The school counselor activity rating scale: an instrument for gathering process data.

The importance of collecting process data describing school counselor practice is widely supported as a component of accountability. However, the lack of practically significant and valid instruments may hamper this practice. The School Counselor Activity Rating Scale was designed to measure how school counselors actually spend their time versus how they would prefer to spend their time in job-related activities. Its development, initial validity and reliability results, and potential applications are presented in this article.

**********

"What do school counselors do?" is not a new question. Indeed, on any given day, practicing school counselors need to provide an answer to this question for teachers, administrators, parents, and in truth, themselves. One way to answer this question is to collect and analyze process data. Process data can describe an important aspect of school counselor practice and effectiveness, that is, how school counselors are spending their time in their day-to-day activities (American School Counselor Association [ASCA], 2003). Given that there is a continual need for school counselors to describe and account for what they do, it follows that an instrument that can be used to collect relevant process data is warranted. The purpose of this study was to examine initial validity, and reliability data on the School Counselor Activity Rating Scale, an instrument that school counselors could use to gather process data.

Information describing what school counselors do has been collected and has received a great deal of attention in school counseling literature. Past research has focused on how school counselors spend time in particular activities (Bonebrake & Borgers, 1984; Carroll, 1993; Hardesty & Dillard, 1994; Vandegrift, 1999) and the difference between how school counselors actually spend their time and how they want to be spending their time (Johnson, 1993; Mustaine, Pappalardo, & Wyrick, 1996; Partin, 1993; Tennyson, Miller, Skovholt, & Williams, 1989; Wilgus & Shelley, 1988). In addition, the difference between school counseling practice and best practice, as advocated by leaders of the profession, has received attention (Burnham & Jackson, 2000; Carter, 1993). Although there has been little consistency in the measurement of time in activities, or the types of activities studied, the consistency of findings reveals that there remains a discrepancy between what is advocated as best practice (the type of activities and the time devoted to various tasks as related to program objectives) and what is actually practiced in schools (Burnham & Jackson; Carter; Hutchinson, Barrick, & Groves, 1986; Mustaine et al.; Partin; Scarborough, 2002).

For the practicing school counselor, the main purpose of collecting process data is for program evaluation. Gysbers and Henderson (1988) stated, "The purpose of evaluation is to provide data to make decisions about the structure and impact of the program and the professional personnel involved" (p. 263). Process data describe the way the school counseling program is structured and conducted (i.e., what activities have been done) and whether prescribed practice was followed (ASCA, 2003). According to The National Standards for School Counseling Programs,

   As a first step to understanding your site's use
   of time, all the school counselors should keep
   track of their time and document activities
   performed throughout their days. This allows
   school counselors and administrators to determine
   the amount of time being spent in each
   of the delivery system components [e.g., consultation,
   coordination, counseling, curriculum]
   and in non-school-counseling activities.
   (ASCA, p. 51)

Assessing the amount of time spent in activities addresses what is happening and then provides an opportunity to discuss what should or could be happening (ASCA). In situations with a discrepancy between actual and prescribed practice, school counselors then can make adjustments that will lead to greater effectiveness.

Outcome studies, on the other hand, provide information on the effectiveness of interventions and of the program being implemented. Empirical research supports the benefits of the four overarching interventions of school counselors (consultation, coordination, counseling, curriculum). In addition, there is research supporting the positive effects of more fully implemented comprehensive developmental school counseling programs on overall student development, including academic, career, and emotional development, academic achievement, as well as quality of life (Gysbers, Lapan, & Blair, 1999; Lapan, Gysbers, & Petroski, 2001; Whiston & Sexton, 1998). The implication is that by providing these essential counseling tasks (process), school counselors contribute to student success in the achievement of essential competencies (outcome).

There may also be a second purpose for documenting school counselor activities. School counselors may share these data to inform constituents about their role and impact. Not only has there been research describing the discrepancy between actual and best practice, but the literature has hypothesized a number of reasons for this discrepancy, including case load, amount of paperwork, and years of experience (Carter, 1993; Johnson, 1993; Mustaine et al., 1996). However, the most consistent reason cited for this discrepancy is the assignment of non-counseling duties to school counselors, or the more general concern that school counselor roles and duties are determined by others (i.e., administration) rather than by best practice (Borders & Drury, 1992; Burnham & Jackson, 2000; Carter; Johnson; Mustaine et al.; Partin, 1993). Indeed, several studies highlight the high percentage of school counselor time spent on "nonguidance" activities (e.g., Burnham & Jackson; Partin; Vandegrift, 1999).

