

.11

.23

.07

.07

.43

.08**

.03

.05*

.03

.01

.06**

.01

.05*

.00

.00

.17***

Note: TRIG-Past R2 = .24 for Step 1 (p < .001); �R

2 = .20 for Step 2 (p < .001).

TRIG-Present R2 = .21 for Step 1 (p < .01); �R

2 = .17 for Step 2 (p < .001).

*p < .05. **p < .01. ***p < .001.

significantly to the prediction of TRIG-Present. For step 2, with all variables

included, the equation was significant, R = .62, F (6, 77) = 7.83, p < .001, with

�R2 = .17 indicating that emotional closeness explained significant additional

variance in TRIG-Present, R2 = .38 (adjusted R2 = .33), Finc(1, 77) = 20.47,

p < .001. Participant age, time since death, and emotional closeness were signifi-

cant contributors to the prediction of TRIG-Present.

Post Hoc Analyses

Because the regression equations were significant, we computed hierarchal

multiple regressions, predicting TRIG-Past and TRIG-Present, for each relation-

ship category (i.e., friend, grandparent, aunt/uncle/cousin), in order to provide the

most information for future research. Again, age, sex, time since death, presence at

death, and suddenness of death were entered together at Step 1, and emotional

closeness was entered at step 2 (see Table 5). For friend (n = 23), the equations for

TRIG-Past were not significant at step 1 F(5, 17) = .98, p > .05 or step 2, F(6, 16) =

2.36, p > .05. However, as indicated in Table 5, ÄR2 = .25 emotional closeness

added to the equation was significant, R2 = .47 (adjusted R2 = .27), Finc(1, 16) =

7.40, p < .05. The TRIG-Present equations were also not significant at step 1,

(n = 23), F(5, 17) = .88, p > .05, or step 2, F(6, 16) = 1.08, p > .05, nor was

the incremental change, ÄR2 for emotional closeness. Nonetheless, emotional

closeness explained additional variance, 25% for TRIG-Past and 8% for the

TRIG-Present.

In contrast, the equations for grandparent (n = 42) on TRIG-Past were sig-

nificant at step 1, F(5, 36) = 2.67, p < .05, and step 2, F(6, 35) = 4.77, p < .001.

Time since death was a significant predictor at step 1; for step 2, time since

death and emotional closeness were significant predictors. For TRIG-Present,

the equation for step 1 was not significant F(5, 36) = 2.31, p > .05, but the equation

for step 2 was significant, F(6, 35) = 4.15, p < .01. Participant age and emotional

closeness significantly predicted TRIG-Present. Emotional closeness explained

additional variance, 18% for TRIG-past and 17% for TRIG-present.

For aunt/uncle/cousin (n = 19), the TRIG-Past equations were significant at

step 1, F(5, 13) = 4.53, p < .05 and step 2, F(6, 12) = 4.39, p < .05, with presence

at death being the only significant predictor for each. The equations for

TRIG-Present were also significant at step 1, F(5, 13) = 46.23, p < .01 and step 2

F(5, 13) = 4.53, p < .05, with age, time since death, and presence at death as

significant predictors.

