Standardization of Reasons for Living Inventory for adolescents (RFL-A) among Iranian adolescents (Kermanshah city)

Introduction:

 The vast majority of research on suicide risk focuses on negative factors that increase the chances of an individual engaging in intentional self-harm (e.g., Beck, Kovacs, & Weismann, 1979; Joiner & Rudd, 1996).

     The different approach to assessment suicide risk arose with the product and development of the Reasons for Living Inventory (RFL; Linehan, Goodstein, Nielsen,& Chiles, 1983). Linehan and colleagues chose to examine the cognitive factors that allow individuals desire to living in the face of hardship and adversity. They say that suicidal individuals lack coping characteristics possessed by normal individuals and have important role in understanding suicide risk. Their new scale allowed to therapist differentiated suicidal and normal individuals to be based on the content of their belief systems. This scale has a 48-item and six valid and reliable subscales: Survival and Coping Beliefs (SCB), Responsibility to Family (RF), Child Related Concerns (CRC), Fear of Suicide (FS), Fear of Social Disapproval (FSD), and Moral Objections (MO).

    The RFL and its psychometric properties have been examined and supported in several studies (Kralik & Danforth, 1992; Osman et al., 1993; Osman, Gregg, Osman, & Jones, 1992; Osman, Jones, & Osman, 1991). Findings on sex differences with the RFL have varied. One study found no differences in scores across subscales (Osman et al., 1991), whereas women scored higher than men on some subscales, such as FS, RF, and MO, in others (Hirsch & Ellis, 1996; Osman et al., 1993; Osman et al., 1992; Osman et al., 1991).

    The RFL has been used in a variety of studies with college students in the countries like United States and Australia as a means of examining protective factors (Hirsch & Ellis, 1996).  Results of Dyck et al (1991) showed weak but significant negative correlations between total RFL score and hopelessness, and they belief that the RFL have a distinct construct have supported. In addition, Hirsch and Ellis (1996) found that suicide Ideators could be distinguished from normal based on their scores on the RFL. Connell and Meyer (1991) grouped college students into categories based on reported history of suicidality and found that the SCB, RF, and MO subscales adequately discriminated between groups. The clinical utility of the RFL has been demonstrated with both adult outpatient (Dyck, 1991) and psychiatric inpatient samples (Strosahl, Chiles, &Linehan, 1992). Dyck concluded that the RFL is less influenced by depression than a commonly used measure of hopelessness and may therefore be a better measure of suicide risk with depressed patients.

    Strosahl and colleagues found that the SCB subscale of the RFL was the best at discriminating across of desire to suicide in a group of patients with a history of suicide. Range, Hall, and Meyers (1993) examined the factor structure, reliability, and validity of the RFL when used with adolescents. Their sample included 128 high school students between the ages of 14 and 17, plus a comparison sample of 153 college students under the age of 20. Their confirmatory factor analysis (CFA) failed to fit the data from either sample to the original RFL six-factor structure or to a five-factor solution (deleting CRC items).

However, Range et al. (1993) were able to derive two unique six-factor solutions accounting for 53.6% of the variance in high school student data and 49.8% of variance in college student data. The authors determined internal consistency reliability of all original RFL subscales except MO to be adequate in both samples (range of Cronbach = .77 to .91).

    Westfield, Cardin, and Deaton (1992) based on original RFL scale produced an RFL-type measure specifically for the college student population. Similar original RFL scale they derived a six-factor solution. But they put college-related concerns factor in new scale and remove child-related concerns factor for a specific, increased importance placed on friends in addition to family. College Student Reasons for Living inventory included: SCB, College and Future-Related Concerns, MO, Responsibility to Friends and Family, FS, and FSD (Westfield et al., 1992). The psychometric properties of the College Student Reasons for Living inventory examined and accepted in the several studies (Rogers & Hanlon, 1996; Westfield, Bandura, Kiel, & Scheel, 1996).

 

    Utility of the RFL with the adolescent population in the several studies examined. Cole (1989) based on five subscale (CRC was dropped) of six subscale RFL compared high school students and adolescences delinquents. They results were consistent with Linehan et al. (1983) but MO failed to significantly correlate with depression, hopelessness, or suicidality in the delinquent adolescences (Cole, 1989).  In the other hand, the high school sample Ideators were distinguished from attempters based on their MO scores.

   Results study of Pinto, Weismann, and Conwell (1998) Instead, exploratory components analysis yielded a five-factor solution accounting for 66.5% of the variance failed to replicate the original RFL factor structure with adolescent psychiatric inpatients.