A common theme in the literature is that school counselors have not taken ownership of this professional issue and have allowed administrators to continue to define their positions (e.g., Burnham & Jackson; Gysbers & Henderson, 2000; Mustaine et al.). Sears and Coy (1991) stated, "School counselors appear to be reluctant or unable to convince principals that they should perform the duties for which they have been trained" (p. 3). Subsequently, the call has been that school counselors become more active in conveying their role and duties to parents, teachers, and administrators (e.g., Carter; Isaacs, 2003; Mustaine et al.; Partin; Sears & Coy; Wilgus & Shelley, 1988).

THE PURPOSE OF THIS STUDY

There is a consistent call and need for process data reflecting what school counselors do. On the one hand, school counselors need to demonstrate their effectiveness, and on the other hand, they need to advocate for themselves to be able to do the activities that most contribute to their effectiveness. Although there has been significant study of how school counselors spend their time, instrumentation has varied and the instruments were not necessarily made available for practical use. Indeed, authors have rarely addressed the validity or reliability of their instruments.

The purpose of this study was to examine the initial reliability and validity of the School Counselor Activity Rating Scale (SCARS) (Scarborough, 2002). The SCARS was developed to assess performance of actual and preferred job duties currently being carried out by practicing school counselors. Preferred activities refers to school counseling activities recommended by ASCA's (2003) National Model, including consultation, coordination, counseling, and curriculum interventions. School counselors, school district administrators, counselor educators, and researchers could use reliable and valid instruments to gather process data about how school counselors spend their time. In addition, data about how school counselors would prefer to spend their time could be gathered with the same instrument, thus allowing for immediate comparisons and discussion of differences.

METHOD

Instrument

The SCARS was developed in two phases. In Phase 1, the instrument was designed, including task statements, rating scale, and format. In Phase 2, the resulting instrument was pretested. The first step in Phase 1 was the development of a list of work activities reflecting the job of school counselors. Task statements were developed to reflect the activities subsumed under the four major interventions described by ASCA (1999) and the National Model for school counseling programs (ASCA, 2003), as well as activities that are reported to be commonly performed by school counselors. Items were selected to describe activities in each of five areas: counseling (individual and group), consultation, coordination, curriculum (classroom lessons), and "other" (clerical and other duties potentially performed by school counselors).

The professional literature was reviewed for common school counselor work activities, including school counseling textbooks (i.e., Baker, 2000; Gibson & Mitchell, 1995; Gysbers & Henderson, 2000; Myrick, 1997), ASCA publications including position statements and The National Standards for School Counseling Programs (Campbell & Dahir, 1997); and journal articles focusing on school counselor activities (e.g., Bonebrake & Borgers, 1984; Borders & Drury, 1992; Burnham & Jackson, 2000; Carter, 1993; Davis, 1999; Hardesty & Dillard, 1994; Hutchinson et al., 1986; Johnson, 1993; Mustaine et al., 1996; Partin, 1993; Wilgus & Shelley, 1988). Based upon these reviews, 50 items were selected to reflect representative activities for the categories of counseling, consultation, coordination, curriculum, and other duties.

In the second step of Phase 1, literature was reviewed regarding instrument design and the measurement of frequency of activities. There has been wide variation in the assessment of how school counselors spend their time. Participants have been asked to estimate the percentage of time spent in activities (Olson, 1983; Partin, 1993; Vandegrift, 1999), to rank order activities according to how they spent their time (Hutchinson et al., 1986), or to complete a frequency scale (Carter, 1993; Johnson, 1993; Loesch & Vacc, 1993; Tennyson et al., 1989).

The SCARS uses a verbal frequency scale in which school counselors are asked how often an activity is performed. A verbal frequency scale is used as a measure of "how often" an action is taken, rather than a Likert scale that measures "strength of agreement," although both are very similar (Alreck & Settle, 1985). The choice to use a frequency scale was based on A Work Behavior Analysis of Professional Counselors (Loesch & Vacc, 1993; see also Davis, 1999) and the Survey of Counselor Functions (Carter, 1993) as were as the work of Tennyson et al. (1989). It is recognized that the verbal frequency scale has limitations (i.e., the specific amount of time is not assessed), but it was chosen for perceived ease, comprehensiveness, and flexibility. Recording actual time spent in tasks, although potentially most accurate, is time-consuming. An alternative, estimating percentages of time spent in tasks, did not seem to lend well to gathering data on specific activities (as differentiated from the more general intervention categories). The use of a frequency scale, however, expanded statistical analysis options for the data.

The SCARS was designed to measure both the frequency with which the school counselor actually performs the activity and the frequency with which the school counselor would prefer to perform each activity. On the first dimension, actual performance, participants rate on a 5-point verbal frequency scale whether they "1--never do this [activity]," "2--rarely do this [activity]," "3--occasionally do this [activity]," "4--frequently do this [activity]," or "5--routinely do this [activity]." Similarly, on the second dimension, preferred performance, participants rate whether they "1--would prefer to never do this [activity]," "2--would prefer to rarely do this [activity]," "3--would prefer to occasionally do this [activity]," "4--would prefer to frequently do this [activity]," or "5--would prefer to routinely do this [activity]." The choice for scale values was based on A Work Behavior Analysis of Professional Counselors (Loesch & Vacc, 1993; see also Davis, 1999). The SCARS was constructed in a way that participants could identify their actual frequency rating and preferred frequency rating for each activity before moving to the next item.