DISCUSSION

The results of this study generally supported the hypotheses. For H1, adoles-

cents reported significantly higher present grief intensity for friend than for

grandparent death loss. For H2, based on the two procedures, adolescents in the

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 157

158 / SERVATY-SEIB AND PISTOLE

Table 5. Summary of Hierarchical Regression Analysis for Variables

Predicting TRIG-Past and TRIG-Present by Relationship Category

TRIG-Past TRIG-Present

Variable B SE B � sr2 B SE B � sr2

Friend

Step 1

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Step 2

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Emotional closeness

Grandparent

Step 1

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Step 2

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Emotional closeness

Aunt/Uncle/Cousin

Step 1

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Step 2

Participant age

Participant sex

Time since death

Presence at death

Suddenness of death

Emotional closeness

–1.32

–2.12

.01

3.10

5.23

–.01

–2.64

.01

1.12

5.48

.11

–1.06

1.03

.01

4.05

–1.42

–1.05

–.82

.01

.68

–1.29

.08

–4.01

–2.86

.01

11.42

–1.11

–3.10

–1.89

.01

9.31

.00

.01

1.80

4.71

.01

5.12

3.61

1.60

4.02

.01

4.42

3.08

.04

.80

2.17

.00

4.26

2.01

.71

2.00

.00

3.89

1.77

.03

1.91

3.12

.01

2.82

2.53

1.95

3.09

.01

3.10

2.56

.04

–.18

–.11

.24

.13

.33

–.01

–.14

.33

.05

.35

.55

–.21

.08

.41

.14

–.11

–.21

–.06

.40

.02

–.10

.46

–.55

–.18

.54

.71

–.08

–.43

–.12

.42

.58

.00

.29

.02

.01

.05

.02

.10

.00

.01

.09

.00

.11

.25*

.04

.00

.16**

.02

.01

.03

.00

.15**

.00

.01

.18**

.13

.02

.12

.47***

.01

.07

.01

.07

.23**

.00

.05

–3.92

4.49

.00

6.71

–.53

–2.75

4.00

.01

4.84

–.29

.10

–3.23

6.60

.01

–.78

–1.89

–3.21

2.80

.01

–7.71

–1.63

.18

–18.07

4.01

.01

18.63

4.36

–16.89

5.27

.01

15.87

5.81

.08

2.97

7.80

.01

8.47

5.97

3.03

7.61

.01

8.38

5.83

.08

1.71

4.63

.01

9.09

4.28

1.53

4.29

.01

8.38

3.81

.06

4.10

6.71

.02

6.06

5.45

4.40

6.98

.02

7.00

5.79

.10

–.33

.14

.07

.18

–.02

–.23

.13

.12

.13

–.01

.32

–.30

.23

.22

–.01

–.07

–.30

.10

.21

–.13

–.06

.45

–1.03

.11

1.02

.49

.13

–.97

.14

.96

.42

.17

.16

.08

.02

.00

.03

.00

.04

.01

.01

.01

.00

.08

.07

.04

.05

.00

.00

.08*

.01

.04

.01

.00

.17**

.43***

.01

.43***

.21**

.01

.34**

.01

.32**

.12*

.02

.02

Note: TRIG-Past R2 = .24 for Step 1 (p < .001); �R

2 = .20 for Step 2 (p < .001).

TRIG-Present R2 = .21 for Step 1 (p < .01); �R

2 = .17 for Step 2 (p < .001).

*p < .05. **p < .01. ***p < .001.

high closeness group reported significantly higher past and present grief intensity,

regardless of relationship category. Finally, for the RQ, emotional closeness added

significantly to the prediction of past and present grief intensity.

The finding of higher grief intensity for friend versus grandparent death

seems inconsistent with Littlefield and Rushton’s (1986) proposition and

the concept of evolutionary salience being central in predicting grief intensity.

However, grandparents’ age must also be considered when interpreting these

results. Because grandparents are likely no longer of reproductive age, they may

not be evolutionary salient to the adolescent at least in terms of self-report and

awareness, despite the index of relatedness. Besides, it is the adolescent and

not the grandparent who is the one who is reproductively valued. On the other

hand, adolescent friendships might be characterized by evolutionary mutual

benefit (see Archer, 1999), for instance, in terms of developmental tasks. If

so, then an evolutionary-based grief approach would be in favor of adolescent

friendships, given that grandparents are no longer reproductively valued and

that the social environment is characterized by relatives living at a distance

rather than functioning together in small groups (Simpson, 1999). However, this

alternate interpretation is highly speculative, and we suspect that closeness is

a more parsimonious variable, though future research could investigate this

speculation. In addition, because of the age difference between deceased friends

versus grandparents, it could be that current grief is more intense for friend

death loss because this loss is more often perceived as preventable, considered

a disenfranchised experience (Oltjenbruns, 1996; Ringler & Hayden, 2000;

Sklar & Hartley, 1990), has an off-time quality, and is likely somehow

violent (Ringler & Hayden, 2000). Or it could be that present grief is higher for

friend death loss because there are reminders of the person in daily life (e.g.,

at school), or interactions and functions (e.g., support) associated with the person

are still missing.