    Based on the available data, it appears that the theoretical base of RFL is adequate to adolescents. However, the results of past studies when the RFL is used with adolescents and college students suggested the need for a unique measure for adolescents (Osman et al., 1996).Therefore decided to develop a new measure, based on the underlying theory of the RFL, specifically for adolescents.

    Improved ways of assessing the level suicide risk in the Iranian adolescents is necessary. Based on annual data for 2006 collected from Social Welfare organization, from 5 attempt for suicide three of them are adolescents between age 12 to 24 (Social Welfare organization 2006).  In the all cities and state of Iran, Kermanshah have upper rate of suicide. The rate of completed suicides in the 15- to 24-year-old age group has deviated from a mean of 6.1 (per100, 000 population) for the 1387(2008) year in the Kermanshah city (Emam Khomeini hospital of treatment suicide). These data suggest that intentional self-harmful behavior and the potential for engaging in such behaviors are a serious concern for young people, parents, teachers and counselors and overall society in the Kermanshah city.

   The RFL–A is a 32-item self-report measure designed specifically to assess adolescents’ adaptive reasons for not committing suicide. It is comprised of five factors: Future Optimism (FO), Suicide-Related Concerns (SRC), Family Alliance (FA), Peer Acceptance and Support (PAS), and Self-Acceptance (SA). Less relevant items (e.g., relating to concerns about the effects of suicide on one’s children) are not included in the RFL–A. The factor structure of the RFL–A is consistent with the multifaceted nature of adolescent suicidality (Osman et al., 1998). The authors also found support for convergent and construct validity. Important group differences on the RFL–A were identified. Specifically, boys had significantly higher SA scores, adolescents   in the normal group scored higher on all subscales than an suicidal group, and a psychiatric no suicidal group scored higher than a psychiatric attempter group. The main purpose of this study was to confirm the factor structure of the RFL–A derived by Osman et al. (1998) in the Iranian adolescents (Kermanshah city).and we tested the hypothesis that: 1) the RFL–A can distinguish adolescent on suicide group from normal. 2) Finally, we hypothesized that the RFL–A would discriminate between suicide attempters and no attempters better than the Beck Hopelessness Scale (BHS; Beck, Weismann, Lester, & Trexler, 1974).

 

Method

 

Participants

 

Participants (189 boys and 211 girls) were recruited from all Kermanshah high schools and patients between age 15 to 24 that because attempt to suicide be care in Farabi hospital. Boys (M age = 15.42, SD = .88) and girls (M age = 15.86, SD = 1.04) did not differ significantly in age, t (221) = .21, p = .83. Most of the participants were Kurd (94.4%), 3.1% were Lack, and 2.5% were Fars. Data collected from the total sample of participants were used to assess the factor structure of the RFL–A. To explore additional psychometric properties of the RFL–A, we collected complete data on the measures used in this study on a subsample (n = 96; 54boys and 42 girls) of participants (see Measures and Procedure section). We assigned these participants to two groups based on information obtained by author and a review of the medical records. In addition to the semi structured (i.e., clinical interviews).

Participants in the group suicide (13 boys and 25 girls) with a history of multiple suicide attempts who were admitted because of a recent (within 1–2 weeks prior to admission) suicide attempt (self-harm or injury with established intent to die) were assigned to the attempter group (n = 14). The method of attempts identified included drug or medication overdoses (n = 5), self-inflicted lacerations (n = 3), hanging (n = 8), attempts to use a gun (n = 3), car accidents (n= 2), and jumping from heights (n =3). Participants in the normal group (176 boys and 186 girls) who had no previous history of suicide attempts.

 

Measures and Procedure

 

Each participant completed a brief demographic questionnaire, the RFL–A, the Beck Suicide Scale Ideation (BSSI), and Oxford Happiness Inventory (Argyle et al, 1987).

Reasons for living inventory for adolescents (RFL-A; Osman et al., 1998). The RFL–A is a 32-item self-report measure designed specifically to assess adolescents’ adaptive reasons for not committing suicide. It is comprised of five factors: Future Optimism (FO), Suicide-Related Concerns (SRC), Family Alliance (FA), Peer Acceptance and Support (PAS), and Self-Acceptance (SA). less relevant items (e.g., relating to concerns about the effects of suicide on one’s children) are not included in the RFL–A. The factor structure of the RFL–A is consistent with the multifaceted nature of adolescent suicidality (Osman et al., 1998). The authors also found support for convergent and construct validity. 

