An Effective Approach to Consultation

Patricia S. Otwell, Doctoral Student
Georgia State University
Elementary School Counselor, Gwinnett County Schools

Fran Mullis
Assistant Professor
Counseling and Psychological Services, Georgia State University

Published in the Professional School Counseling Journal
Published by American School Counselors Association
1:1 October 1997


The purpose of this study was to determine the effects of counselor-led staff development (Active Teaching, Popkin, 1994) as a group consultation strategy. This study investigated effects on student behavior, locus of control, achievement, attendance, and classroom environment. Four of the above variables showed little change. There was, however, a significant improvement in student achievement. Teacher evaluations indicated that the program was a positive, morale-building experience that appeared to increase cohesion among the participants.

Counselor-led Staff Development:
An Efficient Approach to Teacher Consultation

School counselors are human behavior and relationship specialists who make use of the following helping processes: (a) counseling, (b) coordination, and (c) consultation. Consultation is defined as “a cooperative process in which the counselor-consultant assists others to think through problems and to develop skills that make them more effective in working with students” (American School Counselor Association, 1990, p. 1). Consultation can take place in individual or group conferences, or through staff development activities (American School Counselor Association, 1990). This study focused on the effectiveness of counselor-led staff development as a consultation strategy.

In discussing consultative interventions that work, Mathias (1992) wrote, “group consultation may be the most effective. efficient, and powerful approach to a problem” (p. 193). Counselors regularly conduct training courses for groups of teachers on many topics, such as developing positive relationships with students (Rice & Smith, 1993), working with parents of children with disabilities (Berry, 1987), enhancing communication skills (Haws, 1989), and fostering positive self-esteem (Braucht & Weime, 1992; Leonard & Gottsdanker-Willekens, 1987; Maples, 1992). Although most school counselors seem to believe that consulting with teachers in groups is a worthwhile use of their time, there is very little published empirical evidence to support the effectiveness of counselor-led training for teachers.

Limited research on counselor-led training has been conducted in the areas of sexual abuse (Allsopp & Prosen, 1988) and suicide-awareness (Klingman, 1990). One study on counselor-led training (Robinson & Wilson, 1987) examined, with positive results, the effects of a human relations staff development program on student achievement and self-concept. However, as stated by Robinson and Wilson (1987), “more effort needs to be directed toward research in the broad area of consultation” (p.130).

Numerous consultation models and/or theories have been discussed in the literature (Brack, Jones, Smith, White, & Brack, 1993; Bundy & Poppen, 1986; Dustin & Ehly, 1992; Erchul & Conoley, 1991; Fuqua & Kurpios, 1993; Hall & Lin, 1994; Kern & Mullis, 1993; Stoltenberg, 1993; Zins, 1993). The four general models characterized as most applicable to school consultation are process consultation, mental health consultation, behavioral consultation, and advocacy consultation (Conoley & Conoley, 1982). All have problem-solving as a primary goal, while prevention of problems is secondary (Medway, 1989; Zins, 1992).

The teaching of such Adlerian constructs as social interest, encouragement, and goal-directness represents an educative and problem-solving, as well as a preventive, approach (Dustin & Ehly, 1992). Therefore, as a teaching model which emphasizes ways to promote positive behavior in students (Gladding, 1992), the Adlerian approach to consultation may be the most relevant to the school counselor. As might be expected, a research review of school counselors’ effectiveness as consultants concluded that in the studies showing significantly positive effects, the Adlerian consultation approach was a predominant model (Bundy & Poppen, 1986).

Consultation by school counselors has been shown to be an effective, preventive intervention, which can enhance overall academic achievement, increase self-esteem in students, and improve classroom management skills (Bundy & Poppen, 1986; Cecil & Cobia, 1990). As consultants, school counselors not only provide recommendations for helping children with learning and/or behavioral difficulties, they assist staff members to understand individual children, to develop appropriate learning activities, and to encourage productive classroom environments. This assistance is a priority to many teachers who regard consultation as the most beneficial service provided by counselors (Bundy & Poppen, 1986; Hall & Lin, 1994; Wilgus & Shelley, 1988).

