Two counseling interventions to reduce teacher-child relationship stress.
The New Freedom Commission on Mental Health (2003) has recommended the promotion of screening, assessing, and providing services for the mental health of young children, in addition to the improvement and expansion of school mental health. The Commission also proposed the need for empirically based mental health interventions for children and adults. Public and private entities have emphasized the need for researchers and clinicians to demonstrate evidence of treatment effect prior to the dissemination of funding and/or support. The American School Counselor Association (ASCA) also emphasized the need for effectiveness research through the establishment of the National School Counseling Research Center (Lapan, 2005). Lapan further highlighted the weakness in school counseling research due to its history of not effectively addressing validity issues of school counseling treatments.
In the new era of evidenced-based practice, school counselors are pressured to respond with interventions that demonstrate effective change in individual and groups of students, school climate, academic progress, and/or behavioral maladjustment. However, conducting effectiveness research in the schools is problematic due to a limited ability to control certain variables such as variability of presenting problems, consistent location and time, and intervening factors affecting treatment. Owens and Murphy (2004) discussed the problems associated with effectiveness research in the schools but advocated the necessity and benefits of such research. They argued that school-based mental health programs provide a unique environment to explore evidence-based treatments in real-world settings (Owens & Murphy). The research presented in this article represents the joining of evidence-based mental health practice with effectiveness research in the school setting. This research was an attempt to study a well-established mental health practice for young children (child-centered play therapy) and a well-established intervention for teachers (consultation) to explore the impact of these interventions on teacher-child relationship stress in the real-world setting of elementary schools.
TEACHER-CHILD RELATIONSHIP STRESS
The relationship between teacher and student has been linked to the success and progress of students, both academically and personally. The teacher-child relationship is a critical determinant of a student's development, achievement, and classroom functioning, and it is affected by individual characteristics of both teacher and student, as well as dynamics of the relationship itself (Greene, Abidin, & Kmetz, 1997). Hence, behaviors exhibited by both teacher and student can contribute to a high- or low-quality relationship. One factor that contributes to the quality of the teacher-child relationship is the level of stress that exists between teacher and child (Abidin, Greene, & Konold, 2004). High levels of stress are linked to low relationship quality while low levels of stress contribute to a higher-quality relationship.
A large body of elementary school research has correlated negative teacher-child relationships with a continued trajectory of academic and behavioral problems (Hamre & Pianta, 2001; Ladd & Burgess, 2001; Pianta, Steinberg, & Rollins, 1995; Pianta & Stuhlman, 2004). For example, Ladd and Burgess found that sustained teacher-child conflict led to the progression and broadening of child maladjustment as children displayed increased misconduct and decreased academic competence. Research also has revealed that negativity in the teacher-child relationship, as early as kindergarten, predicts problematic academic and behavioral outcomes into middle school (Hamre & Pianta). Pace, Mullins, Beesley, Hill, and Carson (1999) found that childhood emotional and behavioral adjustment problems are associated with less favorable interpersonal relationships with teachers, as teachers rated students with adjustment problems as interpersonally less attractive. The researchers postulated that some children become depressed or behaviorally disturbed due to impairments in interpersonal relationships, providing a circular pattern of negative interactions.
The quality of the teacher-child relationship is impacted by the behavioral characteristics of children in the classroom, suggesting that children who exhibit less tolerable behaviors are often met with frustration by the teacher. Greene, Beszterczey, Katzenstein, Park, and Goring (2002) found that students with attention deficit hyperactivity disorder (ADHD) are rated by teachers as significantly more stressful to teach as compared to non-ADHD students. Further analysis and observation revealed that teacher stress was highly individualized according to the teacher-student dyad with outcome related to the characteristics of each person in each dyad, meaning that the relationship was of a reciprocal nature.
In surveying 415 teachers of students who had emotional and behavioral disorders, Nelson (2001) found that teachers felt less capable in working with students with externalizing and thought-disordered behaviors. He also reported that teachers who expressed confidence in working with externalizing students had lower levels of stress overall. And finally, Hughes, Cavell, and Jackson (1999) tracked levels of aggression in children from first to third grade. They found that the quality of teacher-student relationships predicted aggressive children's developmental trajectories, indicating that a positive teacher-student relationship helps to alleviate subsequent aggression in children.
In summary, teacher-child relationship research demonstrates that the teacher-child relationship has long-lasting effects on the success of the child in school. In addition, the quality of the relationship is individualized to each dyad and dependent on the individual characteristics of the particular teacher and the particular student involved. Problematic student behavior is somewhat stressful for all teachers but how a teacher responds to this behavior may impact the student's academic, psychological, and social development.
CHILD-CENTERED PLAY THERAPY
Due to its focus on the developmental needs of young children, play therapy is widely used as an intervention for children's emotional and behavioral problems. Play can reflect children's own experiences and perceptions of themselves, their relationships, and the world in general. In play therapy, toys are viewed as the child's words and play as the child's language (Landreth, 2002). Children more comfortably, safely, and meaningfully express their inner world through concrete, symbolic representation in play. Through toys or role-playing, children have the opportunity to develop a sense of control over their world as they reenact their experiences in the safety of the playroom.
Child-centered play therapy (CCPT) stemmed from the person-centered theory of Carl Rogers (1942) and was subsequently operationalized as nondirective play therapy by Virginia Axline (1947). In a survey of members of the Association of Play Therapy, most play therapists identified themselves as being trained in the child-centered theory of play therapy (Ryan, Gomory, & Lacasse, 2002). In an earlier survey, Phillips and Landreth (1995) found that play therapists identified CCPT and eclectic models as the most utilized forms of play therapy. The heavy representation of child-centered play therapists can probably be attributed to the formalization of this approach in the classic writings of Axline (1947, 1964), its elaboration in the work of Guerney (1983) and Landreth (2002), and the training available in this approach. The child-centered model offers a solid base for understanding children and how to work with them therapeutically.
Under the ASCA National Model[R] (2005), CCPT is categorized as a direct counseling intervention provided to children and falls under responsive services, one component of the delivery system recognized by ASCA as necessary for fully functioning school counseling programs. Research supporting the effectiveness of play therapy has been conducted extensively in the school setting (Fall, Balvanz, Johnson, & Nelson, 1999; Garza & Bratton, 2005; Packman & Bratton, 2003; Post, 1999; Shen, 2002). Children with learning differences have been positively affected by the use of CCPT as demonstrated through an increase in self-efficacy (Fall et al.) and a decrease in behavioral problems (Packman & Bratton). Children of diverse populations have been positively affected as demonstrated through decreased externalizing problems (Garza & Bratton, 2005), enhanced self-esteem and locus of control (Post), and decreased anxiety following a natural disaster (Shen).
