Raising achievement test scores of early elementary school students through comprehensive school counseling programs.

While the trajectory of institutional change tends to be more gradual in American education, elementary and secondary schools have not been immune to the accelerating pace of societal evolution (e.g., socio-technological advances or changes in the family structure). Moreover, the publication of A Nation at Risk: The Imperative for Educational Reform (National Commission on Excellence in Education, 1983) and the subsequent passage of the Goals 2000: Educate America Act (U.S. Congress, 1994), have further hastened the pace of school reorganization and reform already underway in various states. Genuine modifications, however, to the industrial model of schooling and traditional pedagogy are often times difficult to detect, especially in secondary buildings (Herr, 2002), and educational scholars and pundits remain unconvinced that American schools will ever truly remake themselves for the better (e.g., Kinsler & Gamble, 2001; McNeil, 2000).

During the past two or so decades, the school counseling profession has transitioned through a reorganizational process as well, with its traditional emphases yielding to a more broad-based structure of assisting schoolchildren and their caregivers (see Erford, House, & Martin, 2003; Gysbers, 2001; Gysbers & Henderson, 2000; MacDonald & Sink, 1999; Sink, 2002; Sink & MacDonald, 1998, for further discussions). Paralleling to some extent school reform, the comprehensive school counseling movement introduced a programmatic orientation to working with students and their caregivers (American School Counselor Association, 2003; Campbell & Dahir, 1997; Dahir, 2001; Dahir, Sheldon, & Valiga, 1998; Gysbers; Gysbers & Henderson).

In order for school reform to be deemed efficacious, student performances in a variety of areas must demonstrate real improvement. Similarly, leading scholars in the school counseling profession have argued that counselors must show student growth across these developmental domains: academic, career, and personal-social (e.g., Adelman & Taylor, 2002; Gysbers, 2001; Herr, 2002; House & Hayes, 2002; Paisley & Hayes, 2003). To realize this difficult goal, schools need to realign their counseling interventions and services within the context of a comprehensive school counseling program or CSCP (Gysbers & Henderson, 2001; Lehr & Sumarah, 2002; Sink, 2002).

Over the past decade or so, this programmatic approach to school counseling has emerged as the most commonly deployed organizational structure for the profession (Sink & MacDonald, 1998), endorsed strongly by national leaders in the field and the professional guild, the American School Counselor Association (e.g., ASCA, 1997, 1999, 2003; Campbell & Dahir, 1997; Cobia & Henderson, 2003; Dahir, 2001; Dahir et al., 1998; Erford et al. 2003; Gysbers & Henderson, 2000; Myrick, 2003b; Schwallie-Giddis, ter Maat, & Pak, 2003; Wittmer, 2000a, 2000b). With the recent publication of The American School Counselor Association National Model: A Framework for School Counseling Programs (ASCA, 2003) school counselors now possess a user-friendly blueprint to structure their work with students and their caregivers (Schwallie-Gaddis et al.).

LINKAGE BETWEEN COMPREHENSIVE SCHOOL COUNSELING PROGRAMS AND ACADEMIC ACHIEVEMENT

In keeping with Lapan's (2001) recent recommendations for CSCP evaluation, Green and Keys (2001) and others (e.g., Adelman & Taylor, 2002; Gysbers & Henderson, 2000; Johnson & Johnson, 2003; Johnson & Whitfield, 1991; Myrick, 2003a; Schmidt, 2000) emphasized that school counseling programs (a) align their student targets with the goals of school reform, (b) use evidence-based best practices, and (c) report outcome-based data as way of ensuring accountability of their work with students and their caregivers.

Several prominent scholars in the school counseling profession have recommended that CSCPs should include results-based assessments, where school counseling program outcome data ought to be directly aimed at improving student learning (e.g., Gysbers, 2001; Lapan, 2001; Lapan, Kardash, & Turner, 2002; Paisley & Hayes, 2003). For instance, House and Martin (1998) and House and Hayes (2002), resonating with the positions of Green and Keys (2001) and Gysbers, called for school counselors to provide evidence that they positively impact student achievement and other relevant outcomes (see also Paisley & Hayes). Bemak (2000) pressed the issue a bit farther. He suggested that by revising school counselors' position descriptions to include the advancement of student academic achievement, the highly publicized performance gap among low income, disadvantaged pupils and other student groups would diminish. As if to put a exclamation point on the issue, a leading school counseling researcher and his colleagues (Lapan et al., 2002) posited, in the context of a CSCP: "By performing a more proactive leadership role in empowering students to become self-regulated learners, professional school counselors will both motivate young people to more fully realize their academic potential [italics added] ..." (p. 264).

Glosoff and Koprowicz (1990) offered preliminary research findings in the area of elementary school counseling and its relationship to enhancing academic achievement, student retention rates, and children's socio-emotional behaviors, attitudes, and skills. However, research attesting to the potential efficacy of CSCPs as a scaffold for improving academic achievement, is scarce (House & Hayes, 2002). Program evaluation studies conducted in Missouri (see Gysbers, 2001; House & Hayes; Lapan, 2001, for summaries), and to a lesser extent in Utah (Nelson & Gardner, 1998), indicate that CSCPs can have a positive influence on various noncognitive and cognitive student outcomes, including achievement. It is unclear, however, whether CSCPs affect boys and girls differently. As such, these issues are examined in the current study.

CSCP studies tend to be correlational in nature and rely heavily on self-report surveys which measure student and staff perceptions of self-efficacy or change (Lapan, Gysbers, Multon, & Pike, 1997). For instance, in a large state-wide study of seventh-graders (13- to 14-year-olds), Lapan, Gysbers, and Petroski (2001) reported that in more fully implemented comprehensive school counseling programs, no matter what the socio-economic status (SES) of the surrounding community, middle school students "felt" they were doing better in school (earning higher marks) and were safer. These results appear to be largely consistent with previous CSCP studies conducted by Gysbers, Lapan, Hughey, and their research colleagues (e.g., Hughey, Gysbers, & Starr, 1993; Lapan, Gysbers, Hughey, & Arni, 1993; Lapan, Gysbers, & Sun 1997). Other research reviewed in the school counseling literature support these findings (Borders & Drury, 1992; Whiston, 2003; Whiston & Sexton, 1998).

To recap, initial research has shown that CSCPs are making a difference in students' lives. Modest improvements have been reported largely by secondary students across a variety of developmental domains. The literature, however, has provided little direct evidence that academic achievement is improving in schools with a well-established CSCE In addition, there is a paucity of research conducted using elementary school students. The findings reported here from a large investigation conducted in Washington state's elementary schools begin to fill this void.

RESEARCH AIM AND HYPOTHESES

In this causal comparative study, we attempt to answer in part this broad question: Does school counselors' work in elementary schools with well-established CSCPs promote higher academic achievement in students? To respond to this research query, the following alternative hypotheses were examined: (1) After accounting for SES differences, third and fourth grade students enrolled in elementary schools with a CSCP will significantly out perform those children who were attending schools with no systemic guidance and counseling program on various formal measures of academic achievement (e.g., reading, mathematics, and listening); (2) After accounting for SES differences, third and fourth grade students enrolled continuously (3 or more years) in elementary schools with at least 5 years of CSCP implementation (i.e., a well-established CSCP) will significantly out perform those children who were attending schools with no systemic guidance and counseling program on various formal measures of academic achievement; (3) After accounting for student differences in SES, there are significant gender differences across achievement domains in schools with or without a well-established CSCP.

