The School Counselor Caseload and the High School-to-College Pipeline
by Chenoa S. Woods & Thurston Domina - 2014
Background: Advising students on the transition from high school to college is a central part of school counselorsí professional responsibility. The American School Counselor Association recommends a school counselor caseload of 250 students; however, prior work yields inconclusive evidence on the relationship between school counseling and school-level counseling resources and studentsí college trajectories.
Focus of Study:This study evaluates the relationship between access to school counselors and several critical indicators of student transitions between high school and college.
Research Design: The study utilizes the Education Longitudinal Study of 2002 to explore the relationships between the school counselor caseload and studentsí progress throughout the high school-to-college pipeline. The key indicator is the counselor caseload for students at a given high school, measured as the number of 10th graders per counselor at the high school at which each student is enrolled. The outcome variables are studentsí college expectations, whether students spoke with a counselor about college, taking the SAT, and college enrollment. Logistic and multinomial logistic regression analyses are applied to examine the relationships between these variables.
Findings: Students in schools with small counselor caseloads enjoy greater success at navigating the high school-to-college pipeline. Controlling for student- and school-level characteristics, students in schools where counselors are responsible for advising a large number of students are less likely to speak with a counselor about college, plan to attend college, take the SAT, and enroll in a four-year college.
Conclusions: The findings support the claim that a smaller school counselor caseload may increase studentsí access to key college preparation resources and raise four-year college enrollment rates.
Although college attendance rates have increased greatly in the past several decades, issues of equity in the college preparation and admissions process persist. A large majority of high school students expect to attend college, yet fewer students enroll, and even fewer students graduate with a baccalaureate degree (e.g., Solorzano, Villalpando, & Oseguera, 2005). The road from high school to college is particularly perilous for students from families with limited college experience, because many of these students lack access to information about college entrance requirements, selection criteria, and cost (Kirst & Venezia, 2004). Without this critical information, it is difficult for students to realize their college expectations, and even high-achieving students fail to make the transition from high school to a postsecondary educational institution (Plank & Jordan, 2001).
High school counselors have many duties, including providing students with informational and other assistance in college preparation and attendance. The American School Counselor Association (ASCA), which recommends that high schools maintain a ratio of 250 students per counselor, suggests that the role of school counselors includes responding to students academic, career, and personal/social development. High school counselors face the task of guiding students through adolescence into early adulthood, preparing them to be future citizens, college students, employees, and parents; they implement their counseling program by providing classroom guidance, individual student planning, responsive services, and by collaborating with other educators and community members (ASCA, 2012c, 2012d).
Although counselors are charged with the task of equitably serving all students, many scholars argue that counselors in schools that serve underrepresented students are often unable to effectively advise students because their caseloads are too large (Cabrera & La Nasa, 2000; Corwin, Venegas, Oliverez, & Colyar, 2004; Farmer-Hinton, 2008; Farmer-Hinton & McCullough, 2008; González, Stoner, & Jovel, 2003; Hamrick & Stage, 2004; Lee & Ekstrom, 1987; McDonough, 1997, 2005; Muhammad, 2008; Perna et al., 2008; Rosenbaum, Stephan, & Rosenbaum, 2010). The counselor caseload, sometimes referred to as the studentschool counselor ratio, is usually measured by the number of students in a school divided by the number of counselors in a school. Many counselors manage caseloads double or triple ASCAs recommended caseload. Nationally, the counselor caseload at the average high school is 471 students, and in California, the state with the largest caseload, counselors work with more than 1,000 students (ASCA, 2012b). Reasoning that manageable school counselor caseloads may facilitate more efficient counseling, several states, including Indiana, Maine, and New Hampshire have passed legislation limiting the size of counselor caseloads (ASCA, 2012a). Nevertheless, with caseloads varying in size it is apparent that access to individualized college counseling is not available to all.
Lee, Edwards, Menson, and Rawis (2011) provide a slightly better picture, stating that as of 2008 the student-to-college counselor ratio was 320:1. College counselors are those who are responsible for providing college counseling, either as their sole responsibility or in combination with other counseling tasks. Although the size of average caseloads has decreased by 49 students between 1998 and 2009, most caseloads remain well above the recommended ratio of 250:1 (Clinedinst & Hawkins, 2011; Lee et al., 2011).
The purpose of this paper is to examine the relationship between students access to college guidance, as measured by the size of the school counselor caseload, and college-going behaviors. Situated in the hypothesis that a more manageable school counselor caseload leads to additional opportunities for students to build social capital and progress through the high school-to-college transition, this study uses the Educational Longitudinal Study of 2002 (ELS:2002) to address how the size of the counselor caseload affects students postsecondary educational expectations, preparation, and attendance patterns. The research questions are as follows:
What are the relationships between the counselor caseload and students college planning, information-gathering, and college-preparatory behavior in 10th and 12th grade?
What are the relationships between the counselor caseload and students college enrollment patterns?
THE INFLUENCE OF SCHOOL COUNSELORS ON HIGH SCHOOL-TO-COLLEGE TRANSITIONS
The high-school-to-college pipeline begins as students develop their college expectations, long before students enroll in college. Perna (2006) presents a model of college choice, which draws from both economic and sociological approaches. This model has four layers: habitus (Layer 1); school and community context (Layer 2); higher education context (Layer 3); and social, economic, and policy context (Layer 4). The outer layers inform the inner layers, which then influence college choice. This study focuses primarily on the school and community context, but it stands to reason that the higher education context and macrolevel influences have real impacts for counselors practices in schools. For example, universities changing admissions policies (Level 3) and state funding for school counselor positions (Level 4) can impact the extent to which school counselors can engage in precollege counseling. The school environment helps to develop ones college-going habitus, or their beliefs and worldview, which influence their attitudes, goals, and in this case, educational expectations (McDonough, 1997).