Ray, Bratton, Rhine, and Jones (2001; further detailed in Bratton, Ray, Rhine, & Jones, 2005) conducted the largest meta-analysis on play therapy outcome research to date. This meta-analysis, reviewing 94 studies, statistically calculated an effect size of .80. This was interpreted as a large effect, indicating that children receiving play therapy interventions performed .80 standard deviations above children who did not receive play therapy. Play therapy interventions were found to be effective in increasing children's self-concepts, improving anxiety symptoms, improving social skills, and decreasing clinical behavioral problems (Bratton & Ray, 2000). To date, no literature was found on the impact of play therapy on teacher-child relationship stress.
Teacher consultation is a role performed by most school counselors to support teachers in dealing with individual student or classroom concerns. ASCA (2005) identified consultation as a role required in both delivery components of responsive services and system support. As a responsive service, consultation serves to help teachers respond more effectively with individual students. As a system support, teacher consultation serves the purpose of general support and sharing of system-wide information. Consultation involves three parties, two of whom are working together to benefit the client (Baker & Gerler, 2004). In the case of school counselors, consultation is the working relationship between counselor and teacher with the purpose of serving the needs of a student or group of students.
Teacher consultation is rarely emphasized in school counselor training or school counselor literature (Fall, 1995). Although school counseling textbooks offer models and definitions of consultation, school counselors may not be adequately trained in consultation skills, especially with teachers (Dustin, 1992). For the purposes of this research study, it was difficult to find recent research exploring teacher consultation or even theoretical literature on recent consultation models. In addition to lack of school counselor training in consultation, research may be lacking due to teacher resistance to consultation. Alderman and Gimpel (1996) found that teachers preferred to handle problems in the classroom without assistance from a consultant. In fact, teachers rated consulting with school counselors as their fourth choice of effective intervention in managing student problems, following the interventions of handling the problem on their own, consulting with the principal, and consulting with another teacher. However, teachers identified personal support with contact and listening from the consultant as the most helpful factor of consultation.
PURPOSE OF THE STUDY
The purpose of the present study was to explore the impact of two separate interventions on teacher-child relationship stress. Due to a presumption that the individual characteristics of each person involved in the teacher-child relationship affect the quality of that relationship, this study sought to explore the effect of an intervention planned directly for the child and the effect of an intervention planned directly for the teacher. The first intervention was CCPT designed in this study to impact the problem behaviors of identified children, thereby creating an effect on teacher-child relationship stress. The second intervention was consultation delivered directly to teachers designed to impact the support level of teachers, thereby reducing teacher-child relationship stress.
Because play therapy has been demonstrated to show a large significant positive effect when compared to a control group of no treatment (Bratton ct al., 2005) and because of the ethical dilemma of not serving referred children who need intervention, this study did not employ the use of a control group condition. Two main questions were of concern in this study: (a) Was there an overall change from mean pretest score to mean posttest score on teacher-child relationship stress after application of the interventions? (b) Was there change across time depending on which intervention the child received?
Participants were 93 students and 59 teachers from three elementary schools in the Southwestern United States. The researcher requested that teachers at each of the elementary schools identify students who were experiencing emotional and behavioral difficulties in the classroom. Teachers notified the school counselors of identified students. The school counselors obtained informed consent for participation in the study from parents and presented the informed consents to the researcher. All three schools served students from pre-kindergarten to fifth grade. All three are considered Tide 1 schools targeted by the state for school-wide assistance due to high percentages of children qualifying for free or reduced lunch.
School 1 listed 60.7% of its population as economically disadvantaged; School 2 listed 67.6% of its population as disadvantaged; and School 3 listed 40.2% as disadvantaged. Ethnicity breakdowns for each of the schools are listed as follows: School 1--African American (9.6%), Hispanic (39.8%), Caucasian (49.6%), Native American (.5%), and Asian (.5%); School 2--African American (14.7%), Hispanic (55.7%), Caucasian (28.1%), Native American (.4%), and Asian (1.1%); School 3--African American (10.7%), Hispanic (28.8%), Caucasian (58.3%), Native American (.6%), and Asian (1.7%). Due to playroom space and availability, 43 children were selected for the study from School 1, 30 children were selected from School 2, and 20 children were selected from School 3. Overall, 68 males and 25 females participated in the study.
Children at each school were randomly assigned to one of three treatment groups. Of the males, 24 were assigned to the play therapy-only (PT) treatment condition, 23 were assigned to the play therapy and consultation (PTC) treatment condition, and 21 were assigned to the consultation-only (CO) condition. Of the females, 8 were assigned to the PT group, 9 were assigned to the PTC group, and 8 were assigned to the CO group. Age distribution was as follows: 4-year-olds, 5; 5-year-olds, 19; 6-year-olds, 17; 7-year-olds, 13; 8-year-olds, 20; 9-year-olds, 9; 10-year-olds, 8; and 11-year-olds, 2. Grade-level distribution for the whole study and treatment group was as follows: 5 pre-kindergarten (1 PT, 3 PTC, 1 CO); 24 kindergarten (7 PT, 9 PTC, 8 CO); 18 first grade (11 PT, 1 PTC, 6 CO); 16 second grade (3 PT, 6 PTC, 7 CO); 15 third grade (6 PT, 7 PTC, 2 CO); 9 fourth grade (2 PT, 3 PTC, 4 CO); and 6 fifth grade (2 PT, 3 PTC, 1 CO). Ethnicity breakdowns were as follows: 12 African American (6 PT, 4 PTC, 2 CO); 38 Hispanic (9 PT, 13 PTC, 16 CO); 39 Caucasian (15 PT, 14 PTC, 10 CO); and 4 biracial (2 PT, 1 PTC, 1 CO).
Index of Teaching Stress. The purpose of the Index of Teaching Stress (ITS; Abidin et al., 2004) is to measure stress that a teacher experiences in the relationship with a specific student. The ITS is based on the belief that the relationship between a teacher and student is primary to the academic and personal success of the student. The ITS assesses the independent factors that correlate highly with the quality of the teacher-child relationship, including behavioral characteristics of the student, the teacher's perception of the teaching process, and the teacher's perception of support from others who interact with the child. The ITS includes 90 Likert-scale items and is standardized for use with teachers of students in preschool through 12th grade.
