Current Classroom Placement Practices Discriminate Instruction and Achievement; Assessing a New Instrument and Process with PST

Focused Area

Improving School Climate for Youth-At-Risk

Relevance to Focused Area

When Sanders and Horn first published value added models in 1998, they had developed a very intricate statistical tool to measure educational values over time. Since then numerous other researchers have analyzed the models, developed new models, and compared results to find that measuring education is a truly fuzzy science. In doing so a light has been shown on the assumption of random class rosters; it was Sander’s hypothesis that classes would be randomly created. However based on the research reviewed, classrooms are anything but random, especially at the elementary level. This finding may prove that specific students are receiving specific teachers based on their performance or level of advantage over other students, and not the 12 to 16 other academic and non-academic attributes frequently used on pink and blue cards. It may be that while researchers were looking for a better model to assess staff, they found a better way to identify student segregation. Kalogrides and Loeb have recently written on this topic. Their 2013 article stated that, "Sorting students by achievement level exposes minority and poor students to lower quality teachers and less resourced classmates” (p. 304). It is this researcher’s view that well documented tracking in high school begins with inadvertent ability grouping at the elementary level due to a classroom placement process based on a thoughtful but flawed method. A revised system for developing classroom rosters is needed. This presentation compares the author’s placement procedure with the conventional classroom placement process widely used in most K-8 multiple subject schools today using propensity score analysis.

Primary Strand

Academic Achievement & School Leadership

Relevance to Primary Strand

This presentation will present that our common procedure for designing classroom rosters; the method that sorts students by achievement levels and input from well-meaning stakeholders. In doing so, this presentation will prove that students with specific scores and unique characteristics often reside with specific teachers. Throughout current research, there is evidence that the current practice often segregates students to specific classroom, thereby creating a very different education within the same school, and grade level. It is evident that K – 5 education has perpetuated a classroom placement process that has very likely reduced student achievement for specific learners while creating class rosters that reflect an unfair distribution of at-risk students for specific teachers.

Brief Program Description

Based on recent studies, classrooms are anything but random, especially at the elementary level. In fact, Rothstein’s 2009 research found that the assignment of future teachers predicts the student’s past gains or achievement. This finding may prove that specific students are receiving specific teachers based on their performance or level of advantage over other students and not just the 12 to 16 other academic and non-academic attributes frequently used on pink and blue cards. It may be that while researchers were looking for a better model to assess staff (Value-added Model), they found a better way to identify student segregation. Kalogrides and Loeb have recently written on this topic. Their 2013 article states that, "Sorting students by achievement level exposes minority and poor students to lower quality teachers and less resourced classmates” (p. 304).

Summary

It is evident that elementary and middle schools alike need a classroom placement process that more equitably places students. Student achievement and parent or staff perception cannot continue to be the main elements for building classroom rosters. All stakeholders do not have the same level of input in this practice, so unethical compromises are made by staff and administration.

If classrooms could be created in a different fashion so teachers have similar students in all classrooms of the same grade level, value-added models and other metrics for evaluating staff and students may be more effective. Creating class rosters using an alternative process so students are evenly placed and sorted by staff, allows for less inadvertent tracking and a fair chance at equitable instruction for all learners.

This presentation will allow its audience to see classroom placement from a new perspective and therefore more data to make more informed decisions.

Evidence

Many states and even school districts are moving towards the evaluation of teachers based on their students’ standardized test scores. The federal program for school improvement, Race to the Top and other unique programs from various states utilize various versions of a value-added model to compare individual student achievement each year against a much larger population looking for growth. As shown above, classroom compositions are anything but random, or even balanced. Consequently teachers within the same school, and even in the same grade level will most likely have very different classrooms in respect to student abilities, student outcomes, student self-perceptions, and as a result student potential for growth.

Many researchers believe standardized test scores are the best measure of success and failure in education (Hanushek & Rivkin, 2010). Whether the standardized assessment is the NAEP, TIMMS, SAT, or CAHSEE, there are plenty of assessment tools to measure the success or failure of education. It is evident that past standardized test scores predict future scores too. McGlinchey and Hixson (2004) extended the Stage and Jacobson research in 2001 to state that curriculum-based measurement (CBM), or school based assessments showed a strong correlation to state assessments. Two different studies have found that principal’s assessment of their teachers show a positive correlation to VA models, proving that subjective performance measures also have a place in identifying successful instruction (Harris & Sass, 2007; Jacob & Lefgren, 2005). VA models use various assessment types to predict how each student should score year to year on standardized assessments. Harris (2009) is cautious but believes this is valuable information for school districts, sites, and administrators. The data may be better used to form custom intervention groups or advanced programs depending on each student cohort. In the end the problem is not finding data, it is finding properly sorted classrooms so the data is unaffected by negative impacts from segregation.

Research Evidence of Segregation and its Effect on Class Lists

Through the years, there has been a lot of research and practical usage of tracking or ability grouping with respect to education. Since this literature review has found classroom compositions to be less than balanced or equal, there must be some level of tracking or segregation. Conger noted in his study on segregation and its relationship to randomized class lists, that segregation levels in normal class lists were 2-5 times higher than levels achieved through random practices. He went on to state; “segregation is partially driven by non-random processes, such as systematic segregation or other sorting practices” (p. 233, 2005). This literature review seeks to understand if it is the current placement sorting process that is responsible for increased levels of segregation.