Counselors can potentially play an important role as students develop their educational plans and prepare for life after high school. They can influence students college predispositions, either by encouraging students to consider college and providing information or by acting as gatekeepers and redirecting students with unrealistic educational expectations away from higher education. Counselors can choose the amount and types of college and financial aid information to provide, effectively channeling students towards or away from different types of colleges (Freeman, 1997).
Stanton-Salazar and Dornbusch (1995) describe counselors as institutional agents: people who not only have the ability, but also have the dedication to communicate institutional resources and opportunities (p. 117). One way that counselors do so is by influencing students college-going habitus; primarily shaped by the individuals background and socialization, habitus defines a students view of his or her realistic possibilities (Massé, Perez, & Posselt, 2010). Counselors can influence students habitus directly, using their face-to-face interactions to encourage students to formulate high expectations and provide relevant information. Additionally, counselors can shape the college-going culture of the school itself, creating an institutional-level atmosphere that supports college preparation and attendance for all students (Jarsky, McDonough & Nunez, 2009; McDonough, 1997).
In addition, counselors may increase students social capital. Coleman (1988) defines social capital as relationships that lead to action. When a student develops a relationship with a school counselor, he or she gains access to college knowledge, resources, and advice, which can play an important role in shaping a students future. Social capital factors such as encouragement from others or assistance on college applications are closely linked to students likelihood of attending a four-year college; and for black and Latino students, these social capital factors are as important in predicting college attendance as academic ability (Perna, 2000). Counselors may have more influence in fostering college expectations for first-generation college students and students from low socioeconomic backgrounds compared to their peers with more college-related social and cultural capital (McDonough, 1997). However, negative interactions between students and school counselors can dissuade students from approaching counselors for guidance, thus limiting students access to the potential social and cultural capital counselors hold. Additionally, the degree to which school counselors are able to engage in quality precollege counseling and act as positive institutional agents may be mitigated by larger school-, district-, or state-level regulations or resource limitations.
To effectively influence students habitus and social capital, counselors need to be involved consistently with students throughout their educational careers. Farmer-Hinton and McCullough provide a particularly evocative description of this process, reporting that counselors at a college-prep charter school infuse college by engaging students in informal, yet repetitive conversations about college (2008, p. 85). Accordingly, the relevant literature generally assumes that counselors use their influence to broaden college access, and working with counselors will have positive effects at each stage of the college choice process (Bryan, Moore-Thomas, Day-Vines, & Holcomb-McCoy, 2011). These effects may be especially pronounced for students who have few other sources of college knowledge, (Eccles, Vida, & Barber, 2004; Freeman, 1997) because these students are more dependent on their school counselor to help them receive insider information and guide them through the application process (McDonough, 1997, p. 101; Rosenbaum, Miller, & Krei, 1996).
However, counselors may also play a gate-keeping role, providing concrete encouragement and information to students that they view as college material and withholding information from students that they believe are less likely to succeed in higher education. Studies of students in college prep programs, magnet schools, and charter schools that are organized around college preparatory missions provide some evidence to support this claim (Lee & Eckstrom, 1987). When students enter into these programs and complete the advanced courses that are required for selective college admissions, they may signal their commitment to higher education to counselors, creating a continual loop in which counselors provide encouragement and information, which in turn helps students in these programs advance toward higher education (Farmer-Hinton & McCullough, 2008; Tuitt, Van Horn, & Sulik, 2011). Honors track students report that their counselors knew all of us by name (González et al., 2003, p. 162). In these settings, counselors regularly provide students detailed information about scholarships, internships, and jobs as well as guidance on where to apply and attend college (Freeman, 1997). On the other hand, lower-achieving students report limited access to college resources (Kimura-Walsh, Yamamura, Griffin, & Allen, 2009).
THE CONSEQUENCES OF HIGH COUNSELOR CASELOADS
Encouraging students to develop college expectations and advising students as they progress through the high school-to-college transition process is time-consuming work, particularly for counselors working with students who have access to few other reliable sources of information about college. Counselors faced with large student caseloads, therefore, may find it especially difficult to effectively assist all students. In such a setting, counselors may be more likely to serve a gate-keeping role as they make strategic decisions about which students to devote college advising time and energy toward (Corwin et al., 2004; González et al., 2003; Kimura-Walsh et al., 2009). If this is the case, school-level counselor caseloads may be negatively associated with student college expectations, access to information and, ultimately, students odds of enrolling in higher education.
One explanation for providing students with limited college counseling is that, generally, counselors tend to wear many hats and have a diverse range of responsibilities. Public high school counselors report relatively equal amounts of time on counseling for preparing for postsecondary education, high school course scheduling, and personalsocial counseling, and an additional 14% of their time is allotted to testing (Clinedinst & Hawkins, 2011). School counselors are responsible for a variety of tasks including nonguidance activities and counselors role confusion may lead to discrepancies between accepted counseling models and counselors actual duties (Burnham & Jackson, 2000; Lee et al., 2011).
Counselors work with a diverse group of students who may be gifted, special needs, low income, English language learners, immigrants or children of immigrants, and/or of another population requiring individualized guidance. Counselors report that a wide range of tasks compete with college counseling for their time, including individual counseling; group counseling; classroom guidance; consultation with community agencies; parent groups; support, prevention and intervention strategies; and the administration of diagnostic and other standardized tests, thus some admit to serving only portions of the population, often relatively at-risk students (Burnham & Jackson, 2000; Perna et al., 2008; Sears, 1993). These findings vary with enrollment and the size of the counselor caseload and have changed over time. Counselors in schools with caseloads of 500 students or more spend substantially less time on college counseling than counselors in smaller schools and schools with smaller caseloads (Clinedinst & Hawkins, 2011). Nationally, school counselors time spent on college advising decreased between 2006 and 2009 (Lee et al., 2011).