The finding of significant mean differences in past and present grief for higher

versus lower closeness, regardless of relationship category, is consistent with

Archer’s (1999) premise that closeness, not kinship, is the salient indicator of

grief intensity. This finding is also consistent with Bugen’s (1977) inclusion of

centrality in predicting grief intensity. Nonetheless, future research needs to

determine if this finding remains for other relationship categories (e.g., primary

attachments such as parents or committed romantic partners). In addition, com-

parisons of bereaved adolescents should examine (a) groupings of friend death

loss (e.g., romantic, best friend, friend), (b) groupings of kin death loss (i.e.,

parents/siblings, who share 50% relatedness; grandparents/aunts/uncles, who

share 25% relatedness; and first cousins, who share 12.5% relatedness (Archer,

1999; Simpson, 1999), and (c) groupings of age-related death loss (e.g., similarly

aged, older, and younger individuals). Such research, using path analysis or

structural equation modeling, would allow for a more refined evaluation of

evolutionary based predictions. It could be, for instance, that, in current U.S.

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 159

society, closeness is a proximal factor mediating the more distal evolutionary

salience of kinship.

For the research question, emotional closeness explained an additional 20%

and 17% of the variance in past and present grief intensity, respectively. This

finding is consistent with finding higher past and present grief differences in the

high versus low closeness group, and again indicates that emotional closeness

needs to be considered when examining adolescent death losses. Nonetheless, we

are leery of interpreting the post hoc, relationship category regression analyses

because of the low sample sizes. The primary purpose of these analyses was to

provide researchers with information (e.g., the semipartial correlations) to suggest

future research paths. Nonetheless, we note that emotional closeness was a

positive predictor of TRIG-Past for the friend and grandparent death loss groups.

Although emotional closeness was the primary focus of the present inves-

tigation, it is interesting that time since the death predicted TRIG-Past in several

significant regression analyses. The positive relationship between time since

death and past grief intensity is perhaps counterintuitive, but it is aligned with

previous suggestions that time since death is not a straightforward predictor of

grief intensity for adolescents (Fleming & Balmer, 1996). The current analyses

suggest that the further adolescents are from the death loss, the greater they

perceive their past grief intensity to have been. It could be that: a) adolescents

idealize in their retrospective accounts; b) as they move beyond the initial grief,

they realize how upset they were initially, and appraise and report the past

grief more accurately; and c) the memory of the past grief is stronger in com-

parison to present grief. Future longitudinal research would be useful to examine

the meaning of this finding, and whether it was unique to the present sample or

is a more generalizable result.

Limitations and Future Research

It is important to note limitations of the research. To begin with, the current

sample was relatively homogenous with regard to race, sex, religious affiliation,

and geographic location, which made it difficult to examine the extent to which

these distinctions might affect the generalizability of findings. Likewise, although

the sample is consistent with previous grief-related research, the predominance

of females decreased the power and so did not truly allow for an examination

of sex differences, which may be relevant to evaluating evolutionary based

and closeness hypotheses (Archer, 1999; Marwit & Klass, 1994-1995;

Oltjenbruns, 1996; Silverman, 2000). In addition, a larger N would allow

researchers to group discrete types of kin (e.g., first cousins), whereas this

research, due to the sample size, grouped kin according to the unit that typically

lives under the same roof (e.g., aunt/uncle/cousin) in the cultural group that

was dominant in the sample. Cultural and individual differences are important

for scholars to consider in future research. For example, some adolescents (e.g.,

160 / SERVATY-SEIB AND PISTOLE

poor or marginalized groups) assume adult responsibilities at a much earlier age

than might be likely for this predominately White sample (Preto, 1999). It is not

clear how school and work responsibilities, associated differences in available

caregiving resources in the family, or cultural collectivism would influence the

relatedness of grief intensity, emotional closeness, and relationship category.