The ITS produces a Total Stress score and three domain scores, consisting of ADHD, student characteristics, and teacher characteristics. The ADHD domain measures the teacher's stress level associated with the child's behaviors that are commonly associated with ADHD. The Student Characteristics domain measures the teacher's stress related to the student's temperament and behaviors. The Teacher Characteristics domain measures the teacher's stress as related to self-perception and expectation regarding teaching the particular student. The Total Stress score is a sum of the three domain scores (Abidin et al., 2004).
The ADHD domain consists of 16 items that factor together to form a unique source of relationship stress. The Student Characteristics domain is furthered divided into four areas specific to the teacher's response to student behavior, including the student's level of emotional lability/low adaptability, anxiety/withdrawal, low ability/learning disability, and aggressive/conduct disorder. The Teacher Characteristics domain is also made up of four scales including the teacher's sense of competence/need for support, loss of satisfaction from teaching, disruption of the teaching process, and frustration working with parents.
Abidin et al. (2004) established concurrent and discriminant validity for ITS scores through multiple research studies using the ITS in the areas of teacher stress, ADHD, teacher judgment, teacher health, teacher gender, teacher behavior, and correlational studies with social skills and behavioral checklists. All alpha coefficients for the ITS domain scores and Total Stress scores exceeded .90 for the normative group of the ITS. Because ITS is a relatively new instrument, test-retest reliability is still in question. Abidin et al. reported on one test-retest reliability study in which 42 teachers rated children identified with behavioral problems. Test-retest reliability coefficients for the ITS domains and Total Stress score were reported as .58 for the ADHD domain, .57 for the Student Characteristics domain, .70 for the Teacher Characteristics domain, and .65 for Total Stress.
Demographic data. The teachers completed demographic data on each student. Teachers indicated the student's age, grade, ethnicity, and school. Teachers also reported the number of years that they had been teaching.
Consultation questionnaire. Teachers who participated in consultation were given a post-consultation questionnaire that included three questions. They were asked to share the positives aspects of consultation, the negative aspects of consultation, and to rate the helpfulness of consultation. The third question was a Likert-scale question that asked, "On a scale of 1-5 (with 5 being most helpful), how would you rate your consultation experience?"
Upon receiving the informed consent from each student's parent, the researcher randomly assigned each child to one of three treatment groups. All participants were matched according to school and grade level, then with the use of a table of random numbers, children were assigned to a treatment group (Shadish, Cook, & Campbell, 2002). Each teacher provided informed consent due to their completion of the ITS. One week prior to the beginning of treatment, teachers completed an ITS and demographic form on each student. Students and teachers were scheduled to participate in 8 weeks of treatment. Due to typical school scheduling difficulties such as field trips, standardized testing, and occasional absences, the study was completed in 10 weeks. At the end of 10 weeks, each teacher completed an ITS on each participating student as a postmeasure. Also at posttest administration, teachers who participated in consultation were asked to complete the consultation questionnaire. The following paragraphs describe each treatment condition.
Play therapy only. Thirty-two students were assigned to the PT-only group, which consisted of 16 sessions of play therapy over 8 weeks. Each student received two sessions per week of 30-minute individual CCPT sessions. All play therapists had successfully completed at least two courses in play therapy, and they participated in direct individual or triadic supervision with a counseling faculty member certified in play therapy. Play therapists included two counseling faculty members, seven doctoral-level counseling students, and one advanced master's student. Play therapists were required to review their videotaped play therapy sessions with their supervisors on a weekly basis. Supervisors ensured that the CCPT protocol was being followed and enacted in the play sessions through the use of the Play Therapy Skills Checklist (PTSC; Ray, 2004). Supervisors rated responses on the PTSC confirming that each response fell into a CCPT category. Play therapists were not allowed to discuss the student with the teacher during the study.
Play therapy sessions were conducted in specially equipped playrooms in each of the school settings. Playrooms were equipped with a variety of specific toys to facilitate a broad range of expression, following Landreth's (2002) general guidelines. CCPT is designed to provide specific therapist responses to the child during play therapy. These response sets are clarified in detail in Landreth and in Ray (2004), and both include nonverbal skills and verbal skills. CCPT nonverbal skills include the counselor leaning forward toward the child and being physically directed toward the child at all times. When responding to a child, the counselor's tone is congruent with the child by matching the level of affect displayed by the child. The skill of matching verbal response with nonverbal response is representative of the counselor's level of genuineness with the child. CCPT verbal responses are structured to help facilitate growth in the child. They include the following response categories: (a) tracking behavior, (b) reflecting content, (c) reflecting feeling, (d) facilitating decision-making/returning responsibility, (e) facilitating creativity/spontaneity, (f) esteem building/encouraging, (g) facilitating relationship, and (h) limit setting.
Consultation only. Twenty-nine students were assigned to the CO group, which consisted of eight consultation sessions with a counselor consultant. Each teacher of the assigned students received 10 minutes of person-centered consulting per week over the 8 weeks, totaling eight sessions. All consultants were doctoral-level counselors who were in their second or third year of a counselor education doctoral program. All consultants had completed a Council for Accreditation of Counseling and Related Educational Programs master's program or met equivalencies; two consultants were former school counselors; and one consultant was a former school psychologist. Consultants and play therapists were not the same people, hence consultants were not informed regarding the student in discussion. Students assigned to CO received no direct intervention and were placed on a waiting list to receive play therapy following the study.
The particular model of consultation used for this study was based on previous research and the researcher's experience as a school counselor. A short time period of 10 minutes was chosen due to restraints in the school schedule. It was not deemed feasible to require teachers to participate in 30 minutes of consultation each week but 10 minutes could fit in at lunchtime, during a planning period, or even at recess. The purpose was to offer support to the teachers but not overwhelm them with excessive time requirements.
A person-centered consultation model was chosen for two reasons. The first reason was that person-centered consulting was philosophically aligned with CCPT, which provided a consistent framework for the study. Secondly, Horton and Brown (1990) branded a person-centered approach to consulting as "consultee-centered consultation," which focuses on improving the consultee's ability to deal with all clients, not just the client of focus. Furthermore, consultation research has demonstrated the importance of facilitative characteristics such as empathy, congruence, and unconditional positive regard (Horton & Brown). Further support for a person-centered consultation model was provided by Alderman and Gimpel (1996), who found that teachers identified personal support through contact and listening as the most helpful factor of consultation.