Counselors have been urged to expand their roles as consultants for many years (Dinkmeyer, 1967; Dinkmeyer, Carlson, & Dinkmeyer, 1994; Faust, 1968) because consultation is an efficient method of impacting the well-being and personal development of many more students than can be seen directly by a counselor (Campbell, 1992). The counselor-consultant who consults with one teacher is, in effect, reaching all the students in that teacher’s classroom (Myrick, 1993). Likewise, the counselor who conducts a staff development course for as many as 20-30 teachers at one time may positively impact hundreds of students.

Although the southern Association of Colleges and Schools (SACS) has recently reduced its member elementary schools’ student-to-counselor ratios to 749 students per counselor (Policies, principles, and standards for elementary schools accredited by the Commission on Elementary Schools, 1994-95), this remains a daunting number. Staff development is way to reach more students and may be one solution to the problem of very high student-to-counselor ratios for the average school counselor.

Consulting has become a preferred activity for many counselors (Dustin & Ehly, 1992). In addition to the benefits previously stated, most school counselors believe that changes in a student’s behavior are more likely to be accomplished through changes in behavior of the significant adults in the student’s life than through direct counseling services to that student (Bundy & Poppen, 1986; Campbell, 1992; Myrick, 1993).

The purpose of the following study was to assess the value of a counselor-led training program for teachers and to investigate its effects on classroom environment, student behavior, student achievement, student attendance, and student locus of control. It was assumed that the staff development would be effective, and also that it would be an economical and efficient use of counselor time.

Furthermore, the training of teachers in Adlerian techniques extends the concept of preventive consultation. Effective staff development, in which teachers learn strategies for creating a healthy classroom environment, is proactive and preventive—thus consistent with the overall philosophy of school counselors. With that in mind, the Adlerian-based Active Teaching program (Poplin, 1994) was chosen as the intervention for this study.



Selected classroom teachers and students from a public elementary school situated in a sububan area outside a large metropolitan city in the Southeast (classroom teacher n=14 (12 females and 2 males), student n=245) comprised the sample for this study. The mean age of the teachers was 38.86 and the mean years of teaching experience were 10.57. There was no control group due to administrative restrictions regarding additional paperwork for teachers.

The elementary school (kindergarten-fifth grade) from which this sample was selected primarily serves a lower middle to upper middle class population and has a total of 1099 students. According to this school’s latest Student Ethnic Report, there are 759 Caucasian students (69.1%, 169 African-American students (15.4%), 6 Multi-Racial students (.5%), and 2 Native American students (.2%). The “free or reduced price” lunch list reveals a total of 186 students (17%) receiving financial assistance. Similar percentages of ethnicity and subsidized lunches are reflected in each classroom.

Fourteen teachers voluntarily participating in an Active Teaching (Popkin, 1994) class made up the treatment group. They were paid a stipend of $150 each and they received one Staff Development Unit (SDU) toward certification renewal. The entire class was made up of 29 participants who included 2 paraprofessionals, 1 speech therapist, 1 counselor, 9 special area teachers, and 16 classroom teachers. For the purpose of this study, only 14 classroom teachers (those teaching first through fifth grades) were included. Kindergarten teachers felt that the assessment instruments were conceptually too difficult to obtain valid result from their students. Feedback and participation from all class participants were encouraged throughout the class sessions.


The treatment used was the Adlerian-based Active Teaching program (Popkin, 1994). The program consists of video vignettes and instructor-led discussions of concepts presented in the textbook and on the video. The instructor for this program was one of the two counselors employed at this school. The program instructor, a first-year counselor, was a classroom teacher for 20 years. The treatment group attended five 2-hour, after-school classes totaling 10 hours during January, 1995.


The Behavior Rating Checklist (BRC) was used to assess individual students’ classroom behavior. The instrument contains 10 items rated by teachers on a nine-point scale from “there is no evidence of this behavior” to “this behavior is frequent and typical”. The 10 items are directly related to behaviors discussed in the treatment program. The two scales of the BRC are: (a) Ready, Willing, and Able and (b) Disruptive Classroom Behavior. The instrument was designed to investigate elementary students’ changes in behavior over time; specifically, students’ on-task and productive use of time and students’ in-class behavior.

The BRC is reported to have an internal consistency reliability coefficient of .87 for the Ready, Willing, and Able scale and .88 for the Disruptive Classroom Behavior scale. Test-retest reliability coefficients, which range from .71 to .73 on the two factors, were obtained using a six-month span of time (Edwards, 1993).