After the initial development of the instrument, Phase 2 was conducted. The pretest was conducted to assess for production mistakes, question/statement and scale construction, and readability, and understanding (Dillman, 2000). During the first step of Phase 2, interviews were conducted with two participants. One individual participated in a "think-aloud" interview (Dillman). During this process, the participant completed all procedures of the survey process by reading all information aloud and verbally expressing thoughts during the process. This process was done in the presence of the interviewer.

A retrospective technique was used with a second individual (Dillman, 2000). During the retrospective interview, the participant, in the presence of the interviewer, engaged in the survey process as if the interviewer were not present. The interviewer then observed the participant, noting any skipped questions, confused expressions, or other behavior that would seem to indicate a problem with understanding. After completion, the interviewer asked questions about the noted observations. Both the "think-aloud" and retrospective methods were used to gain understanding of how the directions and questions were being interpreted and whether the intent of the items and instrument was being realized (Dillman). In addition, general questions regarding survey design and general reactions to the survey were asked and discussed.

The second step of Phase 2 involved the review of the instrument by knowledgeable colleagues. This step was employed to elicit feedback based on the experience of the participants with previous surveys and knowledge of study objectives (Dillman, 2000). Five individuals were selected, representing different areas of expertise, to review all aspects of the survey process. One individual was selected as an expert on school counseling with experience as a school counselor and over 10 years of experience as a counselor educator with primary emphasis in training master's-level school counselors. Another individual also was chosen as an expert in school counseling with experience as a mental health counselor in schools and 4 years of experience as a counselor educator primarily focused on training school counselors and researching school counselor practice. Another individual was selected as an expert in research methodology. This individual also had prior experience as a school counselor and had been a counselor educator and counseling researcher for almost 30 years. A fourth individual was chosen for her expertise in counselor education without practice or emphasis in school counseling. Because of a lack of knowledge of current school counseling literature, her understanding of the survey was solely based on the construction and readability of the instrument. The final individual was selected as an expert in survey research and instrument construction.

Based on feedback from participants from Phase 2, changes were made in the content and design of the SCARS. Some task statements were removed due to redundancy and replaced with other task statements identified in Phase 1 and reflective of common school counselor practice. The design of the SCARS was altered based on consistent feedback from the participants. Rather than a random listing of the task statements, which resulted in confusion for the majority of participants, task statements were grouped by general category (counseling, consultation, curriculum, coordination, and other).

The initially tested version of the SCARS contained 50 items that described specific work activities as practiced by school counselors or as suggested as important activities by leaders in the school counseling profession. An example of an item from the counseling category is "counsel with students regarding personal problems." An example from the consultation category is "consult with school staff concerning student behavior." An example from the coordination category is "coordinate and maintain a comprehensive developmental school counseling program." An example of an item from the curriculum category is "conduct classroom lessons addressing career development and the world of work." The item "perform hall, bus, cafeteria duty" is an example from the other category.

Participants

Six hundred elementary; middle, and high school counselors were selected from two Southern states to participate in the study. A total of 300 participants (100 per grade level) were randomly selected from each state. The sample from one state was drawn from a list of school counselors employed by the state. As no such list existed in the other state, a list of members of the state school counselor association was obtained. A total of 361 usable surveys were returned and used for analysis. Of the respondents, 117 were elementary school counselors, 120 were middle school counselors, and 124 were high school counselors. The usable survey response rate was 60%.

The sample of school counselors consisted of 89.7% females and 10.3% males. The participants were African American (10.7%), Asian American (.3%), Native American (1.1%), Hispanic American (.6%), and European American (87.4%). The average age of school counselors was 45.72 (SD = 10.02). On average, participants had 11 years of experience, with 27.9% having 5 or fewer years of experience.

Approximately 90% of respondents indicated their school counseling license was a result of a master's degree, whereas 9% qualified because of add-on certification. Over 90% of participants had a master's degree specifically in school counseling. Nearly half (48.1%) of participants completed their school counseling training before 1990, and 43.5% indicated that they had graduated from a Council for Accreditation of Counseling and Related Educational Programs accredited program. Slightly over 28% were members of the American Counseling Association and 41.6% were members of ASCA.