Consultants for this study were educated in the person-centered model through intense training in person-centered counseling and CCPT. They also attended training on person-centered consultation that focused on content reflection, feeling reflection, encouragement, confrontation, and enlarging the meaning skills. Each consultant presented the following statement to the teachers at the beginning of the consultation relationship:
We'll be meeting for 10 minutes each week. The purpose of this meeting is to provide support for you in whatever way you need. You can choose to talk about the particular student referred for play therapy or any other issues that concern you. I will always begin our conversation with "How is it going with (child's name)?" but you can choose to discuss any issues of concern for you. Our conversation is confidential and will not be shared in whole or in part with any school staff.
Play therapy and consultation. Thirty-two students were assigned to the PTC treatment group. Students assigned to this group received two sessions per week of 30-minute CCPT sessions totaling 16 sessions. Teachers of students assigned to this group received 10-minute consultation sessions each week for 8 weeks totaling eight sessions.
Data analysis was intended to answer research questions using a combined between-within-subjects analysis of variance (i.e., split-plot analysis; Tabachnick & Fidell, 2001). This repeated measures approach was used to demonstrate significant change over time and between groups. In the analysis, the student treatment group (k = 3) served as the between-subjects variable and the time (k = 2) from pretest to posttest served as the within-subjects variable. Analyses were run separately with the ITS Total Stress score, the ADHD domain, the Student Characteristics domain, and the Teacher Characteristics domain as dependent variables. In each combined between-within-subjects analysis of variance (ANOVA), the required assumption of sphericity was assumed given that there were only two points of measurement. Even though random assignment was employed following a school and grade matching procedure, pretest differences appeared to demonstrate the possibility that groups were not equal at their starting point (the PTC group appeared more clinical than the PT and CO groups).
Four separate analysis of covariance (ANCOVA) procedures to control for pretest differences were employed using the posttest Total Stress score, ADHD domain, Student Characteristics domain, and Teacher Characteristics domain as dependent variables and the pretest domain scores as covariates. A multivariate analysis of covariance was not chosen because the data violated the assumption of equality of covariance matrices and normality and because literature did not support a theoretical rationale for combining variables into a multivariate analysis (Henson, 1999). Furthermore, the use of these scale scores will likely occur in univariate fashion, that is, researchers and practitioners often will focus on one scale at a time in their work.
Because each of the employed analyses had its limitations--ANOVA limited by regression to the mean and ANCOVA limited by lack of reliability in change scores--it was decided that a multiple analyses approach would serve to present the data most accurately. Due to the use of multiple analyses, a Bonferroni correction for risk of familywise Type I error was used by dividing the typical .05 alpha level by the number of analyses (8), resulting in a more conservative alpha of .01. It was decided that if statistically significant differences were found with a meaningful effect size between groups, simple main-effects post-hoe analyses would be conducted as needed.
Treatment Group Results
ITS Total Stress. Table 1 presents the ITS means, standard deviations, and sample sizes on the pretest and posttest for all three treatment groups. Results of the ANOVA on Total Stress revealed a statistically significant main effect for time, F(1, 90) = 20.15, p < .01 (partial [[eta].sup.2] = .18); no statistically significant main effect for group, F(2, 90) = .40, p = .67 (partial [[eta].sup.2] = .01); and no statistically significant interaction effect, F(2, 90) = .68, p = .51 (partial [[eta].sup.2] = .02) (see Table 2). Figure 1 graphically displays the main effect for treatment differences and the main effect for time, which indicates significant decline in Total Stress across all three treatment conditions. Because the change was fairly consistent across all three groups, the interaction effect was negligible. The effect size of .18 for change over time indicates a large effect size according to Cohen's (1988) guidelines (see also Henson, in press). There was no significant difference between groups; hence no further simple effects analysis was necessary.
[FIGURE 1 OMITTED]
ITS ADHD domain. Results of the ANOVA on the ADHD domain revealed a statistically significant main effect for time, F(1, 90) = 24.27, p < .01 (partial [[eta].sup.2] = .21); no statistically significant main effect for group, F(2, 90) = .30, p = .74 (partial [[eta].sup.2] = .01); and no statistical significance for interaction effect, F(2, 90) = 2.67, p = .08 (partial [[eta].sup.2] = .06). Again, the effect size for main effect for time (.21) was in the large category.
ITS Student Characteristics domain. Results of the ANOVA on the Student Characteristics domain revealed a statistically significant main effect for time, F(1, 90) = 17.97, p < .01 (partial [[eta].sup.2] = .17); no statistically significant main effect for group, F(2, 90) = 1.82, p = .17 (partial [[eta].sup.2] = .04); and no statistical significance for interaction effect, F(2, 90) = 1.00, p = .37 (partial [[eta].sup.2] = .02). The effect size for main effect for time (.17) was in the large category.
ITS Teacher Characteristics domain. Results of the ANOVA for the Teacher Characteristics domain revealed a statistically significant main effect for time, F(1, 90) = 11.41, p = .01 (partial [[eta].sup.2] = .11); no statistically significant main effect for group, F(2, 90) = .42, p = .66 (partial [[eta].sup.2] = .01); and no statistically significant interaction effect, F(2, 90) = .23, p = .79 (partial [[eta].sup.2] = .01). The effect size for main effect for time (.13) was in the moderate category.
ANCOVA Domain Analyses
A one-way between-groups analysis of covariance was conducted for all ITS domains to compare the effectiveness of the three interventions designed to reduce teacher stress while controlling for pretest differences. The independent variable was the type of intervention (PT, CO, or PTC), and the dependent variables were the posttest scores on Total Stress, ADHD, Student Characteristics, and Teacher Characteristics after treatment. Participants' scores on the domains at pretest were used as the covariates in these analyses. Preliminary checks were conducted to ensure that data met the assumptions of normality, linearity, homogeneity of variance, homogeneity of regression slopes, and reliable measurement of the covariate. All of these analyses found no significant differences between the treatment groups at posttest.
Several supplemental analyses were conducted to investigate the effect of key variables on treatment outcome. A one-way between-groups ANCOVA was conducted to compare the effect of student gender on reduction of teacher stress as measured by the ITS Total Stress. The independent variable was the gender of the student and the dependent variable consisted of posttest scores on the ITS Total Stress. Participants' scores on the pretest administration of the ITS Total Stress were used as the covariate in this analysis. Preliminary checks were conducted to ensure that there was no violation of the assumptions of normality, linearity, homogeneity of variances, homogeneity of regression slopes, and reliable measurement of the covariate. After adjusting for pretest scores, there was no statistically significant difference between females and males on posttest Total Stress scores and the effect was small (F [1, 90] = .56, p = .46 [partial [[eta].sup.2] < .01]).