A second important limitation concerns the measurement of closeness. The

SEC/ECC was designed for this study. These descriptive items did have strong

internal consistency, but it would be useful for the SEC/ECC to be further

investigated in a validity study, one which might compare the SEC/ECC to the

other approaches to measuring emotional closeness. There are different theoretical

frameworks for understanding closeness, and these varied frameworks may lead

to unrelated research questions and disparate findings (Berscheid & Reis, 1998).

In this study, closeness referred to an individual’s subjective impression of the

level of emotional closeness that existed in his or her relationship with the

deceased. A validity study could help determine convergence of as well as

distinctions in meanings and measures. In the meantime, based on the strong

sample specific psychometrics reported here, future research could explore the

use of the SEC/ECC with additional populations of adolescents and perhaps

adults. Finally, it could be that adolescents’ perceptions of their emotional close-

ness may change as a function of the death itself. Perhaps adolescents perceive

themselves as emotionally closer to the deceased the day after the death than

they would have the day before the death. A prospective study would provide

useful information for tackling this question.

In future research, it might be beneficial for researchers to consider attachment,

when examining closeness in adolescent friend and kin relationships. Because

theory and previous research suggest that attachment quality is relevant to grief

(Harris, 1991; Shaver & Tancredy, 2001; Stroebe & Stroebe, 1993; Wayment

& Vierthaler, 2002), investigating closeness along with attachment and affili-

ative bonds (Weiss, 2001) can contribute to the knowledge base. In doing so,

researchers need to assess: a) whether the adolescent has an attachment or an

affiliative bond to the deceased (Hazan, Gur-Yaish, & Campa, 2004), possibly by

using the WHOTO (Hazan & Zeifman, 1994); b) the quality of the attachment

relationship (Ainsworth, 1988); and c) whether the deceased is a highly preferred

or a subsidiary attachment (i.e., what is the hierarchal placement of the deceased

among the adolescent’s attachments) (Bowlby, 1969; Trinke & Bartholomew,

1997). Because most adolescents are in romantic relationships (Carver, Joyner,

& Udry, 2003), which may be attachment relationships with emotional support

and sexual development functions (Allen & Land, 1999; Furman & Shaffer,

2003; Tracy, Shaver, Albino, & Cooper, 2003), research on closeness and

grief intensity should include several kinds peer relationships (e.g., sexual and

non-sexual romantic relationships, “best” friends, friends).

Finally, because of criticisms of the TRIG (Neimeyer & Hogan, 2001), follow

up research might use other, more sophisticated measures of bereavement (e.g.,

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 161

Hogan Inventory of Bereavement, Core Bereavement Items). A longitudinal

design would also be worthwhile.

Practical Application

According to these results, emotional closeness to the deceased needs to be

actually used in an applied and practical sense when working with individual

adolescents on a day-to-day basis. After asking about the formal relationship with

the deceased (e.g., friend, grandparent, aunt/uncle/cousin), teachers, counselors,

after-care providers, and others who work with adolescents need to go a step

further by intentionally and specifically asking bereaved adolescents, “How

close were you to______?” It is simply not enough to know if a grandparent or a

friend has died. Although many grandparents maintain households for their

grandchildren (Casper & Bryson, 1998), especially in some ethnic minority

groups, other grandparent-grandchild relationships may not be close due to

distance and limited contact. The adolescent might be much closer to a

friend. Or the distance and limited contact may be overcome by phone con-

versations or webcam visits with a grandparent. The present findings suggest

that those working with bereaved adolescents can avoid misguided interven-

tions and have an accurate sense of the experienced grief intensity by asking

specifically about closeness.

