Journal of Rural Community Psychology               Volume E7   Number 1   Spring 2004

 

Adolescent Preferences for Help with Mental Health Problems

 

John D. Tyler, Rebecca Green, and Don Newberry

 

University of North Dakota

 

ABSTRACT

 

Predominantly Caucasian students from the seventh, ninth and eleventh grades of a secondary school located in a mid-western community of 80,000 were asked to rate their willingness to recommend several resources for help with mental health problems. Results indicated that mental health professionals were seen as more preferable help providers than family or friends who in turn were rated higher than physicians or clergy. Relying on oneself was seen as least desirable, particularly by females.

 

INTRODUCTION

 

Mental disorders are the major cause of disability among adolescents (National Center for Education and Maternal and Child Health, 1990). Adolescent suicide rates nearly tripled between 1968 and 1985 and depression affects between 7 and 33 percent of this age group (Takanishi, 1993). Also prominent are high risk behaviors (Quadrel, Fischoff & Davis, 1993), and exposure to violence (Hammond & Yung, 1993) are prominent.

 

Given the scope and magnitude of such problems a multi-faceted approach to addressing adolescent mental health is indicated (Kasdin, 1993). Basic research to identify dynamic factors in the upsurge of problematic behaviors is essential. So are more applied and pragmatic attempts to address implicated pathogens. Also important are efforts to increase the access to mental health services of dysfunctional and at risk adolescents.

 

It has been estimated by the U.S. Congress office of Technology Assessment (OTA, 1991) that only one third of the children and adolescents who need mental health services receive them. Various factors including unavailability of services, financial constraints, and fear of peer disapproval may contribute to underuse. However it is also possible that some adolescents forego mental health services because their views about them differ from those of adult care providers. At present there is little information about adolescents' perspectives about the need for and availability of health services generally (Dougherty, 1993). Efforts to promote appropriate utilization of mental health care should be informed by determining adolescent perspectives about appropriate resources for help. The present investigation was undertaken to explore this issue.

 

To assess adolescent perspectives about appropriate treatment/management for mental health problems, a modification of the Sources of Referral Questionnaire (Suan & Tyler, 1990) was used. Adolescent subjects were asked to respond to a set of vignettes descriptive of common emotional and behavioral problems by indicating their preference for each of six resources for assistance: clergy, friend, physician, family member, mental health professional and self.

 

METHOD

 

Subjects

 

Subjects consisted of 183 female and 198 male 7th, 9th and 11th grade students at a Junior/Senior high school in a midwestern community of 80,000. Students were administered questionnaires during regular classes by the second author who identified herself as a psychology graduate student. They were informed that they were not required to participate in the study and could discontinue at any time. In addition, prior to data collection day, passive consent forms describing the study were sent home which invited parents to contact the investigators if they did not wish for their child to participate. Subjects were instructed to omit their names from questionnaires and were assured that their responses would be collected anonymously. Of the total school populations, 85% of 7th graders, 89% of 9th graders and 84% of 11th graders participated and provided usable data.

 

Instruments

 

Subjects completed a booklet containing three questionnaires. The first of these was used to collect general background information about the subject: gender, age, grade, ethnicity, years of education completed by both parents and religious orientation. A second questionnaire containing ninety-nine items inquiring about mental health values was used for another study.

 

The third questionnaire, the Sources of Referral Questionnaire, contained five vignettes, each describing a situation in which "someone" the subject knows is having a personal difficulty. Types of personal difficulties consisted of: a marital problem; a child management difficulty; problems with depression and anxiety; persecutory auditory hallucinations; and hostile, aggressive behavior. For each vignette, the subject was asked to rate the suitability (1 = would not recommend; 7 = would strongly recommend) of six different resources for dealing with the problem. The choices for handling each problem consisted of discussion with: a "priest (clergyman)"; "other family members"; "a close friend(s)"; a "physician"; a "mental health professional (psychologist, social worker, or psychiatrist);" or the recommendation that the person "work it out on his/her own".