An additional one-way between-group ANCOVA was conducted to compare the effect of ethnicity on reduction of teacher stress as measured by the ITS Total Stress score. The independent variable was the ethnicity of student (African American, Caucasian, Hispanic/Latino, biracial) and the dependent variable consisted of posttest scores on the ITS Total Stress. Participants' scores on the pretest administration of the ITS Total Stress were used as the covariate in this analysis. After adjusting for pretest scores, there was not a statistically significant difference among the four identified ethnicities on posttest Total Stress scores and the effect was moderate (F[3, 88] = 1.81, p = .15 [partial [[eta].sup.2] = .06]).
The relationship between Total Stress posttest scores and years of teaching was investigated using a Pearson product-moment correlation coefficient. Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity, and homoscedasticity. There was virtually no correlation between the two variables (r = .04, n = 92, p = .71). Additionally, the relationship between Total Stress post-test scores and age of student was correlated using a Pearson product-moment correlation coefficient, and no relationship was found between the two variables (r = -.06, n = 93, p = .57).
Clinical significance refers to the practical value of an intervention when applied to the everyday life of the client (Kazdin, 2003). Clinical significance is not relevant to the comparison of groups in response to an intervention, that is, presence of a control group (Kazdin, 1999). Rather, clinical significance demonstrates whether clients who receive counseling intervention move toward healthier functioning. This type of evaluation is especially helpful when applied to the present study due to a lack of a control group and the finding of no statistically significant differences between groups. One method of addressing clinical significance is using a comparison method in which client performance is evaluated in relationship to normative samples at the end of treatment (Kazdin, 2003).
With the use of clinical cutoff scores identified by the ITS, participants who scored at clinical levels at pretest on the ADHD, Student Characteristics, Teacher Characteristics, and Total Stress were tracked for progress to determine movement toward nonclinical scores at posttest. On the ADHD domain, teachers identified 22 students (PT = 6; CO = 8; PTC = 8) at or above the clinical functioning level at pretest compared to 11 students (PT = 5; CO = 3; PTC = 3) identified at clinical functioning levels at posttest (50% reduction in clinical stress levels for ADHD). On the Student Characteristics domain, teachers identified 13 students (PT = 5; CO = 2; PTC = 6) at or above clinical functioning level at pretest compared to 6 students (PT = 2; CO = 1; PTC = 3) identified at clinical functioning levels at posttest (54% reduction in clinical stress levels for Student Characteristics). On the Teacher Characteristics domain, only four teachers scored at the clinical functioning level (PT = 2; CO = 2; PTC = 0). No changes were noted at posttest. Seven students were scored at the clinical functioning level on the Total Stress score (PT = 3; CO = 2; PTC = 2). At posttest, five students maintained a clinical level of functioning (PT = 2; CO = 2; PTC = 1), marking a 23% reduction in clinical functioning for Total Stress.
Subjective Results of Consultation
Of the 33 teachers participating in consultation, 27 returned the consultation questionnaire. Of the 27 who returned the questionnaire, 3 rated consultation as a 1 (least helpful), 3 rated consultation as a 2, 1 rated consultation as a 3, 8 rated consultation as a 4, and 9 rated consultation as a 5 (most helpful). Three respondents did not rate consultation on the scale. The mean for all respondents was 3.7. However, as can be seen by the scale distribution, there were few neutral responses toward consultation. The teachers appeared to feel favorably toward consultations or to feel negatively toward consultations.
Negative remarks centered mostly on time issues. One negative comment included "making time to do it because I'm very busy." Another comment: "If I don't have anything to talk about, then I don't feel we should meet just for the sake of meeting. I have too many other obligations to even waste 10 minutes of my 'free' time." Another negative comment was that "I had to talk to her during my lunch time." Other negative remarks appeared to focus on the consultation person-centered model, such as, "I wish the consultation could have been more about my students and their progress in play therapy. Or I would have liked a more structured agenda rather than 'How's your week?'" Another comment: "I feel like there was never anything new to say about my student. It was always same ole, same ole." Another teacher commented, "The negative is that she just listened and I suppose that was her intended role. I found myself rambling a lot and I didn't feel like I had any help resolving situations that were going on."
Overall, responses toward consultation were more positive than negative, such as, "It's good to have someone to listen without bias." Another comment: "[The sessions] were always positive and gave me a chance to 'vent' and by hearing myself I could work through some of the struggles of working with my special kids." Another teacher commented, "At first it was a little awkward, but I really found it helpful. I thought being able to talk through my own thoughts freely was great. It seemed to ease a lot of stress after each session." Other positive comments included the following: "It was nice to have someone to talk to who was just there to listen. Discussing the child helped sort out feelings I was having toward their behaviors." "A positive thing of the consultation was how the consultant could help me turn my 'bad' day around to just relaxing and taking the day." "I had a chance to reflect on the student's progress or lack of. Week after week I felt better about how I responded to the behaviors of the student."
Results of this study indicate that play therapy and consultation are possible effective interventions in reducing teacher-child relationship stress. The three treatment groups, PT, CO, and PTC, all yielded statistically significant reductions in teacher stress response to student characteristics, student behaviors associated with ADHD, and teacher characteristics. The effect size for three of the scores, Total Stress, ADHD, and Student Characteristics, demonstrated a large effect indicating the practical significance of the findings (Henson, in press; Thompson, 2002). Play therapy and consultation, alone and combined, appeared to positively impact the teacher-child relationship. The absence of a statistically significant difference between groups signifies that the treatments had an equal effect on teacher-child relationship stress. However, statistical differences were noted across time, which could indicate that the passage of time alone was a possible variable in determining significant change. An alternate likely explanation for the change over time is the effectiveness of the interventions, a supposition that is based on previous literature that a play therapy intervention yields a large positive effect compared to no intervention (Bratton et al., 2005).
As hypothesized in the design of the study, results revealed that the play therapy intervention had a significant impact on reducing teacher stress as affected by student characteristics and ADHD behaviors. Historically, play therapy research supports the use of play therapy to reduce problem behaviors demonstrated by children. Over the past decade and a half, since 1990, 36 (27 published) research studies on the impact of play therapy have been conducted. These most recent studies have demonstrated the positive impact of play therapy on general behavioral problems (Raman & Kapur, 1999; Shashi, Kapur, & Subbakrishna, 1999); externalizing behavioral problems (Flahive, 2005; Garza & Bratton, 2005; Karcher & Lewis, 2002; Kot, Landreth, & Giordano, 1998; Schumann, 2005); internalizing problems (Packman & Bratton, 2003); self-efficacy (Fall et al., 1999); self-concept (Kot et al.; Post, 1999); anxiety (Baggerly, 2004; Shen, 2002); depression (Baggerly); speech problems (Danger & Landreth, 2005); and diabetes treatment compliance (Jones & Landreth, 2002).