Counselors in schools with unmanageably large student caseloads may focus their energy on advising students at one point in the college choice process at the expense of other points. There is some evidence to suggest that overburdened counselors reserve counseling resources for upperclassmen, leaving fewer resources (i.e., time) for 9th and 10th grade students (Corwin et al., 2004). To the extent to which counselors respond to large caseloads by adopting such an approach, we would expect caseloads to be particularly negatively associated with students odds of meeting with counselors in the early years of high school and their odds of forming early college predispositions.
On the contrary, Rosenbaum et al. (1996), argue that counselors articulate a college-for-all ethos and encourage all students to formulate college expectations, regardless of their assessment of students ability to realize these expectations. Small counselor caseloads may be an indicator of a college-going culture on campus, raising all students educational expectations. Nevertheless, high expectations alone do not communicate to students the details of how and when to prepare for college; professionals are needed to guide students through the relevant college information. While this strategy results in universally high expectations, Rosenbaum et al. argue that counselors do not provide students with adequate information and preparation for success in higher education. As a result, they argue, many students move from high schools to nonselective community colleges, where they lack the skills to succeed academically. Rosenbaum et al.s argument implies that counselor caseloads should correlate negatively with students odds of attending two-year colleges but not their odds of four-year college enrollment.
Despite these conflicting hypotheses, we currently have very little information on the relationship between counselor caseload and student transitions between high school and college. Carrell and Carrell (2006) demonstrate that large caseloads undermine the extent to which counselors can prevent disciplinary problems in schools. However, this analysis provides no evidence about college outcomes. Engberg and Wolniak (2010), meanwhile, examine the effects of counselor caseloads and other indicators of school learning environments on students two- and four-year college enrollment. This study finds that studentcounselor ratios have no direct effect on college outcomes. However, Engberg and Wolniaks analyses include controls for student course taking and social capital. If these factors are mechanisms through which caseloads influence college outcomes, this study may underestimate the relationship between counselor caseloads and student college going.
DATA SOURCE AND MEASURES
We use the ELS:2002 to explore the impact of the school counselor caseload on students educational expectations and their progress throughout the high school-to-college pipeline. The ELS:2002 is based on a stratified sampling design. In the first stage, a sample of 750 schools was drawn from 1,220 eligible schools. Sampled schools provided a roster of their 10th grade students from which the study sampled approximately 30 students from each school. In spring of 2002, when students were sophomores in high school, all sampled students completed paper and pencil questionnaires and standardized mathematics and reading tests. In addition, the study collected transcripts, course catalogues, and information about educational resources and school security from administrators at all sampled high schools. In 2004, when most students were high school seniors, all participants were resurveyed and retested. In 2006, two years after most students have graduated from high school, students were again resurveyed to provide information about their postsecondary experiences (Ingels et al., 2007).
Our analyses use these data to investigate the relationship between counseling resources in students high schools and their progress on the high school-to-college pipeline. Each of these analyses takes the following general form:
In this equation, by indicates that the variable was taken from the base year; f1 indicates that the variable was taken from the first follow-up, and f2 indicates that the variable was taken from the second follow-up. The CASELOAD variable is the standardized counselor caseload and CASELOAD2 is the squared version of this variable. The STUCNTRL term includes the student and family background variables, and the SCHCTRL term captures schools urbanicity.
SCHOOL COUNSELOR CASELOAD
The key indicator in this study is the counselor caseload. This figure, which represents the number of 10th graders per counselor in surveyed students schools, was created by taking the 10th grade enrollment variable (BYG10ER) and dividing it by the number of school counselors employed by a school (BYA23K):
For example, if there are 300 10th grade students in the school and 5 counselors, the studentcounselor ratio is 60:1 or 60 students per counselor. The 10th grade enrollment and school counselor count variables are both reported in the school administrators questionnaire in the base year and the caseload variable is standardized in the regression equations. Note that this variable only accounts for 10th grade students, resulting in smaller ratios than what may be expected if calculated using a school-wide enrollment variable. We maintain that this is the best way to calculate the counselor caseload, given that other measures of school enrollment in ELS: 2002 are not continuous. A squared version of the counselor caseload variable is included to account for nonlinear relationships.
College expectations and plans are measured in 10th and 12th grade, respectively. Two outcome variables, having expectations to attend a postsecondary educational institution (PEI; 10th grade; BYSTEXP) and having plans to attend college directly after high school (12th grade; F1S45) are examined. These dummy variables are coded as one if the student has expectations/plans for college attendance or zero if the student does not have expectations or does not know if he or she plans to attend college; a code of one includes any amount of college attendance, ranging from some college through advanced degree attainment.
COLLEGE INFORMATION AND READINESS
Speaking with a school counselor about college admissions information is one way students gain college knowledge. Logistic regression (Cabrera, 1994) explores the impact of the counselor caseload for whether students discuss college admissions information with counselors in the base year and in the first follow-up (BYS59B and F1S48A, respectively). Dummy variables for 10th and 12th grade are coded as one if the student has spoken with a school counselor about college at least once. The next step along the education pipeline is information gathering and college preparation. Plank and Jordan (2001) determined that preparing for, receiving guidance about, and taking the SAT/ACT increases students chances of enrolling in a PEI. This dummy variable is set equal to one if the student has taken the SAT/ACT by 12th grade (F1S21C).