 

Beck Suicide Scale Ideation (BSSI; Beck et al., 1974). This 19-item scale is designed to assess prior suicide ideation and behavior, frequency of suicide ideation, threats of suicide, and likelihood of attempting

Suicide someday. The BSSI has been used in several investigations with adolescents and young adults. The BSSI was used as a measure of self-reported suicide likelihood in validating the RFL–A scales. In this study we use from BSSI to assess divergent validation.

 

Oxford Happiness Inventory (OHI; Argyle et al, 1987). The OHI contains 29 items designed to assess the happens. It also assesses four dimensions of suicidality: happens, hope and positive expectations about future events. Each OHI item is rated on a 4-point scale ranging from 1 (none or a little of the time) to 4 (most or all of the time). The SPS has good reliability and concurrent validity (Tatman, Greene, & Karr, 1993). We used this scale as a measure to assess convergent validation

Beck Hopelessness Scale (BHS; Beck et al., 1974).The BHS is a 20-item self-report instrument with a true–false response format. As in previous investigations, this scale has been used in several investigations to assess the extent of negative expectations about future events (see Joiner & Rudd, 1996; Marano, Cisler,& Lemuroid, 1993).

We collected data from each participant within 4 weeks of admission. Participation in the study was voluntary. During data collection, the second author or a practicum student in psychology (all trained in the administration of the research package) approached and asked each potential participant to volunteer to participate in the study. Next, the study was briefly explained, informed consent was obtained, and the questionnaire package was administered individually. Approval for conducting the study was obtained from the hospital administrator and the Medical Sciences University of Kermanshah. The protocol also included obtaining adolescent assent and significant other (legal guardians and parents) written informed consent before administering the questionnaire packet and reviewing the medical records.

 

 

 

 

Data analyses:

                 Based goals of study used from below data analyses:

For analyses material of scales used from classic test model
Statically features of material of scales assess by descriptive statistics
Reliability of items each scale assess by kornbakh coefficient and retest
For assess factor validation and determine number factors of scale used from pc style
For calculation divergent validation correlation between RFL-A and BSSI assessed.
For calculation convergent validation correlation between RFL-A and OHI assessed.
 For calculation relationship between RFL-A and other variables like age, gender and education used from T-test and correlation.    
  For calculation discriminate validation and comparison mean of two group (suicide and normal) used from T-test.

Results

 

Reliability Analysis

We examined the internal consistency reliability of the RFL–A total and scales for the combined sample before evaluating the validity of this new instrument. The alpha coefficients for the RFL–A scales were as

Follows: FA = .88, SRC = .92, SA = .91, PAS = .89, and FO = .90. The corrected item-total correlation for each scale was greater than .40. The alpha index for the RFL–A total scale was .93. These findings are consistent with those reported by Osman et al. (1998). And result of retest after 2 weeks on subsample (n=50) was .87. in the table 1 we can see mean, std. deviation, Corrected Item-Total Correlation and Cronbach’s Alpha if Item Deleted  all question of  RFL-A.

 

 

 

 

 

 

 

 

table 1 mean, std. deviation, Corrected Item-Total Correlation and Cronbach’s Alpha if Item Deleted RFL-A

 

Item

Mean

Std. Deviation

Corrected Item-Total Correlation

Cronbach’s Alpha if Item Deleted

 