Doss and Ligon (1979) validated the BRC by asking teachers to rate two groups of students: (a) students with “specific behavior problems” and (b) a random sample of other students in the same schools. The two groups showed significant differences in their mean scores on the two scales, thus validating the instrument.

My Class Inventory (MCI) (Fraser, 1989) was used to assess individual students’ perceptions of classroom environment. The MCI is not only a valid and reliable instrument (Fisher & Fraser, 1981), but it is also economical and easy to administer and hand-score. It is a 25-item instrument to which students respond “Yes” or “No” to each individual item. The MCI was designed for use with children in the 8 to 12 years age range (Fisher & Fraser, 1981) and is well–suited for use with elementary school children because of the low reading levels of its items (Fraser, 1989; Fraser & O’Brien, 1985).

The five scales which make up the MCI are: (a) Satisfaction (extent of enjoyment of class work); (b) Friction (amount of tension and quarreling among students); (c) Competitiveness (emphasis on students competing with each other); (d) Difficulty (extent to which students find difficulty with the work of the class); and (e) Cohesion (extent to which students know, help, and are friendly toward each other) (Fraser, Anderson, & Walberg, 1982; Fraser & Fisher, 1983).

Fraser, Anderson, and Walberg (1982) reported internal consistency reliability estimates (alpha coefficients) of .88 for Satisfaction, .75 for Friction, .81 for Competitiveness, .73 for Difficulty, and .80 for Cohesion, when whole classes were used for the unit of analysis (n=100). Fraser and Fisher (1981) explored the discriminating validity of the MCI by assessing its ability to differentiate between the perceptions of students in different classrooms. Their assumption was that students within the same class should perceive their class relatively similarly, while mean within-class perceptions should vary from one classroom to another. Using a sample of 2305 students in 100 classes, a one-way ANOVA indicated that each scale differentiated significantly (p <.001) between classrooms.

The Child Nowicki-Strickland Internal-External Scale (CNSIE) was used to measure individual students’ generalized locus of control orientations. The CNSIE is a 40-item instrument to which students respond “Yes” or “No” to each individual item, none of which has a reading difficulty higher than the fifth grade level. An item example is: “Do you believe that most problems will solve themselves if you don’t fool with them?”

This instrument is described as a “usable, reliable, and valid measure of generalized locus of control of reinforcement for different-aged children” (Nowicki & Strickland, 1973, p. 153). Reliability and validity have been demonstrated in over 2500 subjects tested. Nowicki and Strickland (1973) reported test-retest reliabilities sampled six weeks apart as .63 for third graders (n=99). In terms of convergent validity support for the CNSIE, Nowicki and Strickland (1973) reported data showing moderate relations between the CNSIE and other measures of locus of control. The CNSIE is considered the best scale for measurement of children’s locus of control (Gordon, 1977).

Unobtrusive measures were used to further assess possible changes in student behavior. Data were gathered from the following sources: (a) Report Card Grades, (b) Attendance Records, and (c) numbers of Discipline Referrals to school administrators.


Participating teachers used the BRC to rate student behavior prior to the beginning of training and three months after training. The MCI and CNSIE were administered to students by their classroom teachers the week of the first class session and three months after training. To allow for individual reading abilities, teachers were instructed to read all items on the MCI and CNSIE aloud while students responded.

The number of Discipline referrals was recorded for comparison purposes using totals before the beginning of training and for the second half of the school year. Likewise, Report Card Grades and Attendance records were recorded and compared for the first and second halves of the school year.


A t-test for correlated samples was used to assess the effectiveness of counselor-led teacher group consultation (Active Teaching, Poplin, 1994) on student locus of control, classroom behavior, perceptions of classroom environment, achievement, school attendance, and discipline referrals. The probability level was set at .05.

Analysis of results from the CNSIE indicated no significant difference in generalized locus of control orientation between the correlated samples when the five grade levels were combined (17.20; 16.68; t=1.84). When the grade levels were examined separately, both the first grade (20.35; 18.21; t=3.04*) and the fifth grade (16.68; 15.24; t=2.16*) showed a significant change toward a more internal locus of control.