Finally, we examine the relationship between the counselor caseload and students college attendance. Specifically, multinomial logistic regression assesses the relationship between the caseload and the type of PEI attended (a two-year college or a four-year college as compared to not attending a degree-granting institution; F2PS1LVL). This outcome is measured in the third wave, or two years after high school.
DEMOGRAPHIC AND CONTROL VARIABLES
All multivariate models include controls for students gender and race, their familys socioeconomic background, their prior academic achievement, and schools urbanicity. We operationalize gender as a dummy variable, based on the categorical ELS: 2002 BYSEX variable. Similarly, we operationalize race as a series of dummy variables. Using data reported in BYRACE, we categorize students as African American, white, Hispanic, Asian, or other race. For ease of interpretation we use the modal categorywhiteas the omitted category in all analyses (Hardy, 1993). We measure students family background using maternal educational attainment (BYMOTHED) and family income (BYINCOME). Maternal education is a series of dummy variables with some college as the reference; income is a standardized continuous variable. Students standardized math (BYTXMSTD) and reading (BYTXRSTD) scores are included as standardized continuous variables in the model to account for past academic performance. School urbanicity (BYURBAN) is included as a dummy variable (with suburban schools as the reference category) to account for schools local context.
For the purposes of this study, the sample is restricted to public school students who did not transfer schools between the base year and the first follow-up and who have nonmissing counselor caseload data (N = 7,800). Due to transferring schools, attending private schools, and/or missing data for the counselor caseload variable, 8,397 students were dropped from the original sample. Despite small differences in the current sample and the full sample, there were no significant differences between the two groups on any of the dependent variables. Including students who had transferred schools may result in including students who were enrolled in schools with different counselor caseloads between the first and second wave of data collection, and restricting the sample in this way holds the counselor caseload constant over the first two waves.
We impute data for cases missing data on one or more control variables using Markov chain Monte Carlo multiple imputation. All analyses were conducted using logistic regression and multinomial logistic regression techniques and were run only for students who have data on the counselor caseload variable and the dependent variable of interest, resulting in different sample sizes depending on the equation. Data are weighted for representativeness to allow for generalizing to the national population. For analyses using only base-year variables, the base-year student probability weight (BYSTUWT) is used, for analyses using variables from the first and second wave the first-year/follow-up panel weight (F1PNLWT) is used, and for analyses using variables from the first and third wave, second follow-up/base-year probability weight is used (F2BYWT).
Our preliminary analyses describe student populations at schools with counselor caseloads of varying sizes. Table 1 separates the counselor caseload into quartiles. The mean counselor caseload for schools with the smallest ratios (the schools at the 25th percentile and below) is 50 students per counselor; the mean for the second quartile is 74 students per counselor; the third quartiles mean is 95 students per counselor; and the mean in the top quartile is 136 students per counselor. White students tend to be concentrated in schools with the smallest caseloads, whereas black and Latino students are concentrated in schools with larger caseloads. Students whose mothers have less than a high school education are more concentrated in the quartile with the largest caseloads, but students whose mothers have an advanced degree represent the smallest proportion of students in the highest quartile. Similarly, students with the lowest family income are much more concentrated in the schools with the largest caseloads. Not surprisingly, schools with the largest enrollment also have the largest counselor caseloads, but the sizeable standard deviations across all four quartiles indicate that schools of various sizes can be found in any of the caseload quartiles.
Table 1 also demonstrates differences on student outcomes for students in the various quartiles. The rate of speaking with a counselor about college in the 10th grade is approximately seven percentage points higher for students in schools with the smallest counselor ratios compared to their peers in schools with the largest caseloads. By 12th grade, that gap expands to 10 percentage points. Similarly, only 59% of students in schools with the largest caseloads take the SAT/ACT, compared to nearly 70% of students in schools with the smallest caseloads. Again, students in schools with the largest caseloads attend 4-year colleges at the lowest rates (39%) compared to students in schools with the smallest caseloads (49%). These findings provide initial descriptive evidence to suggest that students in schools with small counselor caseloads are advantaged in the high school-to-college transition relative to their peers in schools with higher counselor caseloads.
TENTH GRADE OUTCOMES
Table 2 focuses on the relationship between counselor caseloads and 10th graders odds of speaking with a counselor about college and articulating college expectations. This table suggests that there is a significant negative relationship between counselor caseload and students odds of meeting with their counselor prior to 10th grade. This relationship persists even after controlling for student and school background characteristics, suggesting that the relationship is robust. Furthermore, the squared term in both of these models is not significantly different from one, suggesting that the odds of speaking with a counselor decrease incrementally as caseloads increase.
However, the next two models suggest that the associations between counselor caseloads and 10th graders college expectations are much weaker. We recognize that expectations develop through relationships with multiple people and influences that may begin well before high school, yet the number of students school counselors work with does not appear to impact 10th grade students educational expectations. Neither the reduced-form model nor the model with controls for student and school characteristics return a significant relationship between caseload and expectations.
Figure 1 helps to illustrate these findings. The figure represents the predicted probability that students at schools with varying counselor caseloads will speak with a counselor and articulate college expectations in 10th grade, with all controls held constant at the sample mean. Students probability of speaking with counselors declines dramatically as school counselor caseloads increase, pointing to a negative and significant effect of the caseload on students access to conversations with counselors about college. The probability of speaking with counselors for students in schools with the smallest caseloads is nearly 60%, but less than 40% for students in schools with the largest caseloads. By contrast, there is virtually no relationship between counselor caseloads and students college expectations, illustrating the insignificant effect of the counselor caseload on this outcome. In sum, these findings suggest that counselor caseloads influence students odds of speaking with their counselors, but nearly all students expect to attend college, regardless of their access to counselors. This finding is surprising, given scholars attention to the role that counselors can play in forming students college expectations (Krei & Rosenbaum, 2001; McDonough, 1997; Rosenbaum et al., 1996). In light of limited time and resources, school counselors should focus their efforts not solely on encouraging students to attend college but on providing useful assistance about the specific steps students need to take to be adequately prepared.