Item

Mean

Std. Deviation

Corrected Item-Total Correlation

Cronbach’s Alpha if Item Deleted

Q1

3.3595

1.11172

.453

.961

Q17

3.8658

1.44078

.453

.960

Q2

4.0177

1.42216

.607

.960

Q18

1.2380

1.67819

.127

.959

Q3

3.8506

1.32353

.455

.960

Q19

3.6076

1.49464

.680

.959

Q4

4.0658

1.35396

.635

.960

Q20

3.7722

1.67657

.685

.960

Q5

3.9747

1.42115

.170

.960

Q21

3.6608

1.71921

.723

.960

Q6

2.5215

1.72807

.339

.961

Q22

3.7899

1.44226

.697

.959

Q8

3.7722

1.61331

.519

.961

Q23

3.4810

1.62047

.332

.961

Q9

3.3772

1.02109

.393

.961

Q24

3.9316

1.39720

.793

.959

Q10

3.5696

1.80160

.433

.960

Q25

3.4025

1.99173

.521

.959

Q11

3.5266

1.59032

.620

.960

Q26

1.8582

1.89580

.418

.959

Q12

3.3443

1.77774

.687

.960

Q27

3.6076

1.64962

.387

.960

Q13

3.5595

1.68402

.601

.961

Q28

3.4709

1.67121

.700

.959

Q14

3.3696

1.69205

.601

.961

Q29

3.6025

1.63605

.745

.959

Q15

2.0937

1.76941

.409

.959

Q30

3.3392

1.11808

.605

.960

Q16

3.3873

.78004

.375

.959

Q31

2.0532

1.75526

.303

.960

Q17

3.8658

1.44078

.453

.960

Q32

3.8177

1.52881

.746

.959

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PC

Because we can use PC model and achieve this note that  data correlation is not zero we should applied Bartlett Test Of Sphericity before PC model and then used PC to examine the five – factor oblique model reported by Osman et al(1998). However we see in the table 2 KMO is .92 and significant (.0001) and therefore we can do factor analysis in the sample group.

Because extraction factors fit with social and cultural structure of sample group , in the section explorative factor analysis we examine the one – factor, two – factor, three – factor, four – factor and five – factor solution model. In the end of this section appears that factor solution model have better and equated with data and were able to derive unique five-factor solutions accounting for 57.8% of the variance in adolescents data(Table 3). 

 

 

Table 2.KMO and Bartlett’s Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy.

.927

Bartlett’s Test of Sphericity

Approx. Chi-Square

1.193E4

 

df

1134

 

Sig.

.000

 

Table 3.Total Variance Explained

Component

Initial Eigenvalues

Extraction Sums of Squared Loadings

Total

% of Variance

Cumulative %

Total

% of Variance

Cumulative %

1

11.497

35.928

35.928

11.497

35.928

35.928

2

3.127

9.772

45.700

3.127

9.772

45.700

3

1.365

4.267

49.967

1.365

4.267

49.967

4

1.286

4.018

53.984

1.286

4.018

53.984

5

1.253

3.915

57.899

1.253

3.915

57.899

 

For a mixed sample of suicidal and normal adolescents. In addition, we specified and evaluated the fit of 5 competing models, a one- factor, two- factor, three- factor, four- factor and five factors for select best solution way in factor analyses. After do this models it is appears that best solution way for factor analyses is five factors. Table 4 present the RFL-An items internal consistency (alpha coefficients) and descriptive statistics (skewness and kurtosis) for each factor.

                Table 4. Reasons for living inventory for adolescents, internal consistency and descriptive statistics

Factor

alpha coefficients

skewness

kurtosis

1.Family Alliance (FA)

.88

- 0.91

- 0.21

2. Suicide-Related Concerns (SRC)

.92

- 0.59

- 1.08

3. Self-Acceptance (SA)

.91

- 0.86

- 0.16

4. Peer Acceptance and Support (PAS)

.89

- 1.06

0.53

5. Future Optimism(FO)

.90

- 1.32

1.17

Based on results PC model of factor analysis material of scale, number factors of RFL-A in the Iranian population was five factors. Because this scale for first time used in this population lower limit of load factors .35 determined (Hooman, 2001). Factors Structure, coefficient reliability and standard error of measurement each factor presented in Table 5 and in the Table 6 we can see Principal Component Analysis with Promax Rotation Method.

Table 5. Factors Structure, coefficient reliability and standard error of measurement five factor

Factors

Number of questions

coefficient reliability

Std. error of measurement

Family Alliance

7

.88

23/04

 Suicide-Related Concerns

6

.92

10/92

 Self-Acceptance

6

.91

8/64

Peer Acceptance and Support

6

.89

7/98

Future Optimism

7

.90

6/49

Total

32

.93

12/85

                               

                                  Table 6. Principal component analysis Structure Matrix

 

Component

 

1

2

3

4

5

Q24RFL

.833

 

 

 

 

Q25RFL

.722

 

 

 

 

Q30RFL

.716

 

 

 

 

Q17RFL

.712

 

 

 

 

Q7RFL

.706

 

 

 

 

Q12RFL

.575

 

 

 

 

Q1RFL

.472

 

 

 

 

Q13RFL

 

.747

 

 

 

Q20RFL

 

.746

 

 

 

Q15RFL

 

.624

 

 

 

Q4RFL

 

.524

 

 

 

Q11RFL

 

.486

 

 

 

Q28RFL

 

.485

 

 

 