METHOD

Participants

School level. One hundred and fifty public elementary schools from Washington state were randomly selected to participate in the study. The buildings represented small (enrollment from 1 to 365 students, n = 49, or 33%), medium (enrollment from 366 to 499 students, n = 57, 38%), and large (enrollment of 500 or more students, n = 44, 29%) schools and were distributed across the state in rural (n = 59, 39%), suburban (n = 57, 38%), and urban (n = 34, 23%) areas. Schools reported an average of 39% of their students eligible for a U.S. government financial assistance program for low income or "disadvantaged" families. Qualified pupils receive a free or reduced cost lunch, and in some cases, breakfast as well. This percentage has been used as an estimate of a family's level of poverty in several statewide educational studies (personal communication with researchers at the Washington School Research Center, 2002). We use this percentage as an inexact estimate of SES or a poverty index.

School personnel. Of the 150 elementary schools, 31 did not have a formally educated counselor on staff, and therefore, these schools were a priori categorized as non-CSCP buildings and further questioning was not needed. Of the original 150 schools contacted, 119 school personnel were surveyed by telephone. They were primarily female (79%) and European American or White (93%), with an average age of 45.5 years (SD = 9.6), representing the general state-wide demographics for school counselors. Most respondents (95%) also reported having at least one graduate degree (e.g., master's). Those more fully queried about their counseling programs either were Washington State Educational Services Agreement (ESA) certificated school counselors (n = 99 out of 119 or 83%), non-certificated but formally trained counselors (2.5%), ESA certificated school psychologists (10%), ESA certificated counselors who were also teachers (2.5%), or ESA certificated social workers (1.7%). Not uncommon in elementary schools, some respondents reported working full time (58%), while others indicated that they were employed only part time (42%). The school personnel's caseloads averaged 451 students and they had worked on average 9 years total in their specific profession and 5 years in their current educational position. Finally, of the original 150 schools contacted, 67 schools were identified as having implemented a CSCP (i.e., the research group) and 83 schools were identified as not using a CSCP (i.e., the comparison group).

Students--total sample. During the 2000-2001 school year, data on students (N = 20,131) within each of the 150 participating elementary schools were also collected. Specifically, information was garnered from schoolchildren in Grades 3 (n = 9,863, 49%) and 4 (n = 10,268, 51%). In general, the students reflected the ethnic diversity of the state (European American, 72%; Hispanic, 12%; Asian American, 6%; African American, 5%; Native American, 3%; other, 2%). The gender distribution was 51% males and 49% female, with less than 1% missing data.

Research and comparison groups. The total student sample was divided into research (CSCP, n = 9,816) and comparison (non-CSCP, n = 10,315) groups. Overall, the research sample (European American, 68%; Hispanic, 11%; African American, 7%; Asian American, 8%; Native American, 3%; other, 2%) was slightly more ethnically diverse than the comparison group (European American, 76%; Hispanic, 12%; Asian American, 5%; African American, 3%; Native American, 2%; other, 2%). The former group comprised 4,999 (51%) males, 4,805 (49%) females, and another 12 (0.1%) students who did not indicate a particular gender. A similar gender breakdown was reported for the comparison group: 5,308 (52%) males, 4,975 (48%) females, and 35 (0.3%) non-identified students.

In regards to total building enrollment, 46% of the research group's students were enrolled in large elementary schools (enrollment of 500 or more students, n = 4,464), 40% in medium-sized schools (enrollment of 366 to 499 students, n = 3,941), and 14% in small schools (enrollment of i to 365 students, n = 1,411). Thirty-eight percent of the comparison group's students attended large schools, 42% were enrolled in medium-sized schools, and 20% attended small schools. The research sample's schools represented a more even distribution of urban (31%), suburban (42%), and rural (26%) buildings than the comparison group's schools (urban, 16%; suburban, 38%; rural, 44%). Schools in the research group reported an average of 39% of their students eligible for a U.S. government program for low income families that involves receiving a free or reduced cost lunch as compared to an average of 37% for the non-CSCP group.

Research subgroup. The CSCP sample was further broken down into a group of "high" CSCP implementation (5 or more years, n = 3,027). In general, the students in these buildings reflected the ethnic diversity of the larger research group from which it was comprised (European American, 68%; Asian American, 12%; African American, 10%; Hispanic, 8%; Native American, 2%; other, 1%); males (51%) slightly outnumbered females (49%). Students in these well-established CSCP schools represented large schools (500 or more students, n = 1,391 or 46%), medium schools (366 to 499 students; n = 1,295, 43%) and small schools (1 to 365 students, n = 341, 11%). Students in these high-implementation schools represented rural (n = 743, 25%), suburban (n = 1,487, 49%), and urban (n = 797, 26%) areas of the state and reported an average of 37% of their students eligible for a federal lunch assistance program for economically disadvantaged pupils.

Instrumentation

The Comprehensive Guidance and Counseling Programs and Student Success in Washington State Elementary Schools Telephone Survey was constructed by the researchers for this study. The questionnaire, administered to the respondents over the telephone, consists of items that solicit various background (e.g., current educational position, full time equivalent, age, gender, ethnicity, academic degree, years as a school counselor, years in current position) and school data (e.g., location [urban, suburban, or rural], grade levels served, total caseload). If the school or district had a CSCP in place that was aligned largely with the "standard" Gysberian-type model (Gysbers, 2001; Gysbers & Henderson, 2000), school personnel answered additional questions about program specifics.

The Iowa Tests of Basic Skills-Form M (ITBS) are a series of norm-referenced achievement measures designed to provide a broad assessment of student progress in basic academic skills. It is administered annually to Washington state's third, sixth, and ninth grade students. The well-established and widely used group-administered test battery has produced internal consistency and other reliability coefficients in the satisfactory range of low .70s to mid .90s (Hoover, Hieronymus, Frisbie, & Dunbar, 1996). In addition, numerous research studies have generally confirmed the validity of the ITBS (see Riverside Publishing, 1997, for a literature review).

The Washington Assessment of Student Learning (WASL) is a relatively new criterion-referenced test (CRT) group-administered annually to Washington state's fourth, seventh, and tenth grade students (Taylor, 2000). A modified Rasch model (Masters' [1982] partial credit scoring approach; see Taylor for details) was used in developing an equal interval scoring system, where a theta value ([theta]) was computed for each examinee and then transformed to a positive, whole number scale through a linear conversion procedure. The possible scale score range for the WASL's four subtests (Mathematics, Reading, Listening, and Writing) is 150 to 600. The general standard for passing the test is 400; however, this cutoff varies to some extent depending on the subtest and year of test administration. The actual range of scale scores in the content areas can also vary slightly each year.

Recently, Taylor (2000), one of the statistical specialists investigating the instrument, reported on the psychometric properties of the WASL administered to fourth-graders in 1999. The validity of the WASL scores was examined using the intercorrelations among WASL strand scores (rs ranged from .42 to .75) and an exploratory factor analysis. The latter analysis revealed three underlying dimensions, that is, language arts, mathematics, and writing (factor loadings ranging from .62 to .79; Taylor). Internal consistency reliability coefficients for the 1999 Grade 4 WASL were generally satisfactory for the Listening ([alpha] = .61), Reading ([alpha] = .86), Mathematics ([alpha] = .88), and Writing ([alpha] = .81) tests. Inter-rater reliability coefficients for the open-ended questions on the different subtests were high (rs ranged from .97 to .98). Since this CRT continues to be refined each year, the psychometric properties for the other versions of the WASL have not been reported in the professional literature or by the State of Washington's Office of the Superintendent of Public Instruction (OSPI).