Data Collection

The Dillman Total Design Method of survey procedures was used to collect data for this study (Dillman, 2000). MI surveys were numerically coded for the purpose of minimizing the mailing of subsequent surveys to participants who had already completed and returned the instrument. The first phase of data collection involved mailing a pre-notice postcard to all the randomly selected participants. This postcard informed participants that they would be receiving a survey and asked for their participation. The second phase of data collection, implemented approximately 1 week after the pre-notice postcard, was the mailing of the cover letter, the survey instrument, and a self-addressed, stamped return envelope. The cover letter outlined the general purpose of the research study and the procedures the participants were asked to follow. A statement addressing voluntary participation was included. The third phase of data collection was the mailing of a "thank you" postcard 1 week after the initial survey mailing. Appreciation for responding to the questionnaire was expressed, and the postcard also served as a reminder for those individuals who had not yet returned a completed questionnaire. The final phase of data collection occurred 3 weeks after the initial questionnaire mailing. A replacement questionnaire was sent along with a cover letter asking the recipient to respond. Data collection was completed 6 weeks after the mailing of the initial survey packets.

Statistical Analyses

All returned surveys were used in the analyses. Any missing values were handled by pair-wise elimination from calculations. Because the total number of surveys analyzed was greater than 300, an adequate sample size was attained for statistical analyses (Tinsley & Tinsley, 1987).

A principal components factor analysis with an orthogonal transformation using the varimax rotation was performed to identify factors and assess construct validity of the SCARS. Orthogonal rotation was selected because it forces the factors to be independent from one another (Tinsley & Tinsley, 1987). While it is acknowledged that there may be an overlap of skills in interventions (e.g., counseling and consultation), each intervention is considered to be separate and distinct. Because four intervention categories are recognized within the school counseling profession (counseling, consultation, coordination, and curriculum), a decision was made to analyze these items separately from the "other" activities. Therefore, factor analyses were conducted on the 40 items representing the four categories for each scale ("Actual" and "Prefer") of the instrument. Subsequently, factor analyses were conducted on the 10 items of the "other" category for each scale to aid in understanding of these common activities. Kachigan (1986) indicated that factor loadings of .3, .4, or .5 are most often used for meaningful analysis, therefore a minimum factor loading of .4 was used for this study.

Tinsley and Tinsley (1987) recommended that more than one consideration be used in choices to retain factors. Therefore, the choice to retain factors was influenced through consideration of a scree test, explained variance, eigenvalues, as well as comprehensibility (Kachigan, 1986; Tinsley & Tinsley). The scree test was examined to identify the number of factors appearing to be "real" factors (those to the left of the scree) and those that may be attributed to error (those to the right of the scree). Subsequently, the percentage of variance accounted for by the factor solution was examined. The choice is determined by accounting for the amount of explained variance (comprehensiveness) with as few meaningful factors as possible (parsimony) (Kachigan). As is commonly practiced, only factors with eigenvalues greater than 1 were retained (Tinsley & Tinsley). In addition to these statistical determinates, a factor solution was considered in relation to "what makes the most sense in light of what is already known about the subject matter" (Kachigan, p. 389). Therefore, school counseling literature and previous research were considered in the determination of a factor solution. In addition, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett's test (Tinsley & Tinsley) were examined to determine that the data were suitable for factor analytic procedures.

Final decisions for the determination of subscales were based on factor analysis results and further analysis of the individual items that loaded differently on the Actual and Prefer scales. Construct validity was further assessed through the examination of group differences (one-way analysis of variance [ANOVA]) and correlations between resultant subscales and selected demographic variables. Internal consistency reliability was assessed using Cronbach's coefficient alpha.

RESULTS

Based on analyses of the scree plot and explained variance for the Actual scale, a four-factor solution was found for the original 40 items representing the counseling, coordination, consultation, and curriculum categories (results will be presented below for the items making up the "other" category). All factors met Kaiser's criterion with eigenvalues greater than 1. The amount of explained variance of the four factors was 47.27%. The KMO measure of sampling adequacy was .91 and Bartlett's test was significant. The analysis of the Prefer scale of these same items also revealed a four-factor solution. The amount of explained variance of these four factors was 45.22%. The KMO measure of sampling adequacy was .90 and Bartlett's test was significant. All items had loadings greater than or equal to .40. (Item loading results and means and standard deviations for all items are available from the author upon request.)

An analysis of the results of the Actual and Prefer scales was conducted to evaluate the validity of the four subscales. It was found that 32 of 40 items loaded similarly on the Actual and Prefer scales. Based on item loadings, the following names have been assigned to the factors: Curriculum (Factor 1), Coordination (Factor 2), Counseling (Factor 3), and Consultation (Factor 4).