12TH GRADE OUTCOMES
In Table 3, we examine the relationship between counselor caseloads and 12th grade students odds of speaking with counselors about college, odds of articulating college plans, and odds of taking the SAT or the ACT, a major hurdle to four-year college enrollment. The first set of analyses suggests that there is a robust negative association between counselor caseload and 12th graders odds of speaking with their counselor. Furthermore, this association is much stronger than the association between caseloads and 10th graders odds of talking with their counselors. This finding runs contrary to the idea that overburdened counselors focus their attention on students at the end of their high school careers.
As the counselor caseload increases, students likelihood of planning to continue their education directly following high school decreases after accounting for personal and school demographic controls. In stark contrast to the lack of significant effects of the school counselor caseload on 10th grade students college expectations, 12th grade students plans are significantly negatively associated with an increasing counselor caseload.
Additionally, there is a significant negative association between counselor caseloads and 12th graders odds of completing the SAT or ACT. Each of these relationships remains significant after controlling for student and school background characteristics. Interestingly, the caseload squared term is significant and positive in each of the full models in this table, suggesting that for schools with particularly high counselor caseloads, the negative relationship between caseload and 12th graders college plans and behaviors is somewhat truncated.
Figure 2 represents these relationships graphically, by plotting predicted probabilities that students at schools with various caseload sizes speak with counselors about college and articulate college plans with control variables held at the sample mean. As this figure makes clear, 12th graders in schools with large counselor caseloads are less likely to speak with their counselors than similar students in schools with smaller caseloads, although this relationship flattens somewhat at the top of the caseload distribution. Similarly, the figure indicates that students probabilities of articulating college plans decline as counselor caseloads increase, although this relationship reverses among students in schools with the very highest counselor caseloads.
Two points are particularly worth noting here. First, the relationship between caseloads and students odds of speaking with counselors and their odds of articulating college expectations are both much more pronounced in the 12th grade than in the 10th grade. Although there was no relationship between caseload and expectations in the 10th grade, the counselor caseload is significantly negatively related to 12th graders odds of college plans. This finding suggests that even the most overburdened counselors attempt to discuss higher education with a broad array of students at the beginning of their educational careers. In the last year of high school, however, many students in schools with large counselor caseloads do not discuss college with a counselor. One possible explanation for this finding is that overburdened counselors may invest their counselor resources more selectively among high school seniors. Second, the predicted probabilities graphed in Figure 2 suggest that in schools with an average counselor caseload or smaller, students have nearly the same probability of speaking with counselors about college as planning to attend college. However, in schools with larger than average caseloads, these probabilities are divergent, with more students articulating plans to attend college than speaking with a counselor.
Finally, Table 4 considers the relationship between school counselor caseloads and students odds of enrolling in two-year and four-year colleges. These multinomial logistic models indicate that there is no significant association between counselor caseload and students odds of attending a two-year college. However, there is a negative association between counselor caseload and students odds of attending a baccalaureate degree-granting institution, even after controlling for student and school characteristics. This relationship is nonlinear, such that the negative association between counselor caseload and odds of attending a four-year college is less pronounced among schools with particularly high counselor caseloads. This suggests that negative effects of large counselor caseloads on four-year college attendance begin to flatten in the schools with the largest caseloads.
Figure 3 illustrates these findings by graphing students predicted probabilities of enrolling in two-year and four-year colleges, with control variables held at the sample means. This figure indicates that in schools with the largest counselor caseloads, students four-year college attendance rates are lowest and two-year college attendance rates are highest. Students in schools with the smallest caseloads are more than twice as likely (63%) to attend a four-year PEI as they are to attend two-year schools (28%), however students in schools with the largest caseloads are 3% less likely to attend 4-year PEIs (40%) than they are to attend two-year colleges (43%). The tapering of the slope for the line of students who do not attend college indicates that students at schools with the average size caseload have about an 18% probability of not attending college, compared to 10% for students in schools with the smallest caseloads.
DISCUSSION AND CONCLUSION
The purpose of this paper is to explore the relationship between the size of high school counselor caseloads and students college expectations, preparation, and enrollment patterns. A noticeable shortcoming in the current school counseling literature is the limited research on how counselor caseloads affect college access. Carrell and Carrell (2006) even state that as far as they are aware, there is no direct empirical evidence in support of lower student to counselor ratios (p. 2). Although there is the general belief that smaller counselor caseloads would affect the counseling services available (e.g. Bryan, Holcomb-McCoy, Moor-Thomas, & Day-Vines, 2009; Bryan et al., 2011; Clinedinst & Hawkins, 2011; Corwin et al., 2004) there is virtually no research to specify the ideal caseload size, which is surprising because ASCA (2012b) clearly communicates their recommended caseload of 250 students for every counselor.
The descriptive analyses illustrate that students who are already disadvantaged are concentrated in schools with larger counselor caseloads, further limiting their access to knowledgeable adults. Students whose mothers have the lowest educational attainment are most represented in schools with an average of 135 10th grade students per counselor; the findings are similar for students whose parents have the lowest incomes. Black and Latino students, who have been noted to be more likely to rely on school personnel for college information and guidance (Freeman, 1997; Perez & McDonough, 2008), are also more concentrated in the schools with the largest caseloads, whereas white students are overrepresented in schools with the smallest caseloads. Students who already face significant barriers to college (such as limited finances or few college-educated family or community members) also face limited access to their school counselors due to the fact that their counselors typically have many students to manage.