Procedures

Telephone surveys were administered in the spring of 2001 by three trained school counseling doctoral students at a western Washington university. Initially, those schools in the sample without an identified school counselor only completed the demographics section of the telephone survey. These schools were classified as non-CSCP schools. The rationale behind this classification was derived from the work of Gysbers and Henderson (e.g., Gysbers & Henderson, 2000; Henderson & Gysbers, 1998). These writers indicated that school counselors are the chief program coordinators. Hence, it was assumed that schools without a school counselor would not have a CSCP. The other respondents (most likely school counselors) were directly queried about whether the building or district had in place a CSCP. If it was a "CSCP school," the respondents were asked further questions about their programs.

Various data for the 2000-2001 school year were collected from the Washington School Research Center's database (e.g., percentage of students eligible for a free or reduced cost school lunch, the school's enrollment, and the district's total enrollment). The Center yearly receives school census information from Washington's OSPI. Participant demographic data were garnered from the Center's database, including the schoolchildren's gender, ethnicity, number of years enrolled in the school building, and number of years enrolled in the district. Additionally, 2000-2001 student test scores from the ITBS and WASL were collected (i.e., Grade 3 ITBS Vocabulary, Comprehension, Reading, and Mathematics standard scores, and Grade 4 WASL Listening, Reading, Writing, and Mathematics scale scores). It should be noted that Grade 3 students must take the ITBS, but not the WASL; whereas, the opposite is true for the fourth-graders.

Data Analysis

Telephone surveys were coded and tabulated and the numerical data were analyzed for normality. The parametric assumptions underlying the multivariate procedures were also checked using standard techniques (Gall, Gall, & Borg, 2003; Green & Salkind, 2003; Grimm & Yarnold, 1995).

Using the general linear model (GLM), the hypotheses were tested using a series of factorial multivariate analyses of covariances (MANCOVAs; see Green & Salkind, 2003 and Tabachnick & Fidell, 2000, for details). The independent variables included, for example, group (participants in CSCP vs. non-CSCP schools), length of enrollment (i.e., length of time students were continuously enrolled in their particular schools), and gender. To correct for possible Type I errors, Bonferonni adjustments to the alpha level were made as needed. The dependent measures included the student ITBS or WASL achievement test scores. In each multivariate analysis, the percentage of students receiving a free or reduced cost lunch was used as the covariate. This concomitant variable correlated largely in the moderate range with each dependent measure ([r.sub.M]= .35). Partial eta squares ([[eta].sub.p.sup.2]) were reported as estimates of the effect sizes.

RESULTS

The majority (54.6%) of the participating schools with high CSCP implementation used as their template a Gysberian-type program called the Washington State Guidelines for Comprehensive Counseling and Career Guidance Programs from Kindergarten through Community and Technical College (Coats, Ash, & Dorsey, 1998). On average these schools had their CSCP in place for over 6.5 years (M = 6.64, SD = 2.17), almost double the amount of time the non-high usage or general CSCPs had implemented their programs (M = 3.66, SD = 4.96). The mean length of time respondents in the general CSCP group and high CSCP sample worked in their current positions were 4.96 and 5.46 years, respectively, and their total number of years of educational service were comparable (Ms = 9.4 and 10.5 years, respectively). All counselors in CSCP schools generally viewed themselves as moderately to highly involved in their program. For example, the mean rating for counselors in the high-usage CSCP subsample was 6.27, as opposed to 5.85 for the general CSCP group. The perceived levels of CSCP importance and student impact were similar as well, ranging in the moderate to high levels (Ms > 5).

While not specifically tabled here, the magnitude of the average intercorrelations among the dependent variables was in the moderate to strong range. For the third and fourth grade students, the correlations among the covariate and the dependent measures were on average .58 and .43, respectively.

Means and standard deviations were first calculated by Group (general CSCP vs. non-CSCP), Length of Enrollment, and Gender for Grade 3 ITBS standard scores and Grade 4 WASL scale scores (see Tables 1 and 2). These statistics were also computed by Group (high usage CSCP vs. non-CSCP), Length of Enrollment, and Gender for the Grade 3 ITBS standard scores and Grade 4 WASL scale scores (see Tables 3 and 4). Tables 5 and 6 reveal the descriptive statistics for only those third and fourth grade students, respectively, who reported being enrolled in their particular high-CSCP or non-CSCP schools for at least three years. Preliminary trends in these data can be discerned. For example, students enrolled for several years in high usage CSCP schools tended to perform better on various achievement tests than those pupils in non-CSCP buildings. Various gender differences and interaction effects on several dependent variables are noticeable as well.

Grade 3 Findings

To examine the hypotheses, first, a 2 (Group: CSCP vs. non-CSCP) X 4 (Length of Enrollment) X 2 (Gender) MANCOVA with the poverty index as a covariate was computed on Grade 3 ITBS Vocabulary, Comprehension, Reading, and Mathematics standardized test scores. The statistical procedure produced significant (ps = .00, [[eta].sub.p.sup.2] = .00) main effects for Group (favoring the non-CSCP sample on all dependent measures), Length of Enrollment (i.e., the longer the enrollment, the higher the test scores), and Gender. Furthermore, for Gender, only ITBS Vocabulary score was nonsignificant (p >. 10), with the girls scoring significantly higher than the boys on the Comprehension and Reading subscales; whereas, the males did significantly better than the females on the Mathematics test. The Group X Length of Enrollment and Group X Gender interactions were significant (ps = .00, [[eta].sub.p.sup.2] = .00) across the four ITBS Grade 3 subtests. While the latter interaction is not particularly germane to the study, the former shows, not surprisingly, that the longer the students in either group remain in their schools, the higher their test scores. Interestingly, across the four content areas, students in CSCP schools consistently scored lower than pupils in the non-CSCP buildings, but the performance gap disappears gradually over time such that by about the second year of enrollment, students' mean test scores in both groups are virtually identical.

Second, a 2 (Group: high CSCP implementation vs. non-CSCP) X 4 (Length of Enrollment) X 2 (Gender) MANCOVA with the poverty index as a covariate generated a significant main effect for Length of Enrollment across each of the dependent measures (ps = .00, [[eta].sub.p.sup.2] = .01) and a significant Gender effect for ITBS Comprehension, F (1, 5961) = 5.81, p = .02, [[eta].sub.p.sup.2] = .00, and Mathematics, F (1, 5961) = 14.11, p = .00, [[eta].sub.p.sup.2] = .00, subtests. Similar to the results above, these indicate the longer the students in either sample remain in their schools, the better their performances. The Group X Length of Enrollment interaction effects for Vocabulary, Comprehension, Reading, and Mathematics scores were significant (ps = .00, M [[eta].sub.p.sup.2] = .01), where students who were enrolled for several years or more in high usage CSCP schools scored significantly higher on these ITBS measures than comparable students in non-CSCP schools (see Figures 1, 2, 3, and 4, for interaction graphs). The significant Gender effect showed that females significantly outperformed the males on Comprehension and males did better than females on Mathematics.

[FIGURES 1-4 OMITTED]

Third, to examine further the significant Group (high usage CSCP vs. non-CSCP) main effect reported above, a one-way MANCOVA (with poverty index as covariate) was computed on Grade 3 ITBS subtest scores (Vocabulary, Comprehension, Reading, and Mathematics) of schoolchildren who had been continuously enrolled (3 or more years) in their schools. A significant main effect for Group was found, Wilks' A = .99, F (4, 2355) = 5.76, p < .000, [[eta].sub.p.sup.2] = .01. Moreover, significant Group differences favoring the high CSCP students were found for all four of the dependent measures: Vocabulary, F (1, 2358) = 4.16, p < .042, [[eta].sub.p.sup.2] = .002; Comprehension, F (1, 2358) = 6.32, p < .012, [[eta].sub.p.sup.2] = .003; Reading, F (1, 2358) = 5.85, p < .016, [[eta].sub.p.sup.2] = .002; and Mathematics, F (1, 2358) = 17.81, p < .000, [[eta].sub.p.sup.2] = .01.