Students who demonstrate social and aggressive problem behaviors have been found to create more stress among teachers (Greene et al., 2002). Interventions, like CCPT, that seek to reduce behavioral problems in students will likely reduce stress in the teacher-child relationship. The findings of this study support the supposition that by directing CCPT only to the child in an attempt to reduce behavioral maladjustment, teacher stress is reduced as teachers' perceptions of student problem characteristics are reduced.
A practical question that arises from the results of this study is the question of direct intervention for children. Why would school counselors engage in multiple sessions of play therapy that require specialized training and materials if regular short-format teacher consultations are just as effective? The answer to this question lies in the dependent measure that was used for this study. This project investigated the level of teacher-child relationship stress as determined by the reported stress from the teacher. Hence, the teacher's level of stress was the determinant of effectiveness of the three interventions. The child's perspective, parent's perspective of child behavior, or specific child behaviors were not explored by this study. This study did not address whether the child improved behavioral problems, internalizing problems, or emotional states, but it addressed the teacher's response to the child. As presented earlier, the ITS is based on the theory that when teacher characteristics or student characteristics are positively impacted, the teacher-child relationship will improve. Results of this study indicate that an intervention addressing student characteristics, such as play therapy, as well as an intervention addressing teacher characteristics, such as consultation, will help reduce teacher-child relationship stress. The direct effect of play therapy on the behavior problems of children displayed in schools is a separate issue that would need to be explored through multiple dependent measures addressing student needs, not teacher-child relationship stress.
The Teacher Characteristics domain was found to be significantly reduced through interventions of both play therapy and consultation, although to moderate practical significance. This domain addresses the sense of hopelessness and ineffectiveness that a teacher experiences in working with a particular student (Abidin et al., 2004). Ironically, this domain was affected by both a teacher intervention and a student intervention. Because the PT condition allowed virtually no contact with the teacher regarding the student, it might seem odd that a teacher's sense of effectiveness under this condition would be buoyed. One possible explanation for the significant reduction in Teacher Characteristics is that through the provision of any type of intervention, a teacher is bolstered with a sense of hope for the improvement of the situation. However, the lack of teacher involvement in the process of dealing directly with the child (a characteristic of both the PT and CO designs) might have limited the effect size of the interventions.
The teachers' dichotomous reaction to consultation was a unique finding in this study. Teachers who participated in the consultation model held few neutral opinions about the process. Following the study, consultants shared their overall impressions of their consultation experiences. Because several of the consultants had previously worked in the school environment, they felt strongly that a person-centered approach to consultation was the most effective method of helping teachers. However, they also were apprehensive about approaching teachers to maintain a consultation schedule, knowing the stressful schedule of most teachers. By the end of this study, the consultants agreed that the consultation model was successful beyond their initial expectations. They developed strong supportive relationships with their consultees and felt they had delivered a service of benefit to the teachers.
As presented earlier, the majority of teachers felt positively about the consultation experience including the consistent schedule and the person-centered approach. Yet, there were a few teachers who were openly negative toward the consultation experience and toward their consultant. Consultants reported that some teachers would plan other activities during their scheduled times and would assertively complain about having to reschedule. One teacher refused to talk on two occasions and just sat quietly with the consultant. It should be noted that all teachers were informed of the consultation process both verbally and in the written informed consent that they signed prior to participation in the study. This type of response was puzzling to the researcher and to the consultants. Although several scenarios were hypothesized--such as the possibility that these teachers felt threatened by the nature of the consulting relationship, were dealing with their own personal issues that they wished not to share, or were just too stressed to see the consultation process as helpful--the nature of the resistance is still largely unexplained. Nevertheless, results indicated that consultation had a beneficial effect and that most teachers felt positively toward their experience. Yet, school counselors should be cautioned that consultation is obviously not the most effective intervention for all teachers.
Several limitations were noted in this study. The most obvious limitation is the lack of a no-treatment control group. Although this was a limitation of the design, findings further exacerbate this limitation by demonstrating no significant differences between groups. The possibility exists that children demonstrated positive significant change based on time alone. However, other studies indicate that with no intervention, problem relationships with teachers actually increase over time (Hamre & Pianta, 2001; Pianta et al., 1995; Pianta & Stuhlman, 2004). The likelihood that a control group with no intervention would have improved over time is weak, but possible. Another problem in the research design was the lack of time equality across treatments; children who participated in play therapy sessions participated in 16 30-minute sessions while teachers who participated in consultation participated in 8 10-minute sessions. The researcher decided to accept this limitation as part of working in the real-world environment of the school. Requiring teachers to participate in 16 30-minute sessions seemed prohibitive for research success and later replicability of the design. The result of this limitation is the supposition that if teachers had received as much consultation as the play therapy that the students received, the consultation intervention may have resulted in stronger effects.
Another limitation of this study was the use of only one self-report measure. Lapan (2005) has encouraged school researchers to use multiple measures to exhibit good research design. The limitation in using only one measure was a restricted ability to generalize to a broad understanding and definition of relationship stress. The ITS is a self-report measure that is influenced by the perception of the teacher and is perhaps not indicative of behavior that can be measured through objective observers or raters. Abidin (1992) argued that behavioral observations are not adequate in measuring the belief systems of caretakers on children and that self-report measures are more practical to this end.
Implications for Research
The present study represents the implementation of outcome effectiveness research in the real-world setting of elementary schools. Exploration of the effect of treatment interventions on students exhibiting behavioral and emotional problems in early school years demonstrates that, indeed, the field of school counseling has the ability to continue empirical examination of common counseling practices. However, addressing the difficulty of conducting large-scaled research in schools, this study involved the services of 14 play therapists and counselor consultants, which speaks to the need for collaborative partners in research. A school counselor alone would have difficulty enacting such a complex research design with the robust number of students needed for generalization purposes. Partnerships between universities and schools allow the maximal benefit and advantageous use of resources.