Our analyses indicate that these inequalities in access to counselors have important consequences for students as they prepare for college throughout high school and transition from high school to college. Even after controlling for student and school background characteristics, we find that students in schools with high counselor caseloads are less likely to speak to their counselors, less likely to formulate and act on college plans, and less likely to attend four-year colleges. Notably, our findings suggest that the association between counselor caseload and student behavior are more pronounced in the later stages of the high school-to-college transition than the early stages, net of the same control variables. We find that nearly all 10th graders, regardless of school-level counselor caseloads, articulate college expectations. However, as students progress through high school, large caseloads negatively impact students college preparation and attendance behaviors. Previous research indicates that early and consistent expectations for college are an important step in progressing through the preparation process and attending four-year institutions (Klasik, 2012). However, the current study indicates that students are formulating these expectations independently of their access to school counselors. Based on these findings and the extant literature, we advise counselors to guide students through the necessary steps of preparing for and enrolling in college: taking requisite coursework and admissions examinations, applying for financial aid, applying to multiple institutions, and finally enrolling.
The counselor caseload was also significant in predicting students likelihood of taking the SAT/ACT. This is an important part of college preparation because many colleges and universities, in particular closed-access institutions, use standardized tests to evaluate students likelihood for success at their school and may use scores as a placement tool (e.g., indicating whether a student needs to enroll in remedial courses). Taking these exams signals to both teachers and counselors a students intent to apply for and enroll in college (Tuitt et al., 2011). The results demonstrate that counselors guidance impacts student outcomes when they are related to discrete college preparation and attendance milestones such as taking college admissions tests and actually attending college. These results align with the findings that college coaches can increase students enrollment in college by focusing efforts on distinct tasks, such as completing aid applications and applications to three or more institutions (Stephan & Rosenbaum, 2013).
While Rosenbaum et al. (1996) worry that the college for all message conveyed by U.S. high school counselors leads large numbers of unprepared students to enroll in community college, this finding indicates that counseling caseloads are unrelated to two-year college enrollment. However, students enrolled in schools with the lowest counselor caseloads are more than 50% more likely to enroll in four-year colleges than their peers in schools with the highest counselor caseloads, net of controls.
Throughout our analyses, we find evidence to suggest that the relationship between counselor caseloads and student outcomes are nonlinear, such that students in the schools with the highest caseloads are more likely to articulate college plans, take the SAT/ACT, and enroll in four-year colleges than one might expect given the overall relationship between caseload and student college transitions. This finding may indicate that schools with the largest counselor caseloads have developed alternative mechanisms to provide students with college encouragement and information. One possibility is that teachers in these schools fill in for overburdened counselors and help students plan for college. Another possibility is that these schools rely heavily on extracurricular college outreach programs.
IMPLICATIONS, LIMITATIONS, AND FUTURE RESEARCH
Our findings suggest that efforts to reduce counselor caseloads could lead to improved student outcomes and more equitable access to higher education. Smaller caseloads allow for a greater opportunity to develop social capital, providing students with sources of college knowledge and with each counselor potentially having increased availability for precollege counseling. Clinedinst and Hawkins (2011) argue that one way to increase college completion is to implement state policies that will improve middle and high school counseling, which includes maintaining reasonable school counselor and college counselor caseloads and managing counselors time on task. With this study and additional research, school counseling positions may become better defined, more valued and understood, and may enjoy greater support at district, state, and federal levels. By communicating the importance of small counselor caseloads, these findings may lead schools and districts to implement policies that protect counselors time, such as redistributing noncounseling duties to other employees or considering part-time or contracted employees (e.g., test administrators) during high-need times.
Counselors can help create and maintain a pervasive college-going culture, which may increase students college knowledge and improve college attendance rates of traditionally underrepresented students, particularly in schools with new or low college-going traditions. Part of developing a college-going culture may include educating teachers, parents, and other community partners about the importance of preparing early for college and providing tools for how to talk with students about college. Districts, particularly those with a strong college-going culture, may provide additional structural support for precollege counseling and prioritize counselors efforts to maintain or increase access to college and college preparation resources to all students.
However, these analyses are limited in several important ways. First, these correlational analyses cannot conclusively isolate the causal relationship between counselor caseloads and student outcomes. While our analyses control for a limited set of student demographic family background characteristics, as well as students prior achievement, other factors such as wealth, neighborhood quality, and commitment to college may confound observed associations between counselor caseloads and student outcomes. Second, as mentioned, this study takes into account the effects of several demographic differences by controlling for these variables, yet we do so in reference to white and female students and include a continuous variable for income. We call for future research to examine how the counselor caseload distinctly impacts outcomes for students in various racial/ethnic, income, immigrant, and primary language subgroups and school settings (e.g., charter schools and school urbanicity) by analyzing subgroups separately. Third, although this study provides a view of how the size of the counselor caseload affects students postsecondary choice process, this dataset does not have qualitative or frequency measures of students interactions with counselors. Because of this we cannot understand how these studentcounselor interactions occurred, who initiated them, the duration or frequency of the meetings, or the type of guidance provided. It is plausible that several one-on-one meetings with a school counselor could have a very different effect than one large-group assembly. Additionally, students interactions with college coaches or other noncredentialed personnel would not be captured in the current study but could lead to sizable effects in student outcomes. Chicago Public Schools employed college coaches who assisted students in completing key actions in the college preparation process. In these schools more students completed the Free Application for Federal Student Aid (FAFSA) and three or more college applications, as compared to schools without college coaches (Stephan & Rosenbaum, 2013). Employing college coaches may improve college preparation and attendance rates without increasing the number of school counselors.