Practitioners who work in or interface with schools are also likely to find

value in the present findings. These individuals, as an advocate for bereaved

adolescents, can emphasize and educate others about the relatedness of close-

ness and grief intensity. Another basic application of these emotional closeness

findings relates to policies regarding the allowable number of school absences

following death losses. Teachers and counselors may advocate a flexible policy

that considers the adolescent’s perceived emotional closeness to the deceased,

rather than an allotted number of absences based on and associated with the

formal relationship category (e.g., two weeks for parent death loss, two days for

grandparent death loss, no days for friend death loss).

This research contributes to the adolescent death loss and grief intensity

literature because it examined adolescents’ common death losses, investi-

gated grandparent loss and closeness, and compared kin and friend death

loss. The results suggest that emotional closeness is a useful predictor of grief

intensity, at least for adolescent common death losses. Although this finding

may seem obvious, it is imperative that researchers pursue such investiga-

tions so that bereavement services are based on empirical evidence. In addition,

different levels of analysis (e.g., general evolution theory, mid-level theories)

lead to multiple predictions (Simpson, 1999); so additional research is war-

ranted and needed. Nonetheless, the current study sets the stage for more

complex research examining relationship category and closeness in adolescent

death losses.

162 / SERVATY-SEIB AND PISTOLE

APPENDIX A

Scale of Emotional Closeness (SEC)

The level of closeness we feel to others differs from person to person and over

time. Please think about your relationship with the important person who died

while answering the following questions. Using the following scale, circle the

number that corresponds to how much you agree with each statement.

7 6 5 4 3 2 1

very neither very

strongly strongly

agree disagree

1. I felt I could share my most intimate feelings

with this person. 7 6 5 4 3 2 1

2. I kept my distance emotionally from this person. 7 6 5 4 3 2 1

3. It was very easy to talk with this person. 7 6 5 4 3 2 1

4. I felt close to this person. 7 6 5 4 3 2 1

5. It was difficult to talk with this person. 7 6 5 4 3 2 1

6. This person understood me. 7 6 5 4 3 2 1

7. This person shared his/her most personal thoughts

with me. 7 6 5 4 3 2 1

APPENDIX B

Emotional Closeness Continuum

Please make a vertical slash mark on the line below to indicate your level of

closeness with the person who died.

How aware was this person of your most personal feelings?

Completely Completely

unaware of my (–)———————————————(+) aware of my

most personal 0 most personal

feelings feelings

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 163

For example, placing a mark at this point on

the line would indicate that you are unsure.

ACKNOWLEDGMENTS

The authors wish to thank colleagues P. Nora Reilly, Deborah Bennett, Marilyn

Haring, and Deborah Taub for feedback on earlier versions of the manuscript.

REFERENCES

Ainsworth, M. (1988). On security. Retrieved July 20, 2006, from

http://www.psychology.sunysb.edu/attachment/ainsworth/ainsworth_index.html.

Allen, J. P., & Land, D. (1999). Attachment in adolescence. In J. Cassidy & P. R. Shaver

(Eds.), Handbook of attachment: Theory, research, and clinical implication

(pp. 319-335). New York: Guilford.

Archer, J. (1999). The nature of grief: The evolution and psychology of reactions to loss.

Philadelphia, PA: Brunner-Routledge.

Balk, D. (1990). The self-concept of bereaved adolescents: Sibling death and its aftermath.

Journal of Adolescent Research, 5, 112-132.

Balk, D., & Corr, C. A. (1996). Adolescents, developmental tasks, and encounters with

death and bereavement. In C. A. Corr & D. E. Balk (Eds.), Handbook of adolescent

death and bereavement (pp. 3-24). New York: Springer.

Belsky, J. (1999). Modern evolutionary theory and patterns of attachment. In J. Cassidy

& P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical

implication (pp. 141-161). New York: Guilford.

Berscheid, E., & Reis, H. T. (1998). Attraction and close relationships. In D. T. Gilbert,

S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed.,

pp. 193-218). Boston, MA: McGraw-Hill.