 

RESULTS

 

Sample Characteristics

 

Subjects were predominantly Caucasian (88.7%). Also represented were Mexican-American (7.61%), American Indian (2.89%) and African-American (.26%) students.

 

Table 1 contains mean ages, years of education completed by mother and by father, and religious orientation ratings (1 = strong; 5 = not religious) for each of the three grades. One way analyses of variance revealed a significant age difference between boys and girls for grade seven, F(1, 149) = 5.73, p < .05; and a nearly significant age difference for grade eleven, F(1, 93) = 3.90, p = .051; with males slightly older than females in both cases.

 

Table 1

 Sample Characteristics

______________________________________________________________________________

 

 

Grade 7

Grade 9

Grade 11

Sex

 

M

F

M

F

M

F

 

 

 

 

 

 

 

 

Age

n

81

70

69

66

48

47

 

M

12.81

12.60

14.88

14.73

16.81

16.

 

SD

  .59

  .49

  .63

  .54

  .53

  .54

 

 

 

 

 

 

 

 

Mother's Education

n

73

66

69

67

47

47

 

M

14.48

13.88

14.16

14.03

14.08

13.15

 

SD

 2.77

 2.14

 2.60

 2.60

 2.16

 2.67

 

 

 

 

 

 

 

 

Father's Education

n

74

62

66

65

47

46

 

M

13.95

14.36

13.98

13.82

14.34

13.5

 

SD

 3.15

 2.69

 2.78

 2.92

 2.35

2.75

 

 

 

 

 

 

 

 

Religious Orientation               

n

80

70

69

67

48

47

   1 = Strong

M

 2.71

 2.50

 2.91

 2.81

 3.21

2.85

   5 = Not religious

SD

 1.14

 1.28

 1.30

 1.22

 1.15

 .86

 

 

 

 

 

 

 

 

 

 

 

Multivariate analyses of variance were conducted to determine whether there were significant differences in religious orientation or parental education between grade level or gender groupings. There were no significant grade level effects, Hotellings T2 = .04, F(6, 692) = 1.96, p = .07; or Grade x Gender effects, Hotellings T2 = .02, F(6, 692) = 1.15, p = .33. While there was a significant overall MANOVA effect for gender, Hotellings T2 = .03, F(3, 347) = 2.76, p = .04; univariate tests for the significance of mother's education, father's education and religious orientation were all nonsignificant, Fs(1, 349) = 3.44, .23, 3.17; ps = .06, .63 and .08, respectively.

 

Help Preference Ratings

 

A four way MANOVA with two between (gender and grade) and two within (type of problem and help option) factors was run with help option ratings as the dependent measure. Both gender [F(1, 373) = 5.91, p = .015] and grade [F(2, 373)= 6.21, p = .002] main effects were significant. Inspection of Figure 1 reveals that females were generally more prone to strongly recommend obtaining help than males and that the general tendency to recommend use of helping resources increased with age. The Gender x Grade interaction was not significant [F(2, 373) = 2.44, p = .09].

 

Both the type of problem within subject main effect [F(4, 1492) = 15.30, p < .001] and the help option within subject main effect [F(5, 1865) = 198.38, p < .001] were also significant.

 

The Gender x Grade x Type of Problem x Help Option interaction was not significant [Hotellings T2 = .12; F(40, 706) = 1.09; p = .33]. However, a significant Gender x Grade x Help Option effect [Hotellings T2 = .06; F(10, 736) = 2.03; p < .03] was obtained. In addition a significant Grade x Help Option interaction was obtained [Hotellings T2 = .06, F(10, 736) = 2.20, p < .02] and the Gender x Help Option interaction was highly significant [Hotellings T2 = .15, F(5, 369) = 10.82, p < .001]. Ratings of help options averaged over the five vignettes are presented as a function of grade and gender in Figure 1.