Using another dataset, such as the High School Longitudinal Study of 2009, which has a counselor survey in addition to the administrator instrument, could aid our understanding of school counselors use of time, counselors caseloads, and schools college preparation resources. Follow-up studies using future waves of the ELS: 2002 data could also lead to a better understanding of long-term effects of the size of the counselor caseload regarding degree completion and earning potential.
Further investigation using a variety of analytical approaches and settings of how counselor caseloads affect student outcomes is warranted to better illustrate the effectiveness of the school counselor and the impacts of the size of the school counselor caseload. In addition to high schools, middle school counseling programs should also be examined; monitoring counselor caseloads in middle school may increase students access to academic pathways that lead to college eligibility and readiness (Lee et al., 2011). Furthermore, precollege counseling may occur in less formal spaces and from multiple sources. Exploring how counselors and schools partner with community resources such as afterschool programs or local community and industry leaders and how students use in-school and out-of-school resources can result in a better allocation of counselors time and other resources.
Mixed-methods research in this area could provide useful information on how counselors and counseling resources are allocated and utilized and may add to the generalizability of the current findings (Carter & Hurtado, 2007). Comparative analyses between schools with large and small caseloads, and mixed-methods case studies in particular, may provide insight on how counselors employ college guidance techniques, to which students, and the extent to which this guidance translates into observable actions (i.e., ordering and submitting high school transcripts for college). Additionally, accounts from multiple school personnel, (e.g., administrators, counselors, teachers, and paraprofessionals) students, and parents could provide for a well-rounded understanding of how school counselors affect student outcomes. Student and counselor narratives may illuminate the college choice process even further, allowing for a rich understanding of precollege counseling interactions and influences. These suggestions are not exhaustive but can address some limitations of the current dataset and the limited research in the field. This paper attempts to fill the current gap in research about the relationships between the school counselor caseload and student outcomes and may open the door to additional empirical work exploring the profession of high school counseling in regards to college access and attendance, as well as other indicators of student success.
American School Counselor Association. (2012a). State school counseling mandates and legislation. Retrieved from http://www.schoolcounselor.org/school-counselors-members/careers-roles/state-school-counseling-mandates-and-legislation
American School Counselor Association. (2012b). Student-to-school counselor ratio 20102011. Retrieved from http://www.schoolcounselor.org/asca/media/asca/home/ratios10-11.pdf
American School Counselor Association. (2012c). The role of the professional school counselor. Retrieved from http://www.schoolcounselor.org/asca/media/asca/home/RoleStatement.pdf
American School Counselor Association. (2012d). Why secondary school counselors? Retrieved from http://www.schoolcounselor.org/school-counselors-members/careers-roles/why-secondary-school-counselors
Bryan, J., Holcomb-McCoy, C., Moor-Thomas, C., & Day-Vines, N. L. (2009). Who sees the school counselors for college information? A national study. Professional School Counseling, 12(4), 280291.
Bryan, J., Moore-Thomas, C., Day-Vines, N. L., & Holcomb-McCoy, C. (2011). School counselors as social capital: The effects of high school college counseling on college application rates. Journal of Counseling & Development, 89, 190199.
Burnham, J. J., & Jackson, C. M. (2000). School counselor roles: Discrepancies between actual practice and existing models. Professional School Counseling, 4(1), 4148.
Cabrera, A. F. (1994). Logistic regression analysis in higher education: An applied perspective. In John C. Smart (Ed.), Higher education handbook of theory and research (Vol. 10, pp. 225256). New York: Agathon Press.
Cabrera, A. F., & La Nasa, S. M. (2000). Understanding the college-choice process. New Directions for Institutional Research, 107, 522.
Carrell, S. E., & Carrell, S. A. (2006). Do lower student to counselor ratios reduce school disciplinary problems? Contributions to Economic Analysis & Policy, 5(1), 1.
Carter, D. F., & Hurtado, S. (2007). Bridging key research dilemmas: Quantitative research using a critical eye. New Directions for Institutional Research, 133, 2535.
Coleman, J. S. (1988). Social capital in the creation of human capital. The American Journal of Sociology, 94(Supplement: Organizations and Institutions: Sociological and Economic Approaches to the Analysis of Social Structure), S95S120.
College Board. (2011). The completion agenda 2011 progress report. New York: Author.
Corwin, Z. B., Venegas, K. M., Oliverez, P. M., & Colyar, J. E. (2004). School counsel: How appropriate guidance affects educational equity. Urban Education, 39(4), 442457. doi:10.1177/0042085904265107
Eccles, J. S., Vida, M. N., & Barber, B. (2004). The relation of early adolescents college plans and both academic ability and task-value beliefs to subsequent college enrollment. The Journal of Early Adolescence, 24(1), 6377. doi:10.1177/0272431603260919
Engberg, M. E., & Wolniak, G. C. (2010). Examining the effects of high school contexts on postsecondary enrollment. Research in Higher Education, 51, 132153. doi:10.1007/s11162-009-9150-y
Farmer-Hinton, R. L. (2008). Social capital and college planning: Students of color using school networks for support and guidance. Education and Urban Society, 41(1), 127157. doi:10.1177/0013124508321373
Farmer-Hinton, R. L., & McCullough, R. G. (2008). College counseling in charter schools: Examining the opportunities and challenges. The High School Journal, 91(4), 7790.
Freeman, K. (1997). Increasing African Americans' participation in higher education: African American high-school students' perspectives. The Journal of Higher Education, 68(5), 523550.
González, K. P., Stoner, C., & Jovel, J. E. (2003). Examining the role of social capital in access to college for Latinas: Toward a college opportunity framework. Journal of Hispanic Higher Education, 2(2), 146170. doi:10.1177/1538192702250620
Hamrick, F. A., & Stage, F., K. (2004). College predisposition at high-minority enrollment low-income schools. The Review of Higher Education, 27(2), 151168.
