Impact of Class Size on Students' Achievement - Essay Example

? Impact Size on Achievement Alison Flynn Long Island Acknowledgements Table of Contents Introduction ……………………………………………………………………….. 4 Literature Review …………………………………………………………………. 6 Implications ………………………………………………………………………. 15 Applications ………………………………………………………………………. 17 Conclusion ………………………………………………………………………… 18 References …………………………………………………………………………..19 Introduction The state of the education system is a matter of huge concern in any country, with many observers regarding student achievement as the gauge of its effectiveness. Hence, there are a great deal of investigations into all the possible factors that have been identified as affecting students’ achievement, consequently leading to many debates among the stakeholders, adoption and implementation of study recommendations and post implementation analyses. One of the main areas of concern to the stakeholders involved, i.e., policy makers, parents, teachers and researchers, is the question of whether or not class size has an impact on students’ achievement. A great number of resources in terms of financial and human capital have been committed to investigating the impact of class size on the effectiveness of the learning process, thus indicating that this is an area of major importance in the education sector. This is understandable considering that education consumes huge portions of the budgets of parents and the government, both of whom who wish to be assured of quality. Class size is an issue that is largely at the center of the school system, especially since it influences not only the size of classrooms but also the number of teachers in a school (and by extension in the country), hence, it is naturally a concern to parents, teachers, school administrators, the government and all other stakeholders in the education sector (Mosteller, 1995). The research that has been conducted concerning the impact of class size on students’ achievement is particularly wide- and far-reaching, but has on several occasions yielded different and sometimes even conflicting results, thus raising the question, what in fact is the impact of class size on students’ achievement? Another question that arises is whether class size has similar impacts across the grades in elementary education (if impacts are found to be present), and finally whether any present impacts of class size on students’ achievement are residual after the students are integrated back into regular-size classes. Some studies have concluded that class size does have an impact on learning while others dispute this, and this fact alongside the research questions make this an interesting and important area of inquiry. A review of relevant literature based on the premise that class size has an effect on students’ achievement is the backbone of this inquiry, alongside the implications of this literature review and their potential applications in an actual school setting. Literature Review The literature on the effects of class size on students' learning achievement was collected through a search in education journals, databases and relevant sites. This review consists of primary sources and secondary ones that were obtained after refining the search to focus on the impact of class size on elementary school education. A great deal of work has been done in this area over the last three decades, with conflicting results; most studies have suggested that reduction in class size leads to considerable improvements in student learning, while others conclude that class-size reductions may not bring with them any appreciable or significant effects (Pritchard, 1999). Besides this, the other matter that arises is the model of investigations into class size. CEC (2010) believes that it is important to distinguish between class size and student-teacher ratio. He defines class size as the number of students that a teacher handles per class at any one time, while student-teacher ratio is the division of a school’s total enrollment of students by its total number of teachers. It is, however, important to note that several studies have used student-teacher ratios to draw conclusions about the impact of class size. Another issue with regard to study models, and one that would be particularly important in arriving at a conclusion on whether class size has an impact on achievement, is the different approaches to research. Krueger, Hanushek and Rice (2002) state that non-experimental studies rely on many assumptions and are thus subject to numerous misconceptions, while randomized field trials are valid since they involve samples large enough to equally distribute relevant characteristics. Cho et al. (2010) state that for a study to give reliable data on the actual impact of class size on students’ achievement it has to consider both observed and unobserved differences between small- and large-size classes. The observed differences are much easier to account for, while the unobserved ones are altogether a different challenge; for example, overestimation and underestimation of the impact of class size can occur as a result of extra support from parents and learning difficulties, respectively. They suggest that random assignment of the students into treatment and control groups (random assignment of selected children into small and large classes followed by a later comparison of achievement) is the way to overcome this bias problem. Mosteller (1995) appreciates the difficult nature of this subject in terms of designing and executing the investigation in order to come up with valid and reliable results. To begin with, parents may be opposed to variation in treatment of their children. Besides this, it is difficult to ensure that the different groups in the investigation were at the same education level in the beginning hence a huge assumption is made which may potentially interfere with the results. These investigations also require a careful description and execution of the treatment alongside use of suitable and consistent measures of performance that describe students' achievement. The atmosphere must also be healthy, since this is a variety of study that can be easily sabotaged. The Student-Teacher Achievement Ratio (STAR) project is perhaps the best designed study available and at the same time the biggest landmark of class size reduction initiatives (Krueger, 1999). Its findings have had far-reaching effects on education policy since it was not only well designed but was also cross-cutting, applying to poor and rich, rural and urban, and majority and minority children (Mostella, 1995). Chetty et al. (2011) performed a wide-ranging study analyzing the validity of STAR’s experimental design and did not find it wanting. STAR was a four-year longitudinal study covering 79 schools in 42 different systems. Achievement and development of students were analyzed in three types of classes: a small class containing 13 to 17 students handled by one teacher; a regular class containing 22 to 25 students handled by one teacher; and finally, a regular class identical to the preceding one, but now with an accompanying full-time teacher aide. The tests for achievement were Stanford Achievement Test for 3rd grade, STAR’s Basic Skills Criterion Test for the 1st and 2nd grades and the Tennessee Basic Skills Criterion Test for 3rd grade students. The development of the students was measured by the Self-Concept and Motivation Inventory (SCAMIN) test (Word, 1990). The results of the STAR study indicate that small class size does have a significant impact on student performance, and, of major interest, it was found that minority students benefitted the most from class-size reduction (Mostella, 1995). Specifically, STAR results indicated that students in smaller classes outperformed those in bigger classes on standardized and curriculum-based tests--Standardized Achievement Tests and Basic Skills First tests, respectively. This improvement in achievement was initially twice as great among minority students in comparison to majority students before it leveled off later, and these results were deemed reliable because of the large sample size, use of random assignment and an in-school design (Pritchard, 1999). Finn et al. (1990, cited in CEC, 2010) submitted their findings that the students who had taken part in STAR were 6 to 13 months ahead of their counterparts who had not attended small size classes in terms of learning achievement. The achievement in this case covered prowess in mathematics, reading and science studies in 4th grade, 6th grade and 8th grade. Their analysis indicated that it was necessary for students to spend a minimum of three years in a small-size class if all the gains over the regular-size class students in terms of achievement were to be sustained over the next grades. On an interesting note, they found out that the gains of attending a small-size class continued to increase from grade to grade. A follow-up on the STAR project for the students who had participated and were now in regular classes was performed and dubbed the Lasting Benefits Study (LBS). The objective of this longitudinal study was to determine whether the students who had attended a small-size class from 1st to 3rd grade still had an advantage going into 4th grade in a regular-size class. The test used in LBS was the Tennessee Comprehensive Achievement Program (TCAP), which is a consistent test administered by all schools in Tennessee. The results indicated that the students who had been in small-size classes in the first three grades of elementary education maintained a significant advantage in terms of achievement in the 4th grade across all schools. It was the same case for 5th, 6th and 7th grades as the students who had attended small size classes outperformed their fellow students who had not gone through STAR (Nye et al., 1994; cited in Mosteller, 1995). Another analysis by Finn et