Tracking has gained favor and lost its popularity throughout the ebb and flow of educational history. Often the research is in favor of ability grouping when one reads about high ability students, gate programs, and supporting families with substantial economic income (Chiu et al., 2008; Fielder, Lange, & Winebrenner, 2002; Scott, 2001; Tieso, 2003). However if one reviews research completed in low socioeconomic areas, about diverse ethnicities, or around urban centered schools then ability grouping or tracking diminishes in popularity (Biafora & Ansalone, 2008; Burris & Garrity, 2008; Burris & Welner, 2005; Kalogrides & Loeb, 2013). Gutierrez and Kulik had separate but similar research findings stating that ability grouping works for all groups. Both researchers found that to be effective it was imperative that the school district or school site focused their intentions on specific groups; doing so requires time and effort spent on professional development, resources, and specific curriculum for specific groups of students (Gutierrez, 1995; Kulik, 1992). In such cases research showed that tracking by ability was effective. The problem with tracking and ability grouping comes when it is a knee-jerk response, or in response to diminishing resources, or as this paper asserts, happens by accidental or contingency processes.

Tracking becomes detrimental when it happens as a byproduct of high-stakes testing, reduced resources, or as stated above, as an unexpected outcome from pressures exerted on a school or district.. However, grouping students by ability has been found to work best when instructional leaders and curricula are focused on key factors that promote student achievement (Gutierrez, 1995). Biafora & Ansalone utilized a qualitative method to survey 272 principals on the topic of tracking (2008). The authors found that due to decreased resources teachers, schools, and principals become reluctant -but in the end except tracking as an effective strategy for solving instructional dilemmas. In addition they found tracking to be more pervasive in lower SES districts. In fact parents of low SES students show little concern on the issue of their student being grouped with other students of low ability (or high ability); much less than parents of high SES students. Burris & Garrity noted that unless there is an extreme focus on curriculum design, classroom assessment, and professional development homogeneous classrooms and class lists cannot be effective (2008).

Kraemer, Worth & Meyer initiated a qualitative exploratory study to determine how students were assigned to teachers in kindergarten through eighth grade schools across three large urban school districts. Their information was collected from surveys from the principals. All of the principals engaged in yearly practices of “balancing” classes. However results from the study suggested more work needed to be done to effectively create true heterogeneous classrooms (2011). In 1987 another researcher identified methods used to assign students not only to the classrooms but also to their teachers. The researcher used small samples of elementary schools; however each of the schools were ethnically diverse. There was a strong correlation of parental influence on student classroom assignments and how that may or may not relate to specific teachers (Monk, 1987). These pieces of research also denote a level of segregation in a long standing process for creating class rosters.

During more recent times, economic policy has had a large influence on education. In response obtaining high stakes testing data was seen as the best way to assess student achievement (Hanushek & Rivkin, 2010). For the same reason it is often used to divide a cohort into classes; but little has changed. The school staff’s assumption that they can identify the best environment for each student is an even more powerful element in the placement process. Often, teachers that have a special education background will tend to collect students with special needs. Male teachers will attract students that need a strong father figure or structure. And often, there is a teacher who collects many of the more compliant students. Harris found that school sites benignly distribute students this way hoping it is the best environment for student success (2009). This literature review has shown evidence that such distribution may diminish each class’ ability to show the same growth, post the same scores, and therefore affect any given teacher’s evaluation when using value-added metrics.

In the current practice for building classroom lists, it has been shown that specific teachers receive specific students. A level of segregation is apparent due to student achievement scores, advantaged parent input, and the persuasive efforts of various staff members. These pressures impact the heterogeneous nature of the classroom placement process yielding more segregated or unique classroom compositions. As shown above, a contingency placement practice such as this will not likely receive the support it needs to improve student achievement at the same rate as other classrooms. In other words, unique classrooms comprised of inordinate numbers of ELD, low socioeconomic, or difficult students is not likely to receive differentiated support from other classrooms in the same grade, at the same school. In fact, rather than receiving the more mature or highly qualified teacher, this class is more likely to have the novice teacher thereby creating a much different educational experience from similar classrooms (Kalogrides & Loeb, 2013). These unethical classroom compositions could lead to rash conclusions about the instructional effectiveness of specific teachers.

Format

Poster Presentation

Biographical Sketch

Mr. Thomas S. Bruce has been an educator and administrator in Michigan and in California over the last 22 years. Recently he has completed his Ed. D. at Asusa Pacific University in Educational Leadership. Currently he is an administrator for Arcadia Unified School District; where he has been for the last 9 years.

Start Date

11-5-2015 5:45 PM

End Date

11-5-2015 6:45 PM

This document is currently not available here.

Share

COinS
 
Nov 5th, 5:45 PM Nov 5th, 6:45 PM

Current Classroom Placement Practices Discriminate Instruction and Achievement; Assessing a New Instrument and Process with PST

Based on recent studies, classrooms are anything but random, especially at the elementary level. In fact, Rothstein’s 2009 research found that the assignment of future teachers predicts the student’s past gains or achievement. This finding may prove that specific students are receiving specific teachers based on their performance or level of advantage over other students and not just the 12 to 16 other academic and non-academic attributes frequently used on pink and blue cards. It may be that while researchers were looking for a better model to assess staff (Value-added Model), they found a better way to identify student segregation. Kalogrides and Loeb have recently written on this topic. Their 2013 article states that, "Sorting students by achievement level exposes minority and poor students to lower quality teachers and less resourced classmates” (p. 304).