Hardy, M. A. (1993). Regression with dummy variables. Newbury Park, CA: Sage Publications, Inc.
Ingels, S. J., Pratt, D. J., Wilson, D., Burns, L. J., Currivan, D., Rogers, J. E., & Hubbard-Bednasz, S. (2007). Educational longitudinal study of 2002 (ELS:2002) Base-year to second follow-up data file documentation. Washington, DC: U.S. Department of Education, National Center for Education Statistics.
Jarsky, M. K., McDonough, P. M., & Nunez, A. M. (2009). Establishing a college culture in secondary schools through P-20 collaboration: A case study. Journal of Hispanic Higher Education, 8(4), 357373.
Kimura-Walsh, E., Yamamura, E. K., Griffin, K. A., & Allen, W. R. (2009). Achieving the college dream? Examining disparities in access to college information among high achieving and non-high achieving Latina Students. Journal of Hispanic Higher Education, 8, 298315.
Kirst, M. W., & Venezia, A. (Eds.). (2004). From high school to college: Improving opportunities for success in postsecondary education. San Francisco, CA: Jossey-Bass Publishers.
Klasik, D. (2012). The college application gauntlet: A systematic analysis of the steps to four-year college enrollment. Research in Higher Education, 53(5), 506549.
Krei, M. S., & Rosenbaum, J. E. (2001). Career and college advice to the forgotten half: What do counselors and vocational teachers advise? Teachers College Record, 103(5), 823842.
Lee, Jr., J. M., Edwards, K., Menson, R., & Rawls, A. (2011). The college completion agenda: 2011 progress report. College Board Advocacy & Policy Center.
Lee, V. E., & Ekstrom, R. B. (1987). Student access to guidance counseling in high school. American Educational Research Journal, 24(2), 287310.
Massé, J. C., Perez, R. J., & Posselt, J. R. (2010). Revisiting college predisposition: Integrating sociological and psychological perspectives on inequality. Equity & Excellence in Education, 43(3), 279293. doi:10.1080/10665684.2010.492271
McDonough, P. M. (1997). Choosing colleges: How social class and schools structure opportunity. New York: State University of New York Press.
McDonough, P. M. (2005). Counseling and college counseling in Americas high schools. Alexandria, VA: National Association for College Admission Counseling.
Muhammad, C. G. (2008). African American students and college choice: A consideration of the role of school counselors. NASSP Bulletin, 92(2), 8194. doi:10.1177/0192636508320989
National Association for College Admissions Counselors (NACAC). (2011). State of college admission, 2011. Washington, DC: National Association for College Admission Counseling.
Perez, P. A., & McDonough, P. M. (2008). Understanding Latina and Latino college choice: A social capital and chain migration analysis. Journal of Hispanic Higher Education, 7(3), 249265. doi:10.1177/1538192708317620
Perna, L. W. (2000). Differences in the decision to attend college among African Americans, Hispanics, and whites. The Journal of Higher Education, 71(2), 117141.
Perna, L. W. (2006). Studying college access and choice: A proposed conceptual model. In J. C. Smart (Ed.), Higher education: Handbook of theory and research (Vol. XXI, pp. 99157). New York: Springer Press.
Perna, L. W., Rowan-Kenyon, H. T., Thomas, S. L., Bell, A., Anderson, R., & Li, C. (2008). The role of college counseling in shaping college opportunity: Variations across high schools. The Review of Higher Education, 31(2), 131159.
Plank, S. B., & Jordan, W. J. (2001). Effects of information, guidance, and actions on postsecondary destinations: A study of talent loss. American Educational Research Journal, 38(4), 947979. doi:10.3102/00028312038004947
Rosenbaum, J. E., Stephan, J. L., & Rosenbaum, J. E. (2010). Beyond one-size fits-all college dreams. American Educator, 34(3), 213.
Rosenbaum, J. E., Miller, S. R., & Krei, M. S. (1996). Gatekeeping in an era of more open gates: High school counselors views of their influence on students college plans. American Journal of Education, 104(4), 257279.
Sears, S. J. (1993). The changing scope of practice of the secondary school counselor. School Counselor, 40(5), 384388.
Solorzano, D. G., Villalpando, O., & Oseguera, L. (2005). Educational inequities and Latina/o undergraduate students in the United States: A critical race analysis of their educational progress. Journal of Hispanic Higher Education, 4(3), 272-294. doi:10.1177/1538192705276550
Stanton-Salazar, R. D., & Dornbusch, S. M. (1995). Social capital and the reproduction of inequality: Information networks among Mexican-origin high school students. Sociology of Education, 68(2), 116135.
Stephan, J. L., & Rosenbaum, J. E. (2013). Can high schools reduce college enrolment gaps with a new counseling model? Educational Evaluation and Policy Analysis, 35(2), 200219.
Tuitt, F. A., Van Horn, B. N., & Sulick, D. F. (2011). Transmitting the right signals: The continued significance of promoting college entry to students of color. Enrollment Management Journal, 5(1), 1031.
Figure 1. Predicted probabilities of the effect of the counselor caseload on students 10th grade outcomes: whether students expect to attend college and whether a student spoke with a school counselor about college. Predicted probabilities from the full models in Table 2, with control variables held at the mean.
Figure 2. Predicted probabilities of the effect of the counselor caseload on students 12th grade outcomes: whether students plan to attend college and whether a student spoke with a school counselor about college. Predicted probabilities from the full models in Table 3, with control variables held at the mean.
Figure 3. Predicted probabilities of the effect of the counselor caseload on students likelihood of attending college. Predicted probabilities from the full models in Table 4, with control variables held at the mean.