Grade 4 Findings

Fourth, a 2 (Group: CSCP vs. non-CSCP) X 5 (Length of Enrollment) X 2 (Gender) MANCOVA with the percentage of students receiving a free or reduced cost lunch as a covariate was computed using Grade 4 WASL Listening, Reading, Writing, and Mathematics scale scores as dependent variables. Significant main effects favoring the non-CSCP sample on all dependent measures were found for Group (ps = .00, [[eta].sub.p.sup.2] = .00) and Length of Enrollment (ps = .00, [[eta].sub.p.sup.2] = .01). Additionally, a significant Gender effect favoring the girls for the WASL Reading, F (1, 8952) = 133.20, p = .02, [[eta].sub.p.sup.2] = .00, and Writing, F (1, 8952) = 321.62, p = .00, ,qp2 = .03, measures were found. The MANCOVA also yielded significant Group X Length of Enrollment interactions for Grade 4 WASL Listening, F (4, 8952) = 4.91, p = .00, [[eta].sub.p.sup.2] = .00, Reading, F(4, 8,952) = 3.81, p = .00, [[eta].sub.p.sup.2] = .00, and Mathematics, F(4, 8952) = 3.60, p = .00, [[eta].sub.p.sup.2] = .00, scale scores. These interactions reflect those reported for the Grade 3 ITBS data, in that, for Grade 4 WASL Listening, Reading, and Mathematics scores, students in CSCP schools consistently scored lower than pupils in the non-CSCP buildings in the first few years of school enrollment, but the initial significant mean differences diminished over time. By at least the third year of school attendance, students' scores in the high-implementation CSCP group were on par with the performances of the non-CSCP participants.

Fifth, a 2 (Group: high CSCP implementation vs. non-CSCP) X 5 (Length of Enrollment) X 2 (Gender) MANCOVA with the poverty index as a covariate was generated. The analysis indicated a significant Group effect for Grade 4 WASL Writing scale scores, F (1, 6100) = 4.54, p = .03, [[eta].sub.p.sup.2] = .00. The high-implementation CSCP students significantly outperformed those participants in the comparison group. Like the earlier MANCOVAs, Length of Enrollment was a significant main effect across each of the WASL subtests (ps = .00, [[eta].sub.p.sup.2] = .01), where students who were enrolled the longest in their particular school outperformed all others on each of the dependent measures. Significant Gender effects were also found for Listening, F (1, 6100) = 4.78, p = .03, [[eta].sub.p.sup.2] = .00, Reading, F (1, 6,100) = 61.65, p = .00, [[eta].sub.p.sup.2] = .01, and Writing, F (1, 6,100) = 141.66, p = .00, [[eta].sub.p.sup.2] = .02, subtests. The gifts had significantly higher achievement than the boys on the Reading and Writing WASL tests, whereas the boys produced superior mean Listening scores than their female counterparts. Finally, as with the third grade students attending a high implementation CSCP school, those Grade 4 pupils enrolled for multiple years in the same high usage CSCP school significantly outperformed comparable children in non-CSCP schools. Significant Group X Length of Enrollment interaction effects for Grade 4 WASL Listening, F (4, 6,100) = 7.54, p = .00, [[eta].sub.p.sup.2] = .01, Reading, F(4, 6,100) = 4.73, p = .00, [[eta].sub.p.sup.2] = .00, and Mathematics, F (4, 6,100) = 3.87, p = .00, [[eta].sub.p.sup.2] = .00, scale scores were found (see Figures 5, 6, and 7 for interaction graphs).

[FIGURES 5-7 OMITTED]

Sixth, to examine further the significant Group (high usage CSCP vs. non-CSCP) main effect reported above, a one-way MANCOVA (with poverty index as covariate) was computed on Grade 4 WASL subtest scores (Listening, Reading, Writing, and Mathematics) of participants who had been continuously (3 or more years) enrolled in their schools. A significant main effect for Group was generated, Wilks' A = .99, F (4, 3251) = 6.30, p < .00, [[eta].sub.p.sup.2] = .01. Significant Group differences favoring the high CSCP students were also found for each dependent variable: Listening, F (1, 3254) = 6.41, p < .011, [[eta].sub.p.sup.2] = .002; Reading, F(1, 3254) = 6.97, p < .008, [[eta].sub.p.sup.2] = .002; Writing, F (1, 3254) = 13.71, p < .000, [[eta].sub.p.sup.2] = .004; and Mathematics, F (1, 3254) = 21.10, p < .000, [[eta].sub.p.sup.2] = .01.

To review the most salient findings, in their first few years of school enrollment, CSCP students largely obtained significantly lower achievement test scores than those students attending schools with no systemic school counseling program. However, the longer students remained in their CSCP schools, this achievement gap between groups significantly narrowed or disappeared altogether. Moreover, a significant interaction between Group and Length of Enrollment in high usage CSCP versus non-CSCP schools was found for both grade levels. Over several years, participants who stayed in high usage CSCP schools significantly outperformed their counterparts in the comparison schools on the Grade 3 ITBS Vocabulary, Comprehension, Reading, and Mathematics, and Grade 4 WASL Listening, Reading, Writing, and Mathematics tests. While of lesser import here, the Gender main effects for Grades 3 and 4 were generally consistent across achievement dimensions. Females in the non-CSCP and CSCP schools tended to do better on the verbally-loaded measures (e.g., Comprehension, Reading, and Writing), while the boys in both type of schools consistently outperformed girls on the mathematics tests.

DISCUSSION

The principal aim of this study was to examine whether students attending elementary schools with comprehensive school counseling programs would produce higher academic achievement than those children enrolled in non-CSCP buildings. Since there have been no other CSCP investigations published with a comparable participant base and methodology, it is difficult to directly apply the findings reported here to existing school counseling research. Nevertheless, the findings do provide some causal comparative evidence for the research question and have implications for the practical use of CSCPs.

Specifically, the investigation's key findings provide support for the first two hypotheses. For instance, even when early elementary-age children (Grades 3 and 4, ages 8 to 11) attend a school with a less than totally engaged CSCP and have significantly lower initial academic achievement than similar pupils in the comparison group, the CSCP participants, over a 2 to 3 year time period, close the achievement gap. Furthermore, children benefit academically, whether they are economically disadvantaged or not, by remaining in schools for multiple years (at least 3 years) with a well-established (5 or more years of implementation) comprehensive school counseling program. In summary, over time, younger students appear to do better on a variety of norm-referenced and criterion-referenced tests of academic achievement than their peers in schools without a CSCP in place. The findings are now examined in light of previous research.

First, this study extends the general efficacy research conducted in grade schools (see Glosoff & Koprowicz, 1990; Whiston, 2003; Whiston & Sexton, 1998, for reviews). These reviewers concluded that students are helped academically and interpersonally by attending schools with elementary counselors in place. CSCP elementary school counselors' are most likely contributing to students' development.

Second, the results augment the CSCP studies carried out with middle/junior and senior high school students (see e.g., Gysbers, 2001; Lapan, 2001, for reviews). In fact, they reinforce Gysbers' informed view:

[W]hen certified professional school counselors have the time, the resources, and the structure of a comprehensive guidance program in which to work, they contribute to positive student academic and career development as well as the development of positive and safe learning climates in schools. (p. 103)

The differences, however, between the studies cited in Gysbers (2001) and Lapan (2001) and the current investigation not only concern the age range of the samples, but also their design and foci. As mentioned earlier, the empirical studies conducted in Missouri were largely self-report- and correlational/prediction-focused, rather than, as in the current study, directed at testing potential group differences among CSCP and non-CSCP schools across an array of achievement measures. Thus, the evidence provided here supplements the Utah and Missouri CSCP research and should encourage those scholars who advocate for school counseling programs to promote academic and nonacademic gains in students (e.g., House & Hayes, 2002; Lapan et al., 2002; Paisley & Hayes, 2003).