 

Given the significant higher order interactions, additional tests were conducted. In subsequent analyses, results were summed over the type of problem factor. To determine for each gender whether a Help Option x Grade interaction was present two-way ANOVAs with one between subjects factor (grade) and one within subjects factor (help option) were run. For males there was no significant interaction [F(10, 965) = 1.63, p = .094]. For females a significant interaction was found [F(10, 900) = 1.89, p < .043] indicating that help option ratings varied with the grade. Consequently, for girls ANOVAs were run for each of the six helping resource choices at each grade level. Results indicated a significant grade effect for only the Mental Health Professional help option [F(2, 180) = 8.97, p < .001]. Inspection of mean ratings for this option indicates that preference increased with grade level (Ms = 5.35, 5.87, 6.22; SDs = 1.28, 1.09, .91 for seventh, ninth and eleventh grade girls respectively). A Tukey A test (range = .49, p < .05) indicated that the ratings for grade seven were significantly lower than for grades nine and eleven which did not differ significantly.

 

Next separate two-way ANOVAs with one between subjects factor (gender) and one within subjects factor (help option), were run for each grade. Results revealed significant Gender x Help Option interactions for each grade [F(5, 740) = 3.81, p < .002; F(5, 665) = 8.52, p < .001; and F(5, 460) = 4.55, p < .001 for seventh, ninth and eleventh graders respectively]. Given these findings, for each grade level, one-way ANOVA to test for gender effects were run for each of the six help options. For the seventh graders, only the Friend help option showed a significant gender effect [F(1, 149) = 11.83, p < .001] with females (M = 4.70, SD = 1.29) rating this option higher than males (M = 3.93, SD = 1.38).

 

For ninth graders, significant gender effects were found for four of the six help options. Relative to males, females gave higher rankings to the Friend option [Ms = 5.05 vs. 4.40, SDs = 1.26 vs. 1.26, F(1, 134) = 9.07, p < .003], the Physician option [Ms = 4.79 vs. 3.71, SDs = 1.26 vs. 1.54, F(1, 134) = 20.278, p < .0001], and the Mental Health Professional option [Ms = 5.87 vs. 5.22, SD = 1.03 vs. 1.42, F(1, 133) = 8.99, p < .003]. Interestingly, females rated the "Work it out on his/her own" option less favorably than males [Ms = 2.57 vs. 3.40, SDs = 1.33 vs. 1.54, F(1, 134) = 11.19, p < .001].

 

For grade eleven, there were significant gender differences on two help options. Females (M = 2.90, SD = 1.41) rated "Work it out on his/her own" as significantly less preferable [F(1, 93) = 8.58, p < .004] than did males (M = 3.76, SD = 1.45). However, females rated use of a mental health professional (M = 6.22, SD = .91) as significantly more preferable [F(1, 92) = 8.61, p < .004] than did males (M = 5.51, SD = 1.38).

 

To further evaluate differences in preferences for various help resources, selected within subject comparisons were conducted using correlated t tests and Bonferroni critical values (p < .01). Combining all grade levels and genders, subjects rated use of Mental Health Professionals (M = 5.49, SD = 1.36) higher than either the Family Member option (M = 4.89, SD = 1.30) or the Friend option (M = 4.61, SD = 1.32) [t(388, 388) = 7.42, 9.56; ps < .01, .01 respectively]. Also, Bonferroni comparisons indicated that females (M = 2.74, SD = 1.34) rated "Work it out on his/her own" significantly lower than the Priest option (M = 3.40, SD = 1.74) [t(187) = 4.20, p < .01] whereas the corresponding difference for males was not significant (M = 3.44, SD = 1.49 vs. M = 3.31, SD = 1.66) [t(203) = .75, p = .45].