A total of eight items are included in the Curriculum subscale. An example is "conduct classroom lessons on various personal and/or social traits (e.g., responsibility, respect)." The Cronbach's alpha reliability coefficients for the Curriculum subscale were .93 for Actual and .90 for Prefer. Thirteen items were included on the Coordination subscale. "Inform teachers/administrators about the role, training, program, and interventions of a school counselor within the context of your school" and "coordinate orientation process/activities for students" were examples of Coordination items. The Cronbach's alpha reliability coefficients for this subscale were .84 for Actual and .85 for Prefer. The Counseling subscale included nine total items such as "conduct small group counseling addressing relationship/social skills" and "counsel students regarding personal/family concerns." The Cronbach's alpha reliability coefficients for Counseling were .85 for Actual and .83 for Prefer. Seven items were retained for the Consultation subscale. Items included "coordinate referrals for students and/or families to community or education professionals (e.g., mental health, speech pathology, medical assessment)" and "assist in identifying exceptional children (special education)." The Cronbach's alpha reliability coefficients for this subscale were .75 for Actual and .77 for Prefer. One item, "counsel students regarding academic issues," had a negative loading for the Actual scale of the Curriculum subscale (-.41) but loaded on Factor 4 (Consultation) for the Prefer scale (.55). Initially, it was anticipated that this item would load with other Counseling activities. Given the results of the factor analyses and review of literature, this item was not included in any subscale.

"Other" School Counseling Activities

Based on the analysis of the eigenvalues, a two-factor solution was found for the 10 items representing the Actual scale of the "other" category. The amount of explained variance of the two factors was 48%. The KMO measure of sampling adequacy was .73 and Bartlett's test was significant. Analysis of the Prefer scale of these 10 items revealed a three-factor solution. The amount of the explained variance for these three factors was 57.3%. The KMO measure of sampling adequacy was .73 and Bartlett's test was significant. An analysis of the results of the Actual and Prefer scales was conducted to evaluate the validity of the subscales. After examination, the three-factor solution of the Prefer scale was used to develop meaningful subscales (Clerical, Fair Share, and Administrative).

The Clerical subscale consisted of three items. For example, one was "enroll students in and/or withdraw students from school." The Cronbach's alpha reliability coefficients were .84 for Actual and .80 for Prefer. The Fair Share subscale included five items reflecting a broad range of activities that school counselors may perform (e.g., "participate on school committees"). The Cronbach's alpha reliability coefficients for the five items were .53 for Actual and .58 for Prefer. Factor 3 also was composed of activities considered to be "non-school counseling." This subscale was referred to as "Administrative" based on an assumption that administrators are charged with overseeing these responsibilities (e.g., "substitute teach or cover classes," "handle discipline of students"). The Cronbach's alpha reliability coefficients for this subscale were .43 for Actual and .52 for Prefer.

Further Assessment of Construct Validity

Convergent construct validity was established by examining group differences among grade levels of employment on the SCARS subscales (Actual scale). Because job roles and activities may be expected to differ based on grade level of employment (elementary, middle/junior high, high school), group differences were expected (Davis, 1999; Hardesty & Dillard, 1994, Myrick, 1997; Sisson & Bullis, 1992). Given the large sample size (N = 360) and balanced groups (elementary, n = 117; middle/junior, n = 120; high school, n = 124), ANOVA is a robust statistic useful in examining group differences. With a Bonferroni corrected alpha level (.05/7 = .007), analyses revealed a statistically significant effect of grade level on all seven SCARS subscales. Effect sizes measured by [[omega].sup.2] are considered medium or large (Kirk, 1996). See Table 1 for results.

Scheffe's post hoc tests revealed significant differences among all three grade levels on the Curriculum, Coordination, and Counseling scales, with elementary school counselors spending more time in each area than were middle and high school counselors, and middle school counselors spending more time than were high school counselors on those activities as measured by the SCARS. High school counselors, on the other hand, spent significantly more time on Clerical duties than did middle and elementary school counselors, and middle school counselors' means were higher than those of the elementary school counselors. Differences among all three groups were found on the Administrative subscale, with middle school counselors spending more time in those areas than the other two levels, and high school counselors spending the least amount of time in those activities as defined by the SCARS. There were no differences between elementary and middle school counselors on the Consultation and Fair Share subscales; however, each spent significantly more time in those duties than did high school counselors. (See Table 2.)

Discriminant construct validity was established by examining correlations between SCARS subscales (Actual scale) and a demographic variable with which they are not expected to associate. The relationship between years of experience and practice has had mixed empirical support (Brott & Myers, 1999; Carter, 1993; Mustaine et al., 1996; Sink & Yillik-Downer, 2001). Results indicate only two modest correlations--between years of experience and the Coordination subscale (r = .21, p < .001) and the Consultation subscale (r = .19, p < .001).