Third, it is unclear from the study why in both groups, female students generally outperformed boys on the verbal academic measures and the boys surpassed the girls on the mathematics tests (see hypothesis 3). While previous research has shown that elementary-age gifts tend to earn higher classroom grades than boys in stereotypically "feminine" subjects like reading and writing and boys get better mathematics grades, recent studies conducted in elementary schools using standardized measures of achievement report inconsistent results (see e.g., Hay, Ashman, & Van Kraayennoord, 1998; Pomerantz, Altermatt, & Saxon, 2002 for reviews). An earlier investigation, however, provided convincing evidence that there are gender and developmental differences in academic study behaviors of elementary schoolchildren and these may be one of key reasons for gender differences found on educational achievement measures (Hancock & Stock, 1996). We can only speculate here that findings from this study may reflect disparities in academic study strategies between boys and girls. Clearly, further research is needed.

Implications for School Counseling Practice

The findings presented in this article have implications for school counseling practice. Elementary school counselors would do well, in collaboration with other building and district-wide personnel, to design, implement, and further refine their comprehensive programs so that all children are adequately served. The use of the ASCA National Model (ASCA, 2003) in conjunction with Gysbers and Henderson's (2000) text, Developing and Managing Your School Guidance Program, would be an excellent first step. The ASCA National Model is a sound framework with a solid theoretical and research foundation. The publication outlines (a) how to deliver and manage the program, (b) accountability procedures and processes, (c) student developmental competencies, and (d) school counselors' key CSCP responsibilities.

Once the model is firmly established in all elementary schools--hopefully within 3 to 5 years--school counselors can then strategically focus their work on the academic domain. While it is beyond the scope of this article to provide extensive suggestions for practice, there are many resources and publications to assist CSCP school counselors meet this goal (see Clark & Stone, 2000; Cobia & Henderson, 2003; Erford et al., 2003; Gysbers & Henderson, 2000; Henderson & Gysbers, 1998; Myrick, 2003b; Ripley, Erford, Dahir, & Eschbach, 2003; Vanzandt & Hayslip, 2001, for details).

It is important that classroom gnidance and small groups target, in part, student academic competencies. Previous research has shown the positive impact of classroom guidance lessons on student achievement (e.g., Carns& Carns, 1991; Evans & Burck, 1992; Hadley, 1988; Lapan et al. 1993; Lee, 1993). Elementary students can benefit from mastering a "how to succeed in school" guidance curriculum, which could involve, for example, study skills (e.g., self-assessment, organization, and planning), positive attitudes and behaviors toward school and learning, test-taking strategies, effective writing, and homework completion skills. Counselors could also assist other educators in conducting peer tutoring groups for students at risk for school failure. Providing relevant educational interventions, workshops, and in-services for parents, teachers, and staff on methods of enhancing student learning are beneficial as well.

Most importantly, perhaps, for a comprehensive school counseling program to be effective in fostering academic achievement of all schoolchildren, concerted efforts at collaboration, as recommended by Keys, Bemak, and their colleagues (Bemak, 2000; Green & Keys, 2001; Keys, Bemak, & Lockhary, 1998; Keys & Lockhart, 1999) and others (e.g., Rowley, 2001; Sink & Yillik-Downer, 2001), should be facilitated by the district- and school-level administrative staff. The team approach is a consistent theme in these publications, where members of the larger community, school, parent groups, and student associations work systemically to promote healthy student development. School counselors must continue to take on the role of coordinating and leading these partnerships (see, e.g., Cobia & Henderson, 2003; Keys, Green, Lockhart, & Luongo, 2003, for practical details).

Finally, ongoing evaluation is also required to ensure that school counseling programs are fostering important academic outcomes (Lapan, 2001; Myrick, 2003a; Johnson & Johnson, 2003; Thompson, 2002). Although school counselors and their comprehensive programs should not be held directly accountable for the rise or fall of student test scores or grades, action research involving, for instance, relatively simple pretest-posttest designs can be conducted for activities like classroom guidance and small group counseling. Improvement overtime provides evidence that students are mastering various educational/academic competencies. Moreover, the use of student, parent, faculty, and community leader surveys can document perceived changes in student performances across a variety of developmental domains, including academic achievement (Schmidt, 2000). Grades and standardized test scores can be tracked from year to year, showing that students who participated in educationally focused CSCPs are actually learning school curricula. In short, program accountability should be encouraged.

Research Limitations and Suggestions for Future CSCP Investigations

The investigation's shortcomings are enumerated here. First, we did not closely examine the actual day-to-day learning activities and environments of the schools used in the sample. In depth analyses of the teaching-learning processes going on in these schools are obviously needed to understand the differences between those buildings with CSCPs and those without such programs. Second, since the magnitude of the effect sizes (partial [[eta].sup.2]) were modest at best (Green & Salkind, 2003), the practical value of the results cannot be fully assured. Third, the large sample size and the relative increase in statistical power contributed to producing significant group differences. Fourth, the WASL is a relatively new criterion-referenced test and its psychometric properties have yet to be firmly established in the literature. Fifth, the poverty index used in the study was less than ideal. Better measures of economic disadvantage need to be used in subsequent research. Finally, strong causal attributions cannot be asserted given the lack of experimental controls.

In terms of other recommendations for subsequent CSCP research, studies need to be conducted in "high-achieving" CSCP elementary schools in order to clarify which factors explain the most variance in academic performance and perceptions of school climate. Once these dimensions are better understood, they should be set up in "underachieving" CSCP schools. Subsequent research should determine whether positive student outcomes are similar between disparate subgroups of students (e.g., majority vs. minority children or regular education vs. special education students). Finally, potential gender differences on various CSCP student competencies may need to be further explored.

CLOSING REMARKS AND SUMMARY

The findings from this CSCP study and those discussed in previous articles (e.g., Borders & Drury, 1992; Gysbers, 2001; Lapan, 2001) provide school counseling personnel with a body of evidence that supports the usefulness of comprehensive programs to augment the educational objectives of schools. Hopefully, the results documented here will be sustained in other states and at different grade levels. Even though results are promising, we believe CSCP leadership teams for now should target those evaluative dimensions in which important data (e.g., truancy rates, number of discipline referrals, short-term outcomes from guidance and counseling interventions) can be more readily obtained by very busy school personnel. Raising the bar too high and hastily based on the results of several positive empirical studies, particularly in terms of "guaranteeing" achievement gains in CSCP schools, is not yet fully defensible. Evaluation procedures and processes as suggested by Hughes and James (2001) and Schmidt (2000) are workable approaches for current school counseling leadership teams to use in their daily practice (see also Thompson, 2002). CSCPs should promote measurable changes in the student performances, but the assessment goals and tools need to be deliberate and contextual.

In summary, the current study reveals that early elementary-age students enrolled for several years in well-established CSCP schools produce higher achievement test scores over and above those continuously enrolled children in non-CSCP schools. It is perhaps now appealing for elementary school counselors and administrators, in response to mounting public and administrative pressures, to reprioritize program objectives, centering more on improving academic achievement scores rather than on those less "objectifiable" and "marketable" student competencies (e.g., emotional competence, personal-social, multicultural and critical thinking skills, citizenship/character development). Yielding to the educational policies and forces of the "high stakes" testing movement, in advance of a solid CSCP research base, will place school counselors in the difficult position of requiting too much from their programs too soon. In our view, give the ASCA National Model time to become firmly established and researched before program administers and leaders are required to show demonstrable gains in student academic performance. This is a noble and important goal, but one that the school counseling profession must move cautiously toward.