Berscheid, E., Snyder, M., & Omoto, A. M. (1989). The relationship closeness inventory:

Assessing the closeness of interpersonal relationships. Journal of Personality and

Social Psychology, 57, 792-807.

Bowlby, J. (1969). Attachment and loss: Vol. 1: Attachment. New York: Basic Books.

Brent, D. A., Perper, J., Moritz, G., Allman, C., Friend, A., Schwerrs, J., Roth, C., Balach,

L., & Harrington, K. (1992). Psychiatric effects of exposure to suicide among the

friend and acquaintances of adolescent suicide victims. Journal of the Academy of

Child and Adolescent Psychiatry, 31, 629-639.

Bugen, L. A. (1977). Human grief: A model for prediction and intervention. American

Journal of Orthopsychiatry, 47, 196-206.

Burnett, P., Middleton, W., Raphael, B., & Martinek, N. (1997). Measuring core bereave-

ment phenomena. Psychological Medicine, 27, 49-57.

Carver, K., Joyner, K., & Udry, J. R. (2003). National estimates of adolescent romantic

relationships. In P. Florsheim (Ed.), Adolescent romantic relations and sexual

behavior: Theory, research, and practical implications (pp. 23-56). Mahwah, NJ:

Lawrence Erlbaum Associates.

Casper, L. M., & Bryson, K. R. (1998). Co-resident grandparents and their grandchildren:

Grandparent maintained families (Population Division Working Paper No. 26).

Washington, DC: U.S. Bureau of the Census.

Cleiren, M. P. H. D. (1993). Bereavement and adaptation: A comparative study of the

aftermath of death. Washington, DC: Hemisphere.

164 / SERVATY-SEIB AND PISTOLE

Cleiren, M. P. H. D., Diekstra, R. F. W., Kerkhof, A. J. F. M., & van der Wal, J. (1994).

Mode of death and kinship in bereavement: Focusing on “who” rather than “how.”

Crisis: Journal of Crisis Intervention & Suicide, 15, 22-36.

Collins, W. A. (1997). Relationships and development during adolescence: Interpersonal

adaptation to individual change. Personal Relationships, 4, 1-14.

Crenshaw, D. A. (1990). Bereavement: Counseling the grieving through the life cycle.

New York: Continuum.

Davies, B. (1991). Long-term outcomes of adolescent sibling bereavement. Journal of

Adolescent Research, 6, 83-96.

Dillon, D. H., & Brassard, M. R. (1999). Adolescents and parental AIDS death: The role

of social support. Omega, 39, 179-195.

Dyregrov, A., Gjestad, R., Bie Wikander, A. M., & Vigerust, S. (1999). Reactions

following the sudden death of a classmate. Scandinavian Journal of Psychology, 40,

167-176.

Erikson, E. H. (1968). Identity: Youth and crisis. New York: W. W. Norton.

Fanos, J. H., & Nickerson, B. G. (1991). Long-term effects of sibling death during

adolescence. Journal of Adolescent Research, 6, 70-82.

Fashingbauer, T. R., Zisook, S., & DeVaul, R. (1987). The Texas Revised Inventory of

Grief. In S. Zisook (Ed.), Biopsychosocial aspects of bereavement (pp. 109-124).

Washington, DC: American Psychiatric Press.

Fleming, S., & Balmer, L. (1996). Bereaved in adolescence. In C. A. Corr & D. Balk

(Eds.), Handbook of adolescent death and bereavement (pp. 139-154). New York:

Springer.

Furman, W., & Shaffer, L. (2003). The role of romantic relationships in adolescent

development. In P. Florsheim (Ed.), Adolescent romantic relations and sexual

behavior: Theory, research, and practical implications (pp. 3-22). Mahwah, NJ:

Lawrence Erlbaum Associates.

Glass, J. C. (1990). Death, loss, and grief in high school students. High School Journal,

73, 154-160.

Grabowski, J., & Frantz, T. T. (1992-93). Latinos and Anglos: Cultural experiences of

grief intensity. Omega, 26, 273-285.