Q26RFL

 

.406

 

 

 

Q3RFL

 

 

.750

 

 

Q9RFL

 

 

.724

 

 

Q29RFL

 

 

.686

 

 

Q14RFL

 

 

.634

 

 

Q18RFL

 

 

.612

 

 

Q31RFL

 

 

.501

 

 

Q10RFL

 

 

 

.755

 

Q16RFL

 

 

 

.648

 

Q5RFL

 

 

 

.612

 

Q27RFL

 

 

 

.579

 

Q23RFL

 

 

 

.558

 

Q6RFL

 

 

 

.434

 

Q32RFL

 

 

 

 

.670

Q22RFL

 

 

 

 

.604

Q19RFL

 

 

 

 

.555

Q21RFL

 

 

 

 

.548

Q8RFL

 

 

 

 

.514

Q2RFL

 

 

 

 

.488

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Because of the confirmation number factors and explorative factor analysis we used the confirmatory factor analysis and following indexes to evaluate the fit of each model:

1)    A relative robust chi-square of 2 or less.

2)    Bentler and Bonett normed fit index(NFI) of .90 or greater,

3)    Bentler and bonnet non-normed fit index (NNFI) of .90 or greater, robust comparative fit index(R-CFI) of .90 or greater.

4)    Root mean squared residual index (RMSR) of .05 or less (see Bentler, 1995; Bentler & Bonett, 1980; Marsh, Balla, & McDonald, 1988).

Results for the one, two, three, four and five factor solution model tested are presented in the Table 3. The five-factor model provided the best solution way model fit to the data. The Satorra-Bentler index (1.34) was less than 2, and NFI, NNFI, and R-CFI values were greater than .90. Also, the RMSR index was less than .05. These results suggest that the five factor solution way model can be reliably replicated in an adolescent sample.

                    Table 7.Confirmatory Factor Analysis of the Reasons for Living Inventory for Adolescents for five models

 

Model

 

c2

d¦

d¦1c2

GFI

AGFI

CFI

RMSR

Five –Factor

32/2041

1012

859/1

94/.

89/.

92/.

032/.

Four –Factor

15/3468

1024

452/3

87/.

83/.

90/.

075/.

Three –Factor

05/3858

1057

734/3

84/.

84/.

89/.

093/.

Two –Factor

00/2056

1069

151/2

86/.

81/.

88/.

04/1

One-Factor

21/1834

1069

954/2

85/.

79/.

89/.

09/.

 

Discriminate, divergent and convergent validity

We conducted planned comparisons to determine divergent and convergent validity of scale and determine whether the RFL–A scales can distinguish adolescent based on suicide status. Results for achieve divergent and convergent validity in the Table 8 showed there was a positive correlation between, RFL–A and Oxford Happiness Inventory (OHI; Argyle et al, 1987) and negative correlation between  this scale and Beck Suicide Scale Ideation (BSSI; Beck et al., 1974).

 

 

 

 

 

 

 

 

                      Table 8.Correlations between the RFL–A and Concurrent Validity Measures

 

 

RFL–A  Scales

 

Measures

FA

SRC

SA

PAS

FO

 

RFL–A

 

Oxford Happiness Inventory

31*

.37*

 

39*

 

30*

 

.35*

 

.39*

 

Beck Suicide Scale Ideation

–.50*

–.44*

 

–.51*

–.52*

 

–.61*

–.48*

 

Beck Hopelessness Scale

–.60*

 

–.53*

 

–.57*

 

–.63*

 

–.66*

 

–.61*

 

Note: RFL–A = Reasons for Living Inventory for Adolescents; FA = Family Alliance; SRC = Suicide-Related Concerns; SA = Self-Acceptance; PAS = Peer Acceptance and Support; FO = Future Optimism;.

*p < .001, **p < .005

 

For comparisons between the suicide (n = 38) and normal (n = 40) groups, the overall t-test was significant, t = .78, F (5, 168) = 26.04, p < .001; η2= .44. The normal group scored significantly higher than did the attempter group on all five RFL–A scales (all p values < .001). (See Table 9 and Table 10).

Pearson product–moment correlations were computed between the RFL–A total and scales, and the con-current validity measures (the BSSI, BHS and OHI scales). The results are presented in Table 8 for the total sample. The analyses showed that all of the RFL–A total and scale scores were negatively and significantly correlated with scores on BSSI items (range = –.44 to –.61). Similarly, negative and significant correlations were obtained between scores on the RFL–A total and scales and scores on the BHS. And also positive and significant correlations were obtained between scores on the RFL–A total and scales and scores on the OHI. These results showed that RFL-A have good discriminate, divergent and convergent validity and specially can discriminate Suicide Attempters from normal Adolescents.