Teacher ratings of their students’ classroom behavior indicated no significant difference between the pre-and post-test means on the Ready, Willing, and Able subscale of the BRC (46.93; 46.81; t=.31). There was, however, a significant difference between the pre- and post-test means (8.87; 10.05; t=-4.23*) on the Disruptive Classroom Behavior subscale of the BRC, indicating that teachers perceived their students’ classroom behavior as being more disruptive.

The MCI was used to assess students’ perceptions of classroom environment. No significant differences were found in the pre- and post-test means for the Competitiveness (11.92; 11.80; t=.57) and Difficulty (7.30; 6.92; t=2.26) subscales, although the trend was toward less competitiveness and less difficulty. The MCI did show a significant difference in pre- and post-test means between the correlated samples on the subascales of Satisfaction (12.31; 11.62; t=3.52*), Friction (8.91; 10.05; t=-5.50*), and Cohesion (10.20; 9.05; t=5.06*). These scores indicate less satisfaction, more friction, and less cohesion in students’ perception of their classrooms.

Student achievement improved significantly, with the number of A’s and B’s increasing from 125 to 149 between the first and second halves of the academic year (11.36; 13.55; t=_2.53*). The number of students who earned recognition for improvement or effort decreased from 134 to 117 from the first to the second half of the year (9.57; 8.36; t=.85), which was not a statistically significant difference.

Average attendance during the first four months of the academic year was 96.5% and during the first four months of the academic year was 96.5% and during the last five months of the school years was 95.7%. This shows less than a one- percent change.

Discipline referrals increased from 17 to 31 between the first and second halves of the academic year. This difference (1.2; 2.2; t=-1.44) was not statistically significant.


The purpose of this study was to assess the effects of counselor-led staff development (Active Teaching, Popkin, 1994) on student behavior, locus of control, achievement, attendance, and student perception of classroom environment. There was a significant improvement in overall student achievement, and in internalized locus of control in the first and fifth grades, although the other variables showed either no significant change or change in a negative direction. The results of this study were not those expected based on findings from previous research that indicate effective consultation can improve classroom environment and student behavior (Bundy & Poppen, 1986; Cecil & Cobia, 1990) The study did have limitations, however, which could account for these results. In addition, certain unexpected outcomes yielded positive results.

The most obvious limitation is that, for administrative reasons, control groups were not used, and because of this, there is no way to compare students whose teachers did not participate in the program with those who did participate. Another factor, which could have influenced the outcome of the study, was that the post-tests were not completed until the end of the school term. Most people who are involved in the education of children know that student behavior at the end of the year is not always representative of student behavior during the entire academic year.

Another influencing factor is that the students were functioning at a satisfactory level on many of the outcomes being measured before the intervention took place. The consultation intervention used strongly advocates that children make decisions and accept responsibility for the consequences of those decisions. Therefore, locus of control was used as a means of assessing change on this dimension. Overall, there were no statistically significant changes in measurements of students’ locus of control. When grade levels were assessed separately, both the first and fifth grades showed a significant change, with students moving toward a more internal locus of control. Nowicki and Strickland (1973) stated that the results from their research on locus of control could be used as norms for third, fourth, and fifth grade students. In comparing the results from this study with the norms, the third grade was slightly below the norm (more internal locus of control), and the fourth and fifth grade were both well below the norm (more internal locus of control) before the study began. The beginning scores indicate that students were already taking responsibility for their behavior; therefore, the intervention did not result in a significant change overall.

The Satisfaction subscale of the MCI showed a pre-test mean of 12.21 out of a possible score of 15, indicating that students were experiencing satisfaction with their classroom environments before the intervention began. In addition, there was no significant change in students’ Ready, Willing, and Able (RWA) scores on the BRC. Comparing the pre-test mean of 46.93 from the students in this study to the mean of 42.42 from the randomly selected 2540 students assessed during the validation of the BRC (Doss & Ligon, 1979) shows that the students in the current study were already demonstrating behaviors that show a preparedness to do academic work. Although the current study showed students becoming more disruptive in the classroom their pre-test mean of 8.87 and post-test mean of 10.05 on the Disruptive Classroom Behavior scale of the BRC are both lower than the mean of 11.18 obtained in the Doss and Ligon (1979) study. Statistical regression toward the mean must be considered as an influence on these results (Huck, Cormier, & Bounds, 1974)

Because teachers voluntarily participated in this project, it is possible that they agreed to take part because they believed that the objectives of the staff development class were consistent with their teaching philosophies. If that occurred, it is not surprising that their students were already functioning in responsible, cooperative ways.