Table 1. Mean Scores and Standard Deviations for Third Grade ITBS
Standard Scores as a Function of Group (CSCP vs. no CSCP), Enrollment,
and Gender

           Enrolled in    Student    Vocabulary         Comprehension
Group      Building       Gender     M         SD       M         SD

CSCP       This year      Male       181.02    19.23    181.65    20.64
                          Female     179.64    19.61    182.70    19.40
           1 year ago     Male       182.12    21.01    182.58    20.07
                          Female     184.05    20.77    187.10    20.08
           2 years ago    Male       184.29    20.75    184.07    22.06
                          Female     185.71    19.66    186.18    21.02
           3 years ago    Male       187.47    18.10    188.10    20.04
                          Female     186.71    19.14    189.28    19.12
No CSCP    This year      Male       184.68    20.97    185.43    22.50
                          Female     187.24    21.43    18,871    21.17
           1 year ago     Male       183.93    19.88    184.98    19.80
                          Female     184.92    18.68    189.08    20.75
           2 years ago    Male       187.79    19.38    187.79    21.89
                          Female     187.65    19.93    189.54    20.47
           3 years ago    Male       188.59    18.95    189.05    21.16
                          Female     189.04    19.83    191.24    20.43

           Enrolled in    Student    Reading            Math
Group      Building       Gender     M         SD       M         SD

CSCP       This year      Male       181.31    18.86    184.87    18.21
                          Female     181.14    18.28    181.55    16.72
           1 year ago     Male       182.35    19.26    186.33    17.94
                          Female     185.57    19.25    186.44    18.81
           2 years ago    Male       184.18    20.05    187.21    18.76
                          Female     185.91    18.97    186.32    17.88
           3 years ago    Male       187.77    17.68    192.96    17.87
                          Female     187.97    17.91    188.86    17.90
No CSCP    This year      Male       185.03    20.46    189.01    18.97
                          Female     18,793    20.05    187.99    18.94
           1 year ago     Male       184.45    18.54    190.74    17.97
                          Female     186.97    18.37    188.02    18.32
           2 years ago    Male       187.76    19.47    192.03    18.76
                          Female     188.54    18.87    190.33    17.98
           3 years ago    Male       188.82    18.73    193.54    17.29
                          Female     190.14    18.72    190.20    17.90

Table 2. Mean Scores and Standard Deviations for Fourth Grade WASL
Scale Scores as a Function of Group (CSCP vs. no CSCP), Enrollment,
and Gender

           Enrolled in    Student    Vocabulary         Comprehension
Group      Building       Gender     M         SD       M         SD

CSCP       This year      Male       401.83    54.82    396.89    18.42
                          Female     395.88    57.29    402.48    18.14
           1 year ago     Male       408.82    49.69    400.55    17.62
                          Female     407.88    58.95    406.34    18.12
           2 years ago    Male       410.06    49.81    399.58    17.51
                          Female     408.58    55.21    404.93    19.48
           3 years ago    Male       419.90    44.89    404.40    18.84
                          Female     416.59    59.21    409.00    18.74
           4 years ago    Male       418.23    48.65    404.84    17.08
                          Female     417.26    51.98    409.42    17.38
No CSCP    This year      Male       410.71    51.65    401.58    17.78
                          Female     410.22    55.64    406.53    17.52
           1 year ago     Male       416.78    52.41    404.07    19.29
                          Female     415.17    56.91    408.46    19.17
           2 years ago    Male       416.01    54.71    404.55    18.47
                          Female     410.13    56.12    407.67    18.67
           3 years ago    Male       411.15    47.67    403.97    16.49
                          Female     414.89    54.32    409.96    19.08
           4 years ago    Male       420.31    47.46    405.85    17.78
                          Female     420.25    53.42    410.37    18.15

           Enrolled in    Student    Reading         Math
Group      Building       Gender     M       SD      M         SD

CSCP       This year      Male       7.10    2.21    382.77    34.86
                          Female     7.92    2.07    383.38    36.28
           1 year ago     Male       7.30    2.01    387.69    31.36
                          Female     8.32     210    389.58    32.91
           2 years ago    Male       7.10    2.07    383.45    33.24
                          Female     8.31    2.17    391.16    36.27
           3 years ago    Male       7.69    2.31    395.80    34.03
                          Female     8.43    1.87    395.51    34.64
           4 years ago    Male       7.75    2.08    396.01    33.02
                          Female     8.53    1.97    397.55    33.38
No CSCP    This year      Male       7.46    2.08    389.08    34.73
                          Female     8.17     205    390.31    33.64
           1 year ago     Male       7.61    2.06    393.77    34.81
                          Female     8.53    2.05    393.82    36.02
           2 years ago    Male       7.69    2.12    395.28    35.70
                          Female     8.32    2.01    392.90    34.12
           3 years ago    Male       7.64    1.98    396.31    32.72
                          Female     8.60    1.93    396.82    35.69
           4 years ago    Male       7.75    2.05    397.37    32.81
                          Female     8.66    1.94    397.57    34.56

Table 3. Mean Scores and Standard Deviations for Third Grade ITBS
Standard Scores as a Function of Group (high CSCP vs. no CSCP),
Enrollment, and Gender

           Enrolled in    Student    Vocabulary         Comprehension
Group      Building       Gender     M         SD       M         SD

High       This year      Male       178.91    20.68    180.02    20.20
CSCP                      Female     181.70    19.50    184.55    19.88
Use        1 year ago     Male       183.51    21.21    184.52    20.02
                          Female     185.63    20.78    185.71    20.53
           2 years ago    Male       189.50    17.05    190.29    18.11
                          Female     186.85    16.53    188.90    20.81
           3 years ago    Male       192.00    18.13    193.19    20.57
                          Female     189.82    18.57    192.55    19.56
No CSCP    This year      Male       184.68    20.97    185.43    22.50
                          Female     187.24    21.43    188.71    21.17
           1 year ago     Male       183.93    19.88    184.98    19.80
                          Female     184.92    18.68    189.08    20.75
           2 years ago    Male       187.79    19.38    187.79    21.89
                          Female     187.65    19.93    189.54    20.47
           3 years ago    Male       188.59    18.95    189.05    21.16
                          Female     189.04    19.83    191.24    20.43

           Enrolled in    Student    Reading            Math
Group      Building       Gender     M         SD       M         SD

High       This year      Male       179.41    19.40    186.08    18.76
CSCP                      Female     183.09    18.25    184.20    17.86
Use        1 year ago     Male       184.03    19.23    189.56    18.77
                          Female     185.65    19.83    191.23    18.82
           2 years ago    Male       190.09    15.75    191.98    18.37
                          Female     187.86    17.37    189.48    16.32
           3 years ago    Male       192.60    17.97    198.95    17.79
                          Female     191.10    17.98    191.93    18.08
No CSCP    This year      Male       185.03    20.46    189.01    18.97
                          Female     187.93    20.05    187.99    18.94
           1 year ago     Male       184.45    18.54    190.74    17.97
                          Female     186.97    18.37    188.02    18.32
           2 years ago    Male       187.76    19.47    192.03    18.76
                          Female     188.54    18.87    190.33    17.98
           3 years ago    Male       188.82    18.73    193.54    17.29
                          Female     190.14    18.72    190.20    17.90

Table 4. Mean Scores and Standard Deviations for Fourth Grade WASL
Scale Scores as a Function of Group (high CSCP vs. no CSCP),
Enrollment, and Gender

           Enrolled in    Student    Vocabulary         Comprehension
Group      Building       Gender     M         SD       M         SD