 

Table 9.Independent Samples Test For comparisons between the suicide and normal group

 

 

F

t

df

Sig. (2-tailed)

Suicide-normal groups

Equal variances assumed

19.589

4.280

76

.000

 

Equal variances not assumed

 

4.330

65.313

.000

 

 

 

 

 

 

 

Table 10. Means and Standard Deviations on the Reasons for Living Inventory for Adolescents for normal and suicide attempters

 

RFL–A Scale

 

Suicide Attempters

 

Normal group

M

SD

M

SD

1. FA

 

3.23

1.34

4.88

1.10

2. SRC

 

3.11

1.51

4.68

1.29

3. SA

 

3.34

1.28

5.05

0.66

4. PAS

 

4.02

1.39

5.41

0.84

5. FO

 

3.73

1.37

5.38

0.75

   Note: RFL–A = Reasons for Living Inventory for Adolescents; FA = Family Alliance; SRC = Suicide-Related Concerns; SA = Self-Acceptance; PAS = Peer Acceptance and Support; FO = Future Optimism.

 

Discussion:

The purpose of this study was to assess the reliability, validity, and standardization of the Reasons for Living Inventory for Adolescents (RFL–A; Osman et al., 1998) among Iranian Adolescents (Kermanshah city). Several tools have been developed to help counselors and psychologists in determining which adolescents are most likely to attempt to kill themselves and which group is at the greatest risk for suicide. The majority of research and existing measures in the felid of adolescents’ suicide focus on negative predictors of risk like depression, hopelessness and history of prior attempt for suicidal behavior. A major character of Linehan’s RFL (Linehan et al., 1983) is opposite the negative predictor approach that assesses potentially life-threatening crises, this approach allows for the assessment of adaptive reasons for living. Although this measure has been used with varying degrees of success with young adults and adolescents, it has certain limitations (e.g., Cole, 1989; Hirsch & Ellis, 1996; Osman et al., 1996; Pinto et al., 1998). Because these limitations Osman et al. (1998) created a psychometrically sound measure for adolescents based on the theoretical constructs underlying the RFL.

The results of this study add support to existing data (Osman et al., 1998) on the reliability and validity of the RFL–A. Our study also provides preliminary normative data for non suicide (normal) and suicide attempter adolescents in the Kurdish adolescents (city Kermanshah). Although no differences were found on subscale scores within the suicide attempter adolescents, girls in the attempter group scored lower on FA, PAS and SRC, suggesting a pattern of suicide and decrease reasons for living in the Kermanshah that opposite with world pattern (women 3 times more men have successful suicide). Perhaps girls who engage in serious suicidal behavior have lost their connections to family and friends, and may these lower scores indicate increase tolerance of pain (as one components of suicide). It is unclear why boys who have made an attempt would get higher scores in these subscales in the same group. This finding will need to be explored in future

The total RFL–A score was very useful in distinguishing between the non suicidal (normal), and suicide attempter groups. As was expected, adolescents in the non suicidal group had the highest total scores, and suicide attempter groups’ scored lowest. A significant amount of variance in scores on suicide probability were explained by the RFL–A. Low levels reasons for living appear to be indicators of greatest suicide risk. Scores on the RFL–A were also predictors of achieve scores in suicide ideation. Our analyses indicated that more hopeless and more suicide ideation of adolescents have limited optimism about the future, low levels of peer acceptance and support, and a weak sense of alliance with their families. These findings congruent with prior studies (Osman et al, 1998; Gutierrez et al, 2000).

 

In the several areas the RFL–A appears relate to adolescent suicide risk. For example, the SA subscale predicts depression, anger, alienation, and family problems (Gutierrez et al, 2000). The RFL–A can provide specific guidance to clinicians and counselors on where and when do interventions and assessment for improvement desire to life and decrease suicide in adolescents.

Additional support for the discriminative ability of the RFL–A comes from the results of the t scores and discriminate validity analyses. The RFL–A was found to be a significantly better than other scales that widely used tool for this purpose (Joiner & Rudd, 1996) can predicate suicide like hood. Results showed that the RFL–A and the theory underlying measures of adaptive functioning (e.g., Linehan et al., 1983; Osman et al., 1996; Westfield et al., 1992) is useful in assessing suicide risk and this scale more than original RFL (Linehan et al, 1983) have sensitivity, specificity, and predictive value to suicide adolescents. However, the developmentally appropriate RFL–A in the Iran especially in the Kermanshah city because higher statics of suicide is superior for use with adolescents.