Harris, E. S. (1991). Adolescent bereavement following the death of a parent: An explor-

atory study. Child Psychiatry and Human Development, 21, 267-281.

Hazan, C., Gur-Yaish, N., & Campa, M. (2004). What does it mean to be attached? In

W. S. Rholes & J. A. Simpson (Eds.), Adult attachment: Theory, research, and clinical

implications (pp. 55-85). New York: Guilford.

Hazan, C., & Zeifman, D. (1994). Sex and the psychological tether. In K. Bartholomew &

D. Perlman (Eds.), Advances in personal relationships: Vol. 5. Attachment processes

in adulthood (pp. 151-177). London: Kingsley.

Hazan, C., & Zeifman, D. (1999). Pair bonds as attachments. In J. Cassidy & P. R. Shaver

(Eds.), Handbook of attachment: Theory, research, and clinical implication

(pp. 336-354). New York: Guilford.

Irizarry, C. (1992). Spirituality and the child: A grandparent death. In G. R. Cox &

R. J. Fundis (Eds.), Spiritual, ethical, and pastoral aspects of death and bereavement

(pp. 131-146). Amityville, NY: Baywood.

Littlefield, C. H., & Rushton, J. P. (1986). When a child dies: The sociobiology of

bereavement. Journal of Personality and Social Psychology, 51, 797-802.

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 165

Lurie, C. (1993). The death of friends vs. family member in late adolescence: The role of

perceived social support and self-worth. Unpublished master’s thesis, Colorado State

University, Fort Collins.

Martinson, I. M., & Campos, R. G. (1991). Adolescent bereavement: Long-term responses

to a sibling’s death from cancer. Journal of Adolescent Research, 6, 54-69.

Marwit, S. J., & Klass, D. (1994-95). Grief and the role of the inner representation of

the deceased. Omega, 30, 283-298.

McIntosh, J. L. (1993). Control group studies of suicide survivors: A review and critique.

Suicide & Life-Threatening Behavior, 23, 146-161.

McNeil, J. N., Silliman, B., & Swihart, J. J. (1991). Helping adolescents cope with the

death of a peer: A high school case study. Journal of Adolescent Research, 6, 132-145.

Neimeyer, R. A., & Hogan, N. S. (2001). Quantitative or qualitative? Measurement issues

in the study of grief. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.),

Handbook for bereavement research: Consequences, coping, and care. (pp. 89-118).

Washington, DC: American Psychological Association.

Oltjenbruns, K A. (1996). Death of a friend during adolescence: Issues and impacts. In

C. A. Corr & D. E. Balk (Eds.), Handbook of adolescent death and bereavement

(pp. 196-215). New York: Springer.

O’Brien, J. M., Goodenow, C., & Epsin, O. (1991). Adolescent reactions to the death of

a peer. Adolescence, 26, 431-440.

Park, C. L, & Cohen, L. H. (1993). Religious and nonreligious coping with the death of

a friend. Cognitive Therapy and Research, 17, 561-577.

Podell, C. (1989). Adolescent mourning: The sudden death of a peer. Clinical Social

Work Journal, 17, 64-78.

Preto, N. G. (1999). Transformation of the family system during adolescence. In B. Carter

& M. McGoldrick (Eds.), The expanded family life cycle: Individual, family, and

social perspectives (pp. 274-286). Boston: Allyn & Bacon.

Raphael, B. (1983). The anatomy of bereavement. New York: Basic Books.

Reed, M. D., & Greenwald, J. Y. (1991). Survivor-victim status, attachment, and sudden

death bereavement. Suicide and Life-Threatening Behavior, 21, 385-401.

Ringdal, G. I., Jordhoy, M. S., Ringdal, K., & Kaasa, S. (2001). Factors affecting grief

reactions in close family members to individuals who have died of cancer. Journal

of Pain & Symptom Management, 22, 1016-1026.

