Effectiveness of the Group-Level and Individual-Level Variables on Group Therapeutic Factors:A Longitudinal Study of Multilevel Analysis

[[abstract]]Effectiveness of the Group-Level and Individual-Level Variables on Group Therapeutic Factors: A Longitudinal Study of Multilevel Analysis Chun-shu Chen Abstract This study contains two parts: Part I of the study is to investigate members’ beneficial perception of the group therapeutic factors and examine the effects of variables of group-level and individual-level on group therapeutic factors at early, middle and latter group phase; Part II of the study is to examine growth pattern of beneficial perception of group therapeutic factors during group sessions and the effects of variables. The research subjects of this study were 242 members who participated in short-term structured group in a university. The leaders and members were assigned to 12 psycho-educational groups or 13 growth groups by using stratified randomization. Leaders in two types of group were the 2nd year graduate students majoring in guidance and counseling, and had taken eight hours of leader training course in the type of group they belong to. These short-term structured groups met once per week for a 90-minute session and nine times. The research tools included: “Interpersonal Attachment Inventory” filled by the members before they joining the group; “Group Therapeutic Factor Inventory—Short Form” filled immediately after the end of each session, “Group Climate Questionnaire – Short Form, GCQ-S (engagement scale)” and “Trainer Behavior Scale, TBS (four characteristic scale)” filled immediately after the 1st, 2nd, 5th, and 9th group session. Descriptive statistics and hierarchical linear models were used for data analysis. The main results were: 1. In terms of the average score of the beneficial perception for each factor category in the early, middle, and late phases, the factor of “positive feeling toward the groups” has the highest score; factors of “solicitation of action abilities” and “cognitive gains” ranks next; factor of “experience and understanding of family relationship” has the lowest score. The average score of factor of “universality” gets lower in the late phase while factor of “altruism” gets higher in the late phase. 2. In the full model, variables in group-level “group type”, “engaged climate”, “characteristic of leader” and variables in individual-level “preoccupied attachment”, “avoidant attachment”, “dismissive attachment” and “secure attachment” have a significant effect on most group therapeutic factors and beneficial perception for members in three group phases. There is no significant effect on factors “universality” and “experience and understanding of family relationship” in the early phase and factors of “universality” and “providing suggestion” in the middle phase. 3. “Engaged climate” in group-level has the most significant effect on members’ beneficial perception of group therapeutic factors in three phases. Secondly, variable “group type” has a significant effect on specific beneficial perception of group therapeutic factors for different group phases. “Secure attachment” is the most significant individual-level variable for prediction. The moderator effect of “engaged climate” between “avoidant attachment” and specific factors is the most significant in middle and late group phase while the moderator effect of “secure attachment” ranks the second. 4. According to the results of the growth model, the score of group therapeutic factors in each group session supports the cubic curve growth pattern of increasing, followed by decreasing or slowing down, then increasing again. The inflection point with more rapid changes caused by growth rate of factors which occurring between the 5th and 6th group sessions. 5. The variables in group-level and individual-level have a significant effect on the initial status of members’ beneficial perception of each group therapeutic factors; “group type” has an effect on the growth rate of most group therapeutic factors. Based on research results and limitations, several suggestions were proposed for future research and concern regarding the group counseling theory and practice, and educational training for leaders