Doctor of Philosophy / Department of Horticulture, Forestry, and Recreation Resources / Richard H. Mattson / Subjects of this research were 64 adolescents receiving treatment at two behavioral health service institutions located in an urban mid-western city. Both institutions provided horticultural therapy and non-horticultural therapy programs. Research subjects were adolescents with diverse treatment needs and their responses on research questions were inconsistent compared to other related studies with general population. Current research assessed the adolescents with horticultural therapy treatment and without horticultural therapy treatment in three aspects. First, the levels of psychological aspects of adolescents were assessed with the Rosenberg Self-Esteem Scale and Nowicki-Strickland Locus of Control Scale for Children. The levels of self-esteem and locus of control of adolescents with horticultural therapy treatment were not significantly different from those of adolescents without horticultural therapy treatment at both institutions. Based on the different level of worthiness and competence factors, it is recommended to design horticultural therapy programs focused on improving the worthiness factor of self-esteem. Second, the pastoralism disposition of the Children's Environmental Response Inventory was used to assess the level of environmental attitude of the adolescents with and without horticultural therapy treatment. Horticultural experience and environmental attitude had a positive relationship with most subjects. At one institution, the level of environmental attitude of the adolescents with horticultural therapy treatment was significantly higher than the adolescents without horticultural therapy treatment. The adolescents at the horticultural therapy program which was scheduled more frequently showed higher environmental attitude scores. To improve environmental attitude of adolescents, horticultural therapy program should provide diversity and abundant opportunities of horticultural experiences. Third, basic horticultural knowledge was tested with the Basic Horticultural Knowledge Questionnaire. Vegetable/fruit consumption and preference were described with the Vegetable and Fruit Preference and Consumption Survey. Basic horticultural knowledge scores of the horticultural therapy group were significantly higher than that of the non-horticultural therapy group at one institution, but the scores were similar between the two groups at the other institution. Basic horticultural knowledge of subjects was significantly correlated to their vegetable and fruit consumption. To increase vegetable/fruit consumption, horticultural therapy programs should set goals to incorporate nutrition education.