DISCUSSION

This study examined initial data on the reliability and validity of the SCARS. Results supported the utility of the SCARS to be a measure of process data reflecting how school counselors actually may spend their time versus how they would prefer to spend their time in job-related activities. In addition to the examination of individual activities, the SCARS also may be useful in measuring time devoted to the major interventions of a comprehensive, developmental school counseling program (ASCA, 1999, 2003; Borders & Drury, 1992; Cobia & Henderson, 2003). Content validity, construct validity, and reliability were assessed on the 40 items representing the activities associated with the four intervention categories recognized by the school counseling profession (ASCA, 1999, 2003).

Overall, findings supported the curriculum, coordination, counseling, and consultation interventions of school counselors. The curriculum intervention involves school counselors spending time conducting classroom lessons to students. Encompassed in Factor 1 (Curriculum) was the conducting of classroom lessons reflecting competencies addressed in The National Standards for School Counseling Programs (Campbell & Dahir, 1997), such as career development, personal and social characteristics, social skills, conflict resolution, and personal safety. School counselors perform their role as coordinator as they engage in activities to develop, implement, and evaluate the counseling program, or to organize and manage services to students including special events and programs (Myrick, 2003). These activities were represented in Factor 2 (Coordination). School counselors are trained professionals who meet with students individually and in groups to address concerns that may be interfering with the students' learning and personal development. Counseling (individual and group) related to personal and family concerns, school behavior, crisis, relationships, substance abuse, as well as follow-up on clients (students) was represented in Factor 3 (Counseling). School counselors are consultants when they spend time working individually or with groups of teachers and parents, or other mental health and school personnel, for the purpose of identifying and/or meeting the needs of students. These activities were represented in Factor 4 (Consultation).

A separate analysis was conducted on the 10 items reflecting the school counselor activities considered extraneous to the delivery of a comprehensive, developmental school counseling program ("other" duties). Although there is a great variety reflected in these activities, the common element is that they are not directly related to facilitating the academic, career, or personal/social development of students. They do, however, reflect tasks that are necessary within schools. By categorizing these activities more specifically (as opposed to a broad category of "non-school counseling program" tasks), school counselors and their constituents can gain a clearer picture of the activities that school counselors are performing and the amount of time that they usurp from direct interventions focused on students' success and development. The results supported three underlying factors (Clerical, Fair Share, and Administrative) in which "non-school counseling program" activities may be categorized.

An examination of group differences on each of the subscales by grade level of employment lends further validity support as the expected significant relationships were found. Correlations between the subscales and years of experience also were examined. As anticipated, particularly in light of the large sample size, results indicated modest, if any, correlations between years of experience and the subscales.

Limitations

Further research on the SCARS could continue to investigate its psychometric properties. Two of the three subscales reflecting "non-school counselor program" tasks had moderate reliability scores. Although it is possible that the small number of items in each of these categories contributed to this finding, further exploration is warranted. Studies that assess test-retest reliability or reliability over time (i.e., administrations of the instrument at different points during the school year) may be necessary given that different activities may be given different priority based on school schedules across the academic year. Given that the sample on which the analyses were based comprised school counselors from only two states, analyses on school counselors in other states or a national sample is needed.

Implications

Results suggested that the SCARS may be useful in collecting process data describing how school counselors are conducting their program (the Actual scale) as well as comparing this with how they would prefer to be spending their time (the Prefer scale). Although school counselor activity has been measured often and in a variety of ways, a valid and reliable instrument that can be used to gather this information would be relevant and useful for practicing school counselors, counselor educators, and school counseling researchers.

As Whiston (2002) stated,

   This is a critical time for leaders in school
   counseling to invest in the future of the profession
   and support school counseling
   research. School counselors may believe they
   make a difference, but without "hard data" to
   support these claims, school counselors run
   the risks of losing their positions. (p. 153)

School counselors could benefit from a user-friendly means to collect and analyze data to demonstrate how their programs are being implemented (ASCA, 2003; Kuranz, 2002). For example, school counselors could use this information as part of an overall program evaluation report or to educate constituents about their role and activities. Activities can be reported in specific terms (items) easily understood by school counselors and their constituents (i.e., administration, teachers, and principals) or in general terms reflecting professional terminology. (subscales). The SCARS is one example of an instrument that not only provides this information, but the data can be easily analyzed by looking at scale scores and the means and standard deviations of subscales.

Counselor educators could use an instrument such as the SCARS as part of training focused on illustrating the specific activities of school counselors. In addition, discussions of the differences between the "ideal" and "reality" of practice as suggested by research on school counselor professional identity development (Brott & Myers, 1999) may be aided by concrete data reflecting areas of agreement and discrepancy.