Results of this research inspire new ideas for the study of various factors and variables related to the current study outcome. The addition of a no-treatment condition to the replication of the research design would allow researchers to control for the main effect of time. Investigating interventions directly delivered to children or teachers, instead of both, would clarify the effects of intervention on that particular entity. For example, comparison groups of a play therapy condition and a cognitive behavioral technique condition delivered only to students would demonstrate the effect of a student-only intervention on teacher-child relationship stress. Likewise, comparison groups of a person-centered consultation model condition and a solution-focused model condition delivered only to teachers would demonstrate the effect of a teacher-only intervention on relationship stress. Another likely addition to the presented research design is the administration of an objective behavioral measure to compare teacher stress response to child behavior versus actual child behavior. This type of design would establish real versus perceived change in student behavior and its effect on the relationship.
Implications for Practice
Perhaps the most significant outcome of this study was the presentation of two useful interventions to help reduce teacher-child relationship stress. A principal intention of this study was to offer school counselors a research protocol that is replicable in elementary school settings by elementary school counselors. Both interventions of 10-minute teacher consultation and 30 minutes of play therapy are feasible methods that are likely to be delivered by a school counselor. Most school counselors are trained in the area of delivering person-centered counseling to clients. School counseling training programs across the country offer the skill set of providing empathy, unconditional positive regard, and genuineness through the verbal technique of reflection and nonverbal techniques that provide a feeling of acceptance. The results of this research demonstrate that if school counselors use these skills in an intensive one-on-one relationship with teachers on a consistent basis, positive changes will likely take place in the teacher-child relationship. Possibly, when teachers feel understood and supported, they are better able to provide these conditions for their students--a positive parallel process.
Although elementary school counselors report that they are not adequately trained in play therapy, they appear to recognize and support the rationale for offering play therapy to students (Ray, Armstrong, Warren, & Balkin, 2005). The results of this study offer CCPT as a viable option in decreasing stress between teacher and child. CCPT in increments of 30 minutes twice a week is a responsive service that a play therapy--trained school counselor can provide as relief to both the student and the teacher. However, school counselors will need to overcome barriers of space, resources, training, and time allotment. According to Ray et al., school counselors do not identify administrative support as a barrier to play therapy, but they do struggle with the aforementioned obstacles. Results confirm that if the school counselor can overcome these difficulties to provide CCPT to individual children, there can be substantial positive outcomes for the relationship between teacher and child.
The relationship between student and teacher is critical to the academic and personal success of students. Research on teacher-child relationships reveals that early negative relationships with elementary school teachers sets a trajectory for children that has long-lasting impact on their future. School counselors, through their counseling training, hold the skills to build relationships that are collaborative, supportive, and nurturing. By offering responsive services such as play therapy and teacher consultation, a school counselor can help to intervene in stressful academic relationships to increase the quality of the relationship dynamic. Reduction of teacher-child relationship stress allows teachers to be released from destructive interactions with children exhibiting behavioral problems so that learning can take place. The current study lends credibility to the interventions of play therapy and consultation so that school counselors can use these as options in the improvement of teacher-child relationships.
Abidin, R. (1992). The determinants of parent behavior. Journal of Clinical Child Psychology, 21, 407-412.
Abidin, R., Greene, R., & Konold, T. (2004). Index of Teacher Stress: Professional manual. Lutz, FL: Psychological Assessment Resources.
Alderman, G., & Gimpel, G. (1996). The interaction between type of behavior problem and type of consultant: Teachers' preferences for professional assistance. Journal of Educational and Psychological Consultation, 7, 305-313.
American School Counselor Association. (2005). The ASCA national model: A framework for school counseling programs (2nd ed.). Alexandria, VA: Author.
Axline, V. (1947). Play therapy. New York: Ballantine Books.
Axline, V. (1964). Dibs: In search of self. New York: Ballantine Books.
Baggerly, J. (2004). The effects of child-centered group play therapy on self-concept, depression, and anxiety of children who are homeless. International Journal of Play Therapy, 13, 31-51.
Baker, S., & Gerler, E. (2004). School counseling for the twenty-first century (4th ed.). Upper Saddle River, NJ: Pearson. Bratton, S., & Ray, D. (2000).What the research shows about play therapy. International Journal of Play Therapy, 9, 47-88.
Bratton, S., Ray, D., Rhine, T, & Jones, L. (2005). The efficacy of play therapy with children: A meta-analytic review of treatment outcomes. Professional Psychology: Research and Practice, 36, 376-390.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillside, NJ: Lawrence Earlbaum Associates.
Danger, S., & Landreth, G. (2005). Child-centered group play therapy with children with speech difficulties. International Journal of Play Therapy, 14, 81-102.
Dustin, D. (1992). School consultation in the 1990s. Elementary School Guidance & Counseling, 26, 65-76.
Fall, M. (1995). Planning for consultation: An aid for the elementary school counselor. School Counselor, 43, 151-157.
Fall, M., Balvanz, J., Johnson, L., & Nelson, L. (1999). A play therapy intervention and its relationship to self-efficacy and learning behaviors. Professional School Counseling, 2, 194-204.
Flahive, M. (2005). Group sandtray therapy at school with preadolescents identified with behavioral difficulties. Unpublished doctoral dissertation, University of North Texas.
Garza, Y., & Bratton, S. (2005). School-based child-centered play therapy with Hispanic children: Outcomes and cultural considerations. International Journal of Play Therapy, 14, 51-79.
Greene, R., Abidin, R., & Kmetz, C. (1997). The Index of Teaching Stress: A measure of student-teacher compatibility. Journal of School Psychology, 35, 239-259.
Greene, R., Beszterczey, S., Katzenstein,T., Park, K., & Goring, J. (2002). Are students with ADHD more stressful to teach? Journal of Emotional and Behavioral Disorders, 2, 79-90.
Guerney, L. (1983). Play therapy with learning disabled children. In C. E. Schaefer & K. L. O'Connor (Eds.), Handbook of play therapy (pp. 419-435). New York: John Wiley & Sons.
Hamre, B., & Pianta, R. (2001). Early teacher-child relationships and the trajectory of children's school outcomes through eighth grade. Child Development, 72, 625-638.
Henson, R. K. (1999). Multivariate normality: What is it and how is it assessed? In B. Thompson (Ed.), Advances in social science methodology (Vol. 5, pp. 193-211). Stamford, CT: JAI Press.
Henson, R. K. (in press). Effect size measures and meta-analytic thinking in counseling psychology research. The Counseling Psychologist.