A few limitations of this study must be discussed. The majority of participants were Kurdish, making it difficult to       determine the racial and ethnic generalizability of the findings. Future studies should attempt to utilize more ethnically and racially diverse Participants like Turkmen, Fars, Lor and Turk people.

It has long been accepted that adolescent suicide risk is multiply determined (Brent, Moritz, Bridge, Perper, & Canobbio, 1996; Pfeiffer, Newcorn, Kaplan, Mizruchi, & Plutchik,1988). Much of the existing literature focuses on negative risk factors (Beck et al., 1974).  Only focus on risk factors in the assess suicide is incomplete and inadequate. We are believed that an accurate picture of risk can only be constructed from protocols that include both measures of negative factors and protective elements. We results of our study showed that RFL–A as a reliable, valid, and clinically useful tool for assessing adolescent suicide risk in the Kermanshah city.

 

 

References

Argyle, M., Martin, M., R., & Crossland, J. (1989). Happiness as a function of personality and social encounters. In J. P. Frogras & R. J. M. Innes. Recent advances in social psychology: an international prespective.189-203

 

Beck, A. T., Kovacs, M., & Weismann, A. (1979). Assessment of suicidal ideation: The scale for suicide ideation. The Journal of Consulting and Clinical Psychology, 47, 343–352.

Beck, A. T., Weismann, A., Lester, D., & Trexler, M. (1974). The measurement of pessimism: The Hopelessness scale. Journal of Consulting and Clinical Psychology, 42, 861–865.

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness-of-fit in the analysis of covariance structures. Psychological Bulletin, 88, 588–606.

Bentler, P. M., & Wu, E. J. C. (1998). EQS structural equations program (Version 5.7). Encino, CA: Multivariate Software.

Brent, D. A., Moritz, G., Bridge, J., Perper, J., & Canobbio, R.(1996). Long-term impact of exposure to suicide: A three-year controlled follow-up. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 646–653.

Cole, D. A. (1989). Validation of the Reasons for Living Inventory in general and delinquent adolescent samples. Journal of Abnormal Child Psychology, 17, 13–27.

Connell, D. K., & Meyer, R. G. (1991). The Reasons for Living Inventory and a college population: Adolescent suicidal behaviors, beliefs, and coping skills. Journal of Clinical Psychology, 47, 485–489.

Gutierrez, P., King, C. A., & Ghaziuddin, N. (1996). Adolescent attitudes about death in relation to suicidality. Suicide and Life-Threatening Behavior 26, 8–18.

Gutierrez, P., Osman, A., Kopper, B. A., Barrios, F. X., (2000).Why Young People Do Not Kill Themselves: The Reasons for Living Inventory for Adolescents. Journal of Clinical Child Psychology, 2000, Vol. 29, No. 2, 177–187

Hirsch, J. K., & Ellis, J. B. (1996). Differences in life stress and reasons for living among college suicide Ideators and non- ideators.College Student Journal, 30, 377–386.

Hooman, H. A.(2000). Study of validity of ratings. Psychological reports. 51, 1263-420

Ivanoff, A., Jang, S. J., Smyth, N. J., & Linehan, M. M. (1994). Fewer reasons for staying alive when you are thinking of killing yourself: The Brief Reasons for Living Inventory. Journal of Psychopathology and Behavioral Assessment, 16, 1–13.

Joiner, T. E., & Rudd, D. M. (1996). Disentangling the interrelations between hopelessness, loneliness, and suicidal ideation. Suicide and Life Threatening Behavior, 26, 19–26.

Kann, L., Kinchen, S. A., Williams, B. I., Ross, J. G., Lowry, R., Hill, C. V., Grunbaum, J. A., Blumson, P. S., Collins, J. L., & Kolbe,L. J. (1998). Youth risk behavior surveillance—United States, 1997. In CDC Surveillance Summaries, August 14, 1998. Morbidity and Mortality Weekly Reports, 47, 1–89.

Kovacs, M., Goldstone, D., & Gatsonis, C. (1993). Suicidal behaviors and childhood-onset depressive disorders: A longitudinal investigation. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 8–20.