 

Table 2

Mental Health Values Mean Scale Scores for Different Grades

 

 

 

Seventh

Ninth

Eleventh

MHVQ Scale

 

Males

Females

Males

Females

Males

Females

 

 

(N = 80)

(N = 68)

(N = 69)

(N = 67)

(N = 48)

(N = 47)

 

 

M

SD

M

SD

M

SD

M

SD

M

SD

M

SD

Self-Acceptance

4.10

 .43

4.24

 .38

4.17

 .40

4.31

 .38

4.24

 .36

4.39

 .34

Negative Traits

2.22

 .51

2.14

 .44

2.44

 .48

2.14

 .48

2.42

 .42

2.21

 .44

Achievement

 

3.76

 .49

3.90

 .42

3.67

 .38

3.89

 .44

3.75

 .41

3.83

 .45

Affective Control

3.47

 .45

3.49

 .33

3.46

 .45

3.49

 .43

3.55

 .33

3.54

 .38

Interpersonal Relations

4.12

 .52

4.37

 .35

4.22

 .40

4.40

 .42

4.28

 .32

4.32

 .39

Untrust-worthiness

2.14

 .42

1.98

 .38

2.11

 .38

1.98

 .44

2.07

 .29

1.98

 .30

Religious Committment

4.10

 .45

4.20

 .44

3.95

 .47

4.13

 .36

3.99

 .39

4.09

 .39

Unconventional Experiences

2.61

 .60

2.64

 .60

2.41

 .53

2.49

 .50

2.54

 .60

2.37

 .61

 

Table 3

Adolescent versus Psychologist Mental Health Value Differences

 

 

MHVQ Scale

Seventh

Grade

N = 151

 

Ninth

Grade

N = 136

 

Eleventh

Grade

N = 95

 

Psychogists

 

N = 107

 

 

 

F

 

 

df

 

M

SD

 

M

SD

 

M

SD

 

M

SD

 

 

 

Self-

Acceptance

4.17

 .41

 

4.24

 .40

 

 4.31

 .36

 

 4.19

 .31

 

 3.30*

3,485

Negative

Traits

2.19

 .48

 

2.30

 .48

 

 2.32

 .44

 

 2.56

 .24

 

15.97**

3,484

Achievement

3.83

 .46

 

3.78

 .42

 

 3.79

 .43

 

 3.33

 .27

 

37.12**

3,484

Affective

Control

3.48

 .40

 

3.47

 .44

 

 3.55

 .35

 

 3.33

 .36

 

 5.58**

3,484

Interpersonal

Relations

4.24

 .46

 

4.31

 .42

 

 4.30

 .36

 

 4.09

 .33

 

 6.91**

3,485

Untrust-

worthiness

2.06

 .41

 

2.05

 .42

 

 2.02

 .30

 

 2.32

 .23

 

16.24**

3,485

Religious

Committment

4.15

 .44

 

4.04

 .43

 

 4.04

 .39

 

 3.75

 .24

 

22.36**

3,485

Unconventional

Experiences

2.62

 .60

 

2.45

 .52

 

 2.45

 .61

 

 2.02

 .53

 

24.74**

3,485

   

            * p < .025               ** p < .001

 

 

DISCUSSION

 

Adolescents are an underserved population with respect to mental health care. Many factors may contribute to this underuse. For example, the recently challenged belief that psychological turmoil is a normal aspect of adolescence, hence not deserving of special treatment (Kasdin, 1993), the unavailability of parents to accompany adolescent children for services (Dougherty, 1993) and the paucity of research on psychotherapeutic interventions developed and validated specifically for adolescents (Kasdin, 1993). Given the influence adolescents are presumed to wield over other choices which affect them (e.g. music, clothing) it also appears important to examine adolescent attitudes about the desirability of various resources for mental health assistance.

 

Our results yielded several interesting findings about how adolescents perceive resources for help with personal problems. Relative to younger adolescents, older subjects indicated a stronger general tendency to recommend use of helping resources. This trend is particularly interesting in view of evidence (Lewis, 1981) that older adolescents appear more sensitive to risks than younger adolescents. Together both findings suggest that adult caretakers should be targeted by programs intending to promote greater use of mental health care by younger adolescents.