No significant difference overall was shown on the Competitiveness subscale of the MCI, although there was a decrease in this score in 8 of the 14 classes assessed, with the first grade showing a significant decrease on this subscale. There was also no significant difference on the Difficulty subscale, although 9 of the 14 classes did show a decrease. Significant differences appeared on the other three subscales of Satisfaction, Friction, and Cohesion, but the difference was toward less satisfaction, more friction, and less cohesion, which was not expected. As stated previously, the classes began the study with a high rating on Satisfaction and statistical regression toward the mean could have occurred on this subscale. Assessing students at the end of the year could also have been a factor in the higher ratings for Friction, because of behavior problems that tend to occur at this time. Perhaps when summer vacation nears, students prepare to go their separate ways and feel less involved with the classroom group, resulting in lowered scores on the Cohesion dimension of the MCI.

Students’ achievement improved as would be expected from having teachers participate in a program, which emphasizes encouragement techniques. This supports research cited previously (Bundy & Poppen, 1986; Cecil & Cobia, 1990) that consultation can enhance academic achievement. Attendance did not change significantly, but it was not considered a problem at this school, with attendance averaging 96% during the entire school year.

Although the change was not statistically significant, discipline referrals increased. One hypothesis for this increase is that after the training program teachers had a heightened awareness of behaviors which could be re-directed into more appropriate channels, and that they began using the logical consequence of referring students to the principal for behavior infractions. Inspection of data from individual classrooms indicates, however, that the change was probably due to other reasons. For example, in one class the teacher referred one student during the pre-test time interval and referred four students during the post-test interval, with all of the latter referrals occurring the last day of school. Another teacher referred students four times during the pre-test phase and 12 times during the post-test phase. One student accounted for most of those discipline referrals. If those two classes were removed from the analysis, discipline referrals would have increased from 12 during the pre-test phase to 15 during the post-test phase.

One of the most interesting and unexpected findings from the study was the increase in student referrals to the counselors, both of whom were new that year. Referrals for individual counseling increased from 80 to 98 and referrals for group counseling increased from 232 to 293. The total change from 312 to 391 represents a 25% increase. Although this increase may be an end-of-the-year phenomenon, it is possible that this occurred because, after participating in the staff development sessions, teachers recognized the new counselor as someone who had expertise, which could be helpful in redirecting students’ behavior. Teachers may also have started viewing the counselor as a trusted partner who would work with them, rather than as someone who might blame them for the students’ difficulties. If this is what occurred, then consulting with groups of teachers is a beneficial activity for school counselors, and is an intervention that could assist new counselors in becoming a valued member of the academic team.


The counselor-consultant who conducts an Adlerian-based staff development program might choose from three particularly popular ones: (a) Systematic Training for Effective Teaching (STET) (Dinkmeyer, McKay, & Dinkmeyer, 1980); (b) Cooperative Discipline (Albert, 1991); or (c) Active Teaching: Enhancing Discipline, Self-Esteem, and Student Performance (Popkin, 1994). Each is a comprehensive approach toward encouraging teachers to create democratic classrooms where students are understood, encouraged, treated with respect, and given the opportunity to develop self-esteem. In the group setting of such classes, teachers can experience group dynamics from a personal perspective and begin to develop greater understanding of behaviors in a social context.

Teachers’ evaluations of the staff development course in this study (“Very Good” to “Excellent”) were reinforced by such comments as: “It helped me rethink my approach to handling discipline problems.” “It gave me a lot of ideas about classroom management and dealing with conflict.” Furthermore, other teacher reactions supported Dinkmeyer, Carlson, and Dinkmeyer’s (1994) conclusion that the therapeutic effects of such a group may actually outweigh its educational value. Teachers said that: “It was great to share feelings and frustrations with other teachers.” “It was lots of fun—interesting and helpful.”

Teachers on large faculties often do not really know each other. An opportunity to become friends in a common cause can raise teacher morale and b