Kralik, K. M., & Dan forth, W. J. (1992). Identification of coping ideation and strategies preventing suicidality in a college-age sample. Suicide and Life-Threatening Behavior, 22, 167–186.

Linehan, M. M., Goodstein, J. L., Nielsen, S. L., & Chiles, J. A.(1983). Reasons for staying alive when you are thinking of killing yourself: The Reasons for Living Inventory. Journal of Consulting and Clinical Psychology, 51, 276–286.

Marano, C. D., Cisler, R. A., & Lemerond, J. (1993). Risk factors for adolescent suicidal behavior: Loss, insufficient family support, and hopelessness. Adolescence, 28, 851–865.

Marsh, H. W., Balla, J. R., & McDonald, R. P. (1988). Goodness-of fit indexes in confirmatory factor analysis: The effect of sample size. Psychological Bulletin, 103, 391–410.

Orbach, I., Milstein, I., Har-Even, D., Apter, A., Tyano, S., & Elizur,. (1991). A multi-attitude suicide tendency scale for adolescents. Psychological Assessment, 3, 398–404.

Osman, A., Downs, W. R., Kopper, B. A., Barrios, F. X., Besett, T.M., Linehan, M. M., Baker, M. T., & Osman, J. R. (1998). The Reasons for Living Inventory for Adolescents (RFL–A): Development and psychometric properties. Journal of Clinical Psychology, 54, 1063–1078.

Osman, A., Gifford, J., Jones, T., Lickiss, L., Osman, J., & Wenzel, R. (1993). Psychometric evaluation of the Reasons for Living Inventory. Psychological Assessment, 5, 154–158.

Osman, A., Gregg, C. L., Osman, J. R., & Jones, K. (1992). Factor structure and reliability of the Reasons for Living Inventory. Psychological Reports, 70, 107–112.

Osman, A., Jones, K., & Osman, J. R. (1991). The Reasons for Living Inventory: Psychometric properties. Psychological Reports, 69, 271–278.

Osman, A., Kopper, B. A., Barrios, F. X., Osman, A., Besett, T., & Linehan, M. M. (1996). The Brief Reasons for Living Inventory for Adolescents (BRFL–A). Journal of Abnormal Child Psychology, 24, 433–443.

Pfeffer, C. R., Newcorn, J., Kaplan, G., Mizruchi, M., & Plutchik, R.(1988). Suicidal behavior in adolescent psychiatric inpatients. Journal of the American Academy of Child and Adolescent Psychiatry, 27, 357–361.

Piers, E. V., & Harris, D. B. (1996). The Piers–Harris Children’s Self-Concept Scale. Los Angeles: Western Psychological Services.

Pinto, A., Weismann, M. A., & Conwell, Y. (1998). Reasons for living in a clinical sample of adolescents. Journal of Adolescence, 21, 397–405.

Range, L. M., Hall, D. L., & Meyers, K. (1993). Factor structure of adolescents’ scores on the Reasons for Living Inventory. Death Studies, 17, 257–266.

Rogers, J. R., & Hanlon, P. J. (1996). Psychometric analysis of the College Student Reasons for Living Inventory. Measurement and Evaluation in Counseling and Development, 29, 13–24.

Satorra, A., & Bentler, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis. In A. Van Eye & C. C. Clogg (Eds.), Latent variable analysis: Applications for developmental research (pp. 399–519)..

Strosahl, K., Chiles, J. A., & Linehan, M. M. (1992). Prediction of suicide intent in hospitalized parasuicides: Reasons for living, hopelessness, and depression. Comprehensive Psychiatry, 33, 366–373.

Social Welfare and Rehabilitation Organization (2006). Suicide: Final data for 2006. National Vital Statistics Report for suicide, 21 (SWRO Publication No. 3 99–112).

Westfield, J. S., Bandura, A., Kiel, J. T., & Scheel, K. (1996). The College Student Reasons for Living Inventory: Additional psychometric data. Journal of College Student Development,37, 348–350.

Westfield, J. S., Cardin, D., & Deaton, W. L. (1992). Development of the College Student Reasons for Living Inventory. Suicide and Life-Threatening Behavior, 22, 442–452.

 

Osman Mahmoudi M.A of Family CounselingUniversity of Social Welfare and Rehabilitation of IranRahmanberdi Ozoni Davaji, Anahita Khodabakhshi KoolaeeDepartment of Counseling, University of Social Welfare and Rehabilitation of Iran
Article Source

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>