It is necessary to study school counselor practice and the impact of counselor activities. Although studies have consistently found that school counselors are not spending their time as they prefer and that much of what they do is not reflective of what is currently advocated as best practice, this problem still remains. Yet, studies assessing the relationship between fully implemented school counseling programs and student outcomes have reported positive and hopeful results (Lapan, Gysbers, & Sun, 1997). Data collected from an instrument that reflects how school counselors spend their time could be used as dependent (or criterion) variables in studying the factors related to practice. In addition, studies could be designed to test interventions related to moving school counselors from current practice to becoming more aligned with best practices.

Table 1. Analysis of Variance Results for SCARS Subscales by
Grade Level of Employment (Elementary, Middle, High School)

Source          df      SS       MS        F      [[omega].sup.2]

Curriculum
Between           2  16302.16  8151.08  240.63 *        .57
Within          358  12126.91    33.84

Coordination
Between           2   3481.43  1740.71   23.12 *        .11
Within          358  26955.69    75.30

Counseling
Between           2   3627.68  1813.84   61.81 *        .25
Within          358  10505.69    29.34

Consultation
Between           2    939.59   469.80   22.77 *        .11
Within          358   7387.47    20.62

Clerical
Between           2   4931.73  2465.86  376.49 *        .68
Within          358   2344.75     6.55

Fair Share
Between           2   1519.14   759.57   49.19 *        .21
Within          358   5528.29    15.44

Administrative
Between           2    127.73    63.86   32.67 *        .15
Within          358    699.85     1.95

Note. * p < .007 (Bonferroni adjustment).
Effect size measured by [[omega].sup.2].

Table 2. Comparisons Among SCARS Subscale Means by Grade Level

                      Elementary           Middle

Subscale            Mean       SD      Mean       SD

Consultation     3.51 (a)     .64   3.60 (b)     .61
Coordination     3.17 (a)     .62   2.90 (a)     .69
Counseling       3.46 (a)     .65   3.14 (a)     .62
Curriculum       3.79 (a)     .75   2.58 (a)     .87
Clerical         1.64 (a)     .88   4.06 (a)     .99
Fair Share       3.62 (a)     .84   3.53 (b)     .79
Administrative   1.84 (a,c)   .70   2.10 (b,c)   .83

                         High
                                        F
Subscale            Mean       SD    (2,358)

Consultation     3.08 (a,b)   .69   240.63 *
Coordination     2.59 (a)     .68    23.12 *
Counseling       2.60 (a)     .53    61.81 *
Curriculum       1.74 (a)     .53    22.77 *
Clerical         4.43 (a)     .66   376.49 *
Fair Share       2.71 (a,b)   .74    49.19 *
Administrative   1.39 (a,b)   .54    32.67 *

Note. Means sharing the same subscript (a or b) differ
at p < .01 (Scheffe's test). Means sharing the same
subscript (c) differ at p < .05 (Scheffe's test).

* p < .007 (Bonferroni adjustment).

References

Alreck, P. L., & Settle, R. B. (1985). The survey research handbook. Homewood, IL: Richard D. Irwin.

American School Counselor Association. (1999). Role of the school counselor. Alexandria, VA: Author.

American School Counselor Association. (2003). The ASCA National Model: A framework for school counseling programs. Alexandria, VA: Author.

Baker, S. B. (2000). School counselor for the twenty-first century (3rd ed.). Upper Saddle River, NJ: Prentice-Hall.

Bonebrake, C. R., & Borgers, S. B. (1984). Counselor role as perceived by counselors and principals. Elementary School Guidance & Counseling, 18, 194-199.

Borders, L. D., & Drury, S. M. (1992). Comprehensive school counseling programs: A review for policymakers and practitioners. Journal of Counseling & Development, 70, 487-498.

Brott, P. E., & Myers, J. E. (1999). Development of professional school counselor identity: A grounded theory. Professional School Counseling, 5, 339-348.

Burnham, J. J., & Jackson, C. M. (2000). School counselor roles: Discrepancies between actual practice and existing models. Professional School Counseling, 4, 41-49.

Campbell, C. A., & Dahir, C. A. (1997). The national standards for school counseling programs. Alexandria, VA: American School Counselor Association.

Carroll, B.W. (1993). Perceived roles and preparation experiences of elementary counselors: Suggestions for change. Elementary School Guidance and Counseling, 27, 216-226.

Carter, R. B. (1993). School counselor role statements: Fact or fiction. Education, 114, 45-53.

Cobia, D. C., & Henderson, D. A. (2003). Handbook of school counseling. Upper Saddle River, NJ: Pearson Education.

Davis, K. M. (1999). An analysis of counseling work behaviors by school counselors. Unpublished doctoral dissertation, University of North Carolina at Greensboro.

Dillman, D. A. (2000). Mail and Internet surveys: The tailored design method (2nd ed.). New York: John Wiley and Sons.

Gibson, R. L., & Mitchell, M. H. (1995). Introduction to Counseling and Guidance (4th ed.). Englewood Cliffs, N J: Prentice-Hall.