High       This year      Male       412.58    55.88    400.10    20.22
CSCP                      Female     389.60    58.26    400.10    19.15
Use        1 year ago     Male       402.83    48.09    399.46    16.19
                          Female     401.60    54.95    406.11    15.19
           2 years ago    Male       421.60    46.12    398.53    19.82
                          Female     414.68    53.45    408.31    19.47
           3 years ago    Male       427.47    47.22    406.29    19.96
                          Female     426.05    55.46    410.69    16.01
           4 years ago    Male       424.95    47.80    408.17    16.26
                          Female     423.36    49.84    413.10    16.42
No CSCP    This year      Male       410.71    51.65    401.58    17.78
                          Female     410.22    55.64    406.53    17.52
           1 year ago     Male       416.78    52.41    404.07    19.29
                          Female     415.17    56.91    408.46    19.17
           2 years ago    Male       416.01    54.71    404.55    18.47
                          Female     410.13    56.12    407.67    18.67
           3 years ago    Male       411.15    47.67    403.97    16.49
                          Female     414.89    54.32    409.96    19.08
           4 years ago    Male       420.31    47.46    405.85    17.78
                          Female     420.25    53.42    410.37    18.15

           Enrolled in    Student    Reading         Math
Group      Building       Gender     M       SD      M         SD

High       This year      Male       7.66    2.09    390.76    32.99
CSCP                      Female     8.04    2.18    383.88    37.87
Use        1 year ago     Male       7.51    1.84    389.84    28.11
                          Female     8.60    1.72    392.79    29.34
           2 years ago    Male       7.29    2.17    382.76    37.52
                          Female     8.79    2.18    395.74    38.20
           3 years ago    Male       8.17    2.24    403.98    32.97
                          Female     8.56    1.75    401.75    32.90
           4 years ago    Male       8.11    1.93    404.60    29.74
                          Female     9.02    1.90    404.98    32.82
No CSCP    This year      Male       7.46    2.08    389.08    34.73
                          Female     8.17    2.05    390.31    33.64
           1 year ago     Male       7.61    2.06    393.77    34.81
                          Female     8.53    2.05    393.82    36.02
           2 years ago    Male       7.69    2.12    395.28    35.70
                          Female     8.32    2.01    392.90    34.12
           3 years ago    Male       7.64    1.98    396.31    32.72
                          Female     8.60    1.93    396.82    35.69
           4 years ago    Male       7.75    2.05    397.37    32.81
                          Female     8.66    1.94    397.57    34.56

Table 5. Descriptive Statistics for Grade 3 ITBS Standard Scores as a
Function of Group (high CSCP vs. no CSCP) With Only Continuously
Enrolled Students

                    Vocabulary       Comprehension    Reading
Group               M        SD      M        SD      M        SD

High CSCP Use (a)   191.01   18.34   192.90   20.19   191.92   17.98
No USCP (b)         188.81   19.38   190.11   20.83   189.46   18.73

                    Math
Group               M        SD

High CSCP Use (a)   195.77   18.24
No USCP (b)         1.91.92  17.66

Note. Continuously enrolled students were those participants who
attended the same school for three or more years.

(a) N = 552 (b) N = 1809

Table 6. Descriptive Statistics for Grade 4 WASL Scale Scores as a
Function of Group (high CSCP vs. no CSCP) With Only Continuously
Enrolled Students

                     Listening          Reading            Writing
Group                M         SD       M         SD       M       SD

High CSCP Use (a)    424.80    49.48    410.12    16.93    8.52    1.98
No CSCP (b)          418.63    50.66    407.79    18.10    8.17    2.04

                     Math
Group                M         SD

High CSCP Use (a)    404.30    31.68
No CSCP (b)          397.26    33.78

Note. Continuously enrolled students were those participants who
attended the same school for three or more years.

(a) N = 786 (b) N = 2471

The authors sincerely appreciate the assistance they received from the Washington School Research Center and from Rich Lapan (University of Missouri, Columbia) and Ken Hughey (Kansas State University).

References

Adelman, H. S., & Taylor, L. (2002). School counselors and school reform: New directions. Professional School Counseling, 5, 235-244.

American School Counselor Association. (1997). The professional school counselor and comprehensive school counseling programs. Retrieved August 6, 2001, from http://www.schoolcounselor.org/pdf/counsel.pdf

American School Counselor Association. (1999). The role of the professional school counselor. Retrieved August 6, 2001, from http://www.schoolcounselor.org/role.htm

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

Bemak, F. (2000). Transforming the role of the counselor to provide leadership in educational reform through collaboration. Professional School Counseling, 3, 323-331.

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

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

Carns, A. W., & Carns, M. R. (1991). Teaching study skills, cognitive strategies, and metacognitive skills through self-diagnosed learning styles. The School Counselor, 38, 341-346.

Clark, M., & Stone, C. (2000). The developmental school counselor as educational leader. In J. Wittmer (Ed.), Managing your school counseling programs: K-12 developmental strategies (2nd ed., pp. 85-81). Minneapolis, MN: Educational Media.

Coats, R. Q., Ash, M., & Dorsey, C.J.H. (1998). Washington state guidelines for comprehensive counseling and guidance programs from kindergarten through community and technical college. Olympia, WA: Office of the Superintendent of Public Instruction.

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

Dahir, C. A. (2001). The national standards for school counseling programs: Development and implementation. Professional School Counseling, 4, 320-327.

Dahir, C. A., Sheldon, C. B., & Valiga, M. J. (1998). Vision into action: Implementing the national standards for school counseling programs. Alexandria, VA: American School Counselor Association.

Erford, B.T., House, R., & Martin, P. (2003). Transforming the school counseling profession. In B.T. Erford (Ed.), Transforming the school counseling profession (pp. 1-20). Upper Saddle River, NJ: Merrill Prentice Hall.

Evans, J. H., Jr., & Burck, H. D. (1992). The effects of career education interventions on academic achievement: A meta-analysis. Journal of Counseling and Development, 71, 63-68.

Gall, M. D., Gall, J. P, & Borg, W. R. (2003). Educational research: An introduction (7th ed.). New York: Longman.

Glosoff, H. L., & Koprowicz, C. L. (1990). Children achieving potential: An introduction to elementary school counseling and state-level policies. Alexandria, VA: American Association for Counseling and Development and National Conference of State Legislatures (Denver, CO). (ERIC Document Reproduction Service No. ED330917)

Green, A., & Keys, S. G. (2001). Expanding the developmental school counseling paradigm: Meeting the needs of the 21st century student. Professional School Counseling, 5, 84-95.

Green, S. B., & Salkind, N. J. (2003). Using SPSS for Windows: Analyzing and understanding data (3rd ed.). Upper Saddle River, NJ: Prentice-Hall.

Grimm, L. G., & Yarnold, P. R. (1995). Reading and understanding multivariate statistics. Washington, DC: American Psychological Association.

Gysbers, N. C. (2001). School guidance and counseling in the 21st century: Remember the past into the future. Professional School Counseling, 5, 96-105.

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

Gysbers, N. C., & Henderson, R (2001). Comprehensive guidance and counseling programs: A rich history and a bright future. Professional School Counseling, 4, 246-256.

Hadley, H. R. (1988). Improving reading scores through a self-esteem intervention program. Elementary School Guidance and Counseling, 22, 248-252.

Hancock, T. E., & Stock, W. A. (1996). Gender and developmental differences in the academic study behaviors of elementary school children. Journal of Experimental Education, 65, 18-39.

Hay, I., Ashman, A. F., & Van Kraayennoord, C. E. (1998). The influence of gender, academic achievement and non-school factors upon pre-adolescent self-concept. Educational Psychology, 18, 461-470.