Horton, G. E., & Brown, D. (1990).The importance of interpersonal skills in consultee-centered consultation: A review. Journal of Counseling and Development, 68, 423-426.
Hughes, J., Cavell, T., & Jackson, T. (1999). Influence of the teacher-student relationship on childhood conduct problems: A prospective study. Journal of Clinical Child Psychology, 28, 173-184.
Jones, E., & Landreth, G. (2002). The efficacy of intensive individual play therapy for chronically ill children. International Journal of Play Therapy, 11, 117-140.
Karcher, M.,& Lewis, S. (2002). Pair counseling: The effects of a dyadic developmental play therapy on interpersonal understanding and externalizing behaviors. International Journal of Play Therapy, 11, 19-42.
Kazdin, A. (1999). The meanings and measurement of clinical significance. Journal of Consulting and Clinical Psychology, 67, 332-339.
Kazdin, A. (2003). Clinical significance: Measuring whether interventions make a difference. In A. Kazdin (Ed.), Methodological issues & strategies in clinical research (3rd ed., pp. 691-710). Washington, DC: American Psychological Association.
Kot, S., Landreth, G., & Giordano, M. (1998). Intensive child-centered play therapy with child witnesses of domestic violence. International Journal of Play Therapy, 7, 17-36.
Ladd, G., & Burgess, K. (2001). Do relational risks and protective factors moderate the linkages between childhood aggression and early psychological and school adjustment? Child Development, 72, 1579-1601.
Landreth, G. (2002). Play therapy: The art of the relationship (2nd ed.). New York: Brunner-Routledge.
Lapan, R. (2005). An editor's top ten wish list. Professional School Counseling, 8, ii-iv.
Nelson, J. (2001). Sources of occupational stress for teachers of students with emotional and behavioral disorders. Journal of Emotional and Behavioral Disorders, 9, 123-131.
New Freedom Commission on Mental Health. (2003). Achieving the promise: Transforming mental health care in America. Final report (DHHS Publication No. SMA-03-3832). Rockville, MD: U.S. Department of Health and Human Services.
Owens, J., & Murphy, C. (2004). Effectiveness research in the context of school-based mental health. Clinical Child and Family Psychology Review, 7, 195-209.
Pace, T., Mullins, L., Beesley, D., Hill, J., & Carson, K. (1999). The relationship between children's emotional and behavioral problems and the social responses of elementary school teachers. Contemporary Educational Psychology, 24, 140-155.
Packman, J., & Bratton, S. (2003). A school-based group play/activity therapy intervention with learning disabled preadolescents exhibiting behavior problems. International Journal of Play Therapy, 12, 7-29.
Phillips, R., & Landreth, G. (1995). Play therapists on play therapy: I. A report of methods, demographics, and professional practices. International Journal of Play Therapy, 4, 1-26.
Pianta, R., Steinberg, M., & Rollins, K. (1995). The first two years of school: Teacher-child relationships and deflections in children's classroom adjustment. Development and Psychopathology, 7, 295-312.
Pianta, R., & Stuhlman, M. (2004). Teacher-child relationships and children's success in the first years of school. School Psychology Review, 33, 444-458.
Post, P. (1999). Impact of child-centered play therapy on the self-esteem, locus of control, and anxiety of at-risk 4th, 5th, and 6th grade students. International Journal of Play Therapy, 8, 1-18.
Raman, V., & Kapur, M. (1999). A study of play therapy in children with emotional disorders. NIMHANS Journal, 17, 93-98.
Ray, D. (2004). Supervision of basic and advanced skills in play therapy. Journal of Professional Counseling: Practice, Theory, and Research, 32, 28-41.
Ray, D., Armstrong, S., Warren, S., & Balkin, R. (2005). Play therapy practices among elementary school counselors. Professional School Counseling, 8, 360-365.
Ray, D., Bratton, S., Rhine, T., & Jones, L. (2001). The effectiveness of play therapy: Responding to the critics. International Journal of Play Therapy, 10, 85-108.
Rogers, C. (1942). Counseling and psychotherapy. Boston: Houghton Mifflin.
Ryan, S., Gomory, T., & Lacasse, J. (2002). Who are we? Examining the results of the Association for Play Therapy membership survey. International Journal of Play Therapy, 11, 11-41.
Schumann, B. (2005). Effects of child-centered play therapy and curriculum-based small group guidance on the behaviors of children referred for aggression in an elementary school setting (Doctoral dissertation, University of North Texas, 2004). Dissertation Abstracts International, 65/12, 4476.
Shadish, W., Cook, T., & Campbell, D. (2002). Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton Mifflin.
Shashi, K., Kapur, M., & Subbakrishna, D. (1999). Evaluation of play therapy in emotionally disturbed children. NIMHANS Journal, 17, 99-111.
Shen, Y. (2002). Short-term group play therapy with Chinese earthquake victims: Effects on anxiety, depression, and adjustment. International Journal of Play Therapy, 11, 43-63.
Tabachnick, B., & Fidell, L. (2001). Using multivariate statistics (4th ed.). Boston: Allyn and Bacon.
Thompson, B. (2002)."Statistical," "practical," and "clinical": How many kinds of significance do counselors need to consider? Journal of Counseling and Development, BO, 64-71.
Dee C. Ray, Ph.D., is director of the Child and Family Resource Clinic and assistant professor in the Department of Counseling, Development, and Higher Education at the University of North Texas, Denton. E-mail: email@example.com
Table 1. ITS Total Stress Means, Standard Deviations, and Sample Sizes for Treatment Groups Variable M SD n Pretest PT 150.28 59.63 32 CO 153.79 47.11 29 PTC 163.13 52.18 32 Posttest PT 139.97 52.64 32 CO 134.31 42.41 29 PTC 144.38 43.28 32 Table 2. Summary of Split-Plot Analysis of Variance for Total Stress Scores According to Group Assignment Source df SS MS F p Between subjects Intercept 1 4045751.76 4045751.76 917.47 <.01 Group 2 3538.27 1769.13 .40 .67 Error 1 90 396872.35 4409.70 Within subjects Time 1 12149.80 12149.80 20.15 <.01 Time x group 2 814.18 407.09 .68 .51 Error 2 90 54253.06 602.81
|Printer friendly Cite/link Email Feedback|
|Author:||Ray, Dee C.|
|Publication:||Professional School Counseling|
|Date:||Apr 1, 2007|
|Previous Article:||Conducting adolescent violence risk assessments: a framework for school counselors.|
|Next Article:||An examination of school counselor involvement in school-family-community partnerships.|