Gysbers, N. C., & Henderson, P. (1988). Developing and managing your school guidance program. Alexandria, VA: American Association for Counseling and Development.

Gysbers, N. C., & Henderson, P. (2000). Developing and managing your school guidance program (3rd ed.). Alexandria, VA: American Counseling Association.

Gysbers, N. C., Lapan, R.T., & Blair, M. (1999). Closing in on statewide implementation of a comprehensive guidance program model. Professional School Counseling, 2, 357-366.

Hardesty, P. H., & Dillard, J. M. (1994).The role of elementary school counselors compared with their middle and secondary school counterparts. Elementary School Guidance & Counseling, 29, 83-91.

Hutchinson, R. L., Barrick, A. L., & Groves, M. (1986). Functions of secondary school counselors in the public schools: Ideal and actual. The School Counselor, 34, 87-91.

Isaacs, M. L. (2003). Data-driven decision making: The engine of accountability. Professional School Counseling, 5, 288-295.

Johnson, L. (1993). School counseling: Standards versus practice. The New York State Journal for Counseling and Development, 8, 55-69.

Kachiga n, S. K. (1986). Statistical analysis: An interdisciplinary introduction to univariate & multivariate methods, New York: Radius Press.

Kirk, R. E. (1996). Practical significance: A concept whose time has come. Educational and Psychological Measurement, 35, 746-759.

Kuranz, M. (2002). Cultivating student potential. Professional School Counseling, 5, 172-179.

Lapan, R. T., Gysbers, N. C., & Petroski, G. F. (2001). Helping seventh graders be safe and successful: A statewide study of the impact of comprehensive guidance and counseling programs. Journal of Counseling & Development, 79, 320-330.

Lapan, R. T., Gysbers, N. C., & Sun, Y. (1997). The impact of more fully implemented guidance programs on the school experiences of high school students: A statewide evaluation study. Journal of Counseling and Development, 75, 292-302.

Loesch, L. C., & Vacc, N. A. (I 993). A work behavior analysis of professional counselors. Muncie, IN: Accelerated Development.

Mustaine, B. L., Pappalardo, S., & Wyrick, B. K. (1996).The discrepancy between actual and preferred time on task for Ontario school counsellors. Guidance & Counseling, 11 (2), 32-35.

Myrick, R. D. (1997). Developmental guidance and counseling: A practical approach (3rd ed.). Minneapolis, MN: Educational Media Corporation.

Myrick, R. D. (2003). Developmental guidance and counseling: A practical approach (4th ed.). Minneapolis, MN: Educational Media Corporation.

Olson, B. L. (1983). The counselor's role: Actual and ideal time spent on guidance functions. The Guidance Clinic, 16(3), 11-16.

Partin, R. L. (1993). School counselors' time: Where does it go? The School Counselor, 40, 274-281.

Scarborough, J. L. (2002). Between practice and preference: Factors related to discrepancies in how school counselors spend their time. Unpublished doctoral dissertation, University of Virginia, Charlottesville.

Sears, S. J., & Coy, D. R. (1991). The scope of practice of the secondary school counselor. (ERIC Document Reproduction Service No. ED328830)

Sink, C. A., & Yillik-Downer, A. (2001). School counselors' perceptions of comprehensive guidance and counseling programs: A national survey. Professional School Counseling, 4, 278-288.

Sisson, C. F., & Bullis, M. (1992). Survey of school counselors' perceptions of graduate training priorities. The School Counselor, 40, 109-117.

Tennyson, W. W., Miller, G. D., Skovholt, T. G., & Williams, R. C. (1989). How they view their role: A survey of counselors in different secondary schools. Journal of Counseling & Development, 67, 399-403.

Tinsley, H. E. A., & Tinsley, D. J. (1987). Uses of factor analysis in counseling psychology research. Journal of Counseling Psychology, 34, 414-424.

Vandegrift, J. A. (1999). Are Arizona public schools making the best use of school counselors? Results of a three-year study of counselors' time use (Report No. CG028489). (ERIC Document Reproduction Service No. ED430178)

Whiston, S. C. (2002). Response to the past, present, and future of school counseling: Raising some issues. Professional School Counseling, 5, 148-155.

Whiston, S. C., & Sexton, T. L. (1998). A review of school counseling outcome research: Implications for practice. Journal of Counseling & Development, 76, 412-426.

Wilgus, E., & Shelley, V. (1988).The role of the elementary-school counselor: Teacher perceptions, expectations, and actual functions. The School Counselor, 35, 259-266.

Janna L. Scarborough, Ph.D., NCC, NCSC, is an assistant professor and school counseling program coordinator, Counseling and Human Services Department, Syracuse University, Syracuse, NY. E-mail: [e-mail]