 

A second general trend was that girls were more favorably disposed toward most of the helping options studied than were boys. Whether this reflects greater sensitivity to problems, a greater need to find help for them, a greater willingness to rely on others for help or some other factor deserves further study. Consistent with this trend girls at both the ninth and the eleventh grade level rated "Work it out on his/her own" significantly less favorably than did their male counterparts.

 

An unexpected finding (Figure 1) was the fact that at all grade levels professional mental health workers were seen as the most desirable resource to use with various personal problems. In contrast the choice of solving problems oneself or using a clergy person generally was ranked lowest. Given that the experimenter identified herself as a psychology graduate student, it is possible that this finding reflects a bias toward pleasing the experimenter. However the fact that subjects were asked to respond anonymously was intended to mitigate this possibility.

 

 

If one assumes that rankings below 4 (midpoint on the 7 point rating scale) indicate lack of confidence in a particular referral source, both clergy and "Work it out on his/her own" were viewed as negative choices by respondents while physicians were viewed more neutrally. Family, friends and mental health professionals were viewed as positive choices.

 

The finding that relative to males ninth and eleventh grade female subjects gave significantly lower ratings for the "Work it out on his/her own" option is consistent with Gilligan's (1982) observation that during early adolescence, females undergo a significant reduction in their general self-confidence. This gender difference did not manifest at the seventh grade level.

 

Also interesting is the fact that relative to boys, seventh and ninth grade girls were more positive about seeking help from friends. Examination of Figure 1 indicates that from grade seven to grade eleven boys rated peer use increasingly more positively while girls maintained about the same level of preference across grades. Interesting too is the fact that relative to boys, girls were more positive about use of physicians as resources for mental health problems, a finding that only achieved significance for ninth graders, however.

 

It should be noted that our results are based on a relatively homogeneous sample of predominantly Caucasian middle class adolescents from a moderate sized mid-western community. They may not be representative of individuals from different racial, socioeconomic or regional populations. Given this caution several practical implications are tentatively offered.

 

Present findings indicate that at least for adolescents similar to our sample, professional helping resources are more attractive than family members or peers. Perhaps for similar populations it would be better to improve access to professional mental health resources than to develop alternatives such as peer counseling programs. Given that many rural teenagers live at some distance from professional mental health services and may not drive, programs to improve physical access to such care are desirable. Such programs should take into account adolescent sensitivities about negative peer judgements and emphasize privacy and confidentiality  Further research to explore the views of adolescents about barriers to accessing professional mental health care and acceptable remedies would be useful.

 

REFERENCES

 

Dougherty, D. N. (1993). Adolescent health: Reflections on a report to the U.S. Congress.

            American Psychologist, 48, 193-201.

 

Gilligan, C. (1982). In a different voice: Psychological Theory and women's           

            development.  Harvard University Press: Cambridge, MA.

 

Hammond, W. R., & Yung, B. (1993). Psychology's role in the public health response to  

            assaultive violence among young African-American males. American

            Psychologist, 48, 142-154.

 

Kasdin, R. E. (1993). Adolescent mental health: Prevention and treatment programs.

            American Psychologist, 48, 127-141.

 

Lewis, C. (1981). How adolescents approach decisions: Changes over grades seven to      

            twelve and policy implications. Child Development, 53, 538-544.

 

National Center for Education in Maternal and Child Health (1990). The health of            

            America's youth. Washington, DC: Author.

 

Quadrel, M. J., Fischoff, B. and Davis, W. (1993). Adolescent (In) vulnerability.  

            American Psychologist, 48, 102-116.

 

Suan, L. V. & Tyler, J. D. (1990). Mental health values and preference for mental health

            resources of Japanese-American and Caucasian students. Professional Psychology:

            Research and Practice, 21, 291-296.

 

Takanishi, R. (1993). The opportunities of adolescence--research, interventions and

             policy.  American Psychologist, 48, 85-87.

 

U.S. Congress, Office of Technology Assessment (1991). Adolescent Health

            (OTA-H-468). Washington, DC; U.S. Government Printing Office.

 

Figure Caption

Figure 1. Preference ratings of resource for help options by grade and gender.