Ringler, L. L., & Hayden, D. C. (2000). Adolescent bereavement and social support:

Peer loss compared to other losses. Journal of Adolescent Research, 15, 209-230.

Russac, R. J., Steighner, N. S., & Canto, A. I. (2002). Grief work versus continuing bonds:

A call for paradigm integration or replacement? Death Studies, 26, 463-478.

Sanders, C. (1979-1980). A comparison of adults bereavement in the death of a spouse,

child, and parent. Omega, 10, 303-322.

Schachter, S. (1991-1992). Adolescent experiences with the death of a peer. Omega,

24, 1-11.

Shaver, P. R., & Tancredy, C. M. (2001). Emotion, attachment, and bereavement: A

conceptual commentary. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut

(Eds.), Handbook for bereavement research: Consequences, coping, and care

(pp. 63-88). Washington, DC: American Psychological Association.

Silverman, P. R. (2000). Never too young to know: Death in children’s lives. New York:

Oxford.

166 / SERVATY-SEIB AND PISTOLE

Simpson, J. A. (1999). Attachment theory in modern evolutionary perspective. In

J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and

clinical implication (pp. 115-140). New York: Guilford.

Sklar, F., & Hartley, S. F. (1990). Close friends as survivors: Bereavement patterns in

a “hidden” population. Omega, 21, 103-112.

Smith, R. J., Lingle, J. H., & Brock, T. C. (1978-79). Reactions to death as a function

of perceived similarity to the deceased. Omega, 9, 125-138.

Stroebe, M. S., & Stroebe, W. (1993). The mortality of bereavement: A review. In M. S.

Stroebe, W. Stroebe, & R. O. Hansson (Eds.), Handbook of bereavement: Theory,

research and intervention (pp. 175-195). New York: Cambridge University Press.

Swanson, E. A., & Bennett, T. F. (1982-1983). Degree of closeness: Does it affect the

bereaved’s attitudes toward selected funeral practices? Omega, 13, 43-50.

Toray, T., & Oltjenbruns, K. A. (1996). Children’s friendships and the death of a friend.

In C. A. Corr & D. M. Corr (Eds.), Handbook of childhood death and bereavement

(pp. 165-178). New York: Springer.

Tracy, J. L., Shaver, P. R., Albino, A. W., & Cooper, M. L. (2003). Attachment styles

and adolescent sexuality. In P. Florsheim (Ed.), Adolescent romantic relations and

sexual behavior: Theory, research, and practical implications (pp. 137-159).

Mahwah, NJ: Lawrence Erlbaum Associates.

Trinke, S. J., & Bartholomew, K. (1997). Hierarchies of attachment relationships in young

adulthood. Journal of Social and Personal Relationships, 14, 603-625.

Wass, H. (1995). Death in the lives of children and adolescents. In H. Wass, & R. A.

Neimeyer (Eds.), Dying: Facing the facts (pp. 269-301). Washington, DC: Taylor

& Francis.

Wayment, H. A., & Veirthaler, J. (2002). Attachment style and bereavement reactions.

Journal of Loss & Trauma, 7, 129-149.

Webb, N. B. (1993). Helping bereaved children: A handbook for practitioners. New

York: Guilford.

Weiss, R. S. (2001). Grief, bonds, and relationships. In M. S. Stroebe, R. O. Hansson,

W. Stroebe, & H. Schut (Eds.), Handbook of bereavement research: Consequences,

coping and care (pp. 47-62). Washington, DC: American Psychological Association.

Worden, J. W., & Silverman, P. R. (1996). Parental death and the adjustment of school-age

children. Omega, 33, 91-102.

Zisook, S., & Lyons, L. (1988). Grief and the relationship to the deceased. International

Journal of Family Psychiatry, 9, 135-146.

Direct reprint requests to:

Heather L. Servaty-Seib

Counseling & Development

Department of Educational Studies

Purdue University

West Lafayette, IN 47907

e-mail: [email protected] or [email protected]

ADOLESCENT GRIEF: RELATIONSHIP AND CLOSENESS / 167