Henderson, P., & Gysbers, N. C. (1998). Leading and managing your school guidance program staff: A manual for school administrators and directors of guidance. Alexandria, VA: American Counseling Association.

Herr, E. L. (2002). The impact of national policies, economics, and school reform on comprehensive guidance programs. Professional School Counseling, 4, 236-245.

Hoover, H. D., Hieronymus, A. N., Frisbie, S. D., & Dunbar, S. B. (1996). Norms and score conversions with technical information: Iowa Tests of Basic Skills. Itasca, IL: Riverside.

House, R. M., & Hayes, R. L. (2002). School counselors becoming key players in school reform. Professional School Counseling, 5, 249-256.

House, R. M., & Martin, P. J. (1998). Advocating for better futures for all students: A new vision for school counselors. Education, 119, 284-291.

Hughes, D. K., & James, S. H. (2001). Using accountability data to protect a school counseling program: One counselor's experience. Professional School Counseling, 4, 306-309.

Hughey, K., Gysbers, N. C., & Starr, M. (1993). Evaluating comprehensive school guidance programs: Assessing the perceptions of students, parents, and teachers. The School Counselor, 41, 31-35.

Johnson, S. K., & Johnson, C. D. (2003). Results-based guidance: A systems approach to student support programs. Professional School Counseling, 6, 180-184.

Johnson, S. K., & Whitfield, E. A. (Eds.). (1991). Evaluating guidance programs: A practitioner's guide. Iowa City, IA: The American College Testing Program and the National Consortium of State Career Guidance Supervisors.

Keys, S. G., Bemak, R, & Lockhart, E. J. (1998). Transforming school counseling to serve the mental health needs of at-risk students. Professional School Counseling, 2, 381-388.

Keys, S., Green, A., Lockhart, E., & Luongo, R (2003). School consultation. In B. Erford (Ed.), Transforming the school counseling profession (pp. 171-190). Saddle River, NJ: Merrill Prentice-Hall.

Keys, S. G., & Lockhart, E. J. (1999). The school counselor's role in facilitating multisystemic change. Professional School Counseling, 3, 101-107.

Kinsler, K., & Gamble, M. (2001). Reforming schools. London: Continuum.

Lapan, R.T. (2001). Results-based comprehensive guidance and counseling programs: A framework for planning and evaluation. Professional School Counseling, 4, 289-299.

Lapan, R.T., Gysbers, N. C., Hughey, K., & Arni, T. J. (1993). Evaluating a guidance and language arts unit for high school students. Journal of Counseling and Development, 71, 444-451.

Lapan, R.T., Gysbers, N. C., Multon, K. D., & Pike, G. R. (1997). Developing guidance competency self-efficacy scales for high school and middle school students. Measurement and Evaluation in Counseling and Development, 30, 4-17.

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

Lapan, R.T., Kardash, C. M., & Turner, S. (2002). Empowering students to become self-regulated learners. Professional School Counseling, 5, 257-265.

Lee, R. S. (1993). Effects of classroom guidance on student achievement. Elementary School Guidance and Counseling, 27, 163-171.

Lehr, R., & Sumarah, J. (2002). Factors impacting the successful implementation of comprehensive guidance and counseling programs in Nova Scotia. Professional School Counseling, 5, 292-297.

MacDonald, G., & Sink, C. A. (1999). A qualitative developmental analysis of comprehensive guidance programmes in schools in the United States. British Journal of Guidance and Counselling, 27, 415-430.

Masters, G. N. (1982). A Rasch model for partial credit scoring. Psychometrika, 47,149-174.

McNeil, L M. (2000). Contradictions of school reform: Educational costs of standardized testing. New York: Routledge.

Myrick, R. D. (2003a). Accountability: Counselors count. Professional School Counseling, 6, 174-179.

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

National Commission on Excellence in Education. (1983). A nation at risk: The imperative for educational reform. Washington, DC: U.S. Government Printing Office.

Nelson, D. E., & Gardner, J. L. (1998). An evaluation of the comprehensive program in Utah public schools. Salt Lake City, UT: Utah State Office of Education.

Paisley, P. O., & Hayes, R. L. (2003). School counseling and the academic domain: Transformations in preparation and practice. Professional School Counseling, 6, 198-204.

Pomerantz, E. M., Altermatt, E. R., & Saxon, J. L. (2002). Making the grade but feeling distressed: Gender differences in academic performance and internal distress. Journal of Educational Psychology, 94, 396-404.

Ripley, V., Erford, B. T., Dahir, C., & Eschbach, L. (2003). Planning and implementing a 21st century comprehensive developmental school counseling program. In B.T. Erford (Ed.), Transforming the school counseling profession (pp. 63-119). Upper Saddle River, NJ: Merrill Prentice Hall.

Riverside Publishing. (1997). Validity information for the Iowa Tests of Basic Skills (ITBS) and Iowa Tests of Educational Development (ITED): Forms K, L, M. Itasca, IL: Author.

Rowley, W. J. (2001). Expanding collaborative partnerships among school counselors and school psychologists. Professional School Counseling, 3, 315-322.

Schmidt, J. J. (2000). Counselor accountability: Justifying your time and measuring your worth. In J. Wittmer (Ed.), Managing your school counseling programs: K- 12 developmental strategies (2nd ed., pp. 273-291). Minneapolis, MN: Educational Media.

Schwallie-Giddis, P., ter Maat, M., & Pak, M. (2003). Initiating leadership by introducing and implementing the ASCA National Model. Professional School Counseling, 6, 170-173

Sink, C. A. (2002). In search of the profession's finest hour: A critique of four views of 21st century school counseling. Professional School Counseling, 5, 156-163.

Sink, C. A., & MacDonald, G. (1998). The status of comprehensive guidance and counseling in the United States. Professional School Counseling, 2, 88-94.

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

Tabachnick, B. G., & Fidell, L. S. (2000). Using multivariate statistics (4th ed.). New York: Harper Collins.

Taylor, C. S. (2000). Washington assessment of student learning, grade 4, 1999, technical report. Olympia, WA: Office of the Superintendent of Public Instruction.

Thompson, R. A. (2002). School counseling: Best practices for working in the schools. New York: Brunner-Routledge.

United States Congress. (1994). Goals 2000: Educate America Act. Washington, DC: U.S. Government Printing Office.

Vanzandt, Z., & Hayslip, J. B. (2001). Developing your school counseling program: A handbook for systemic planning. Belmont, CA: Wadsworth.

Whiston, S. C. (2003). Outcomes research on school counseling services. In B.T. Erford (Ed.), Transforming the school counseling profession (pp. 435-447). Upper Saddle River, NJ: Merrill Prentice Hall.

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

Wittmer, J. (2000a). Developing school guidance and counseling: Its history and reconceptualization. In J. Wittmer (Ed.), Managing your school counseling programs: K-12 developmental strategies (2nd ed., pp. 2-13). Minneapolis, MN: Educational Media.

Wittmer, J. (2000b). Implementing a comprehensive developmental school counseling program. In J. Wittmer (Ed.), Managing your school counseling programs: K-12 developmental strategies (2nd ed., pp. 14-34). Minneapolis, MN: Educational Media.

The results of this study were first presented at the 2002 annual conference of the British Educational Research Association (University of Exeter, UK). An earlier version of the study is available as a research report through the Washington School Research Center (http://www.spu.edu/orgs/research/index.html).

Christopher A. Sink, PhD, NCC, LMHC, is professor and chair, Department of School Counseling and Psychology at Seattle Pacific University, WA. E-mail: [e-mail]

Heather R. Stroh is a researcher with the Washington School Research Center, Lynnwood, WA, and a doctoral student in school counseling at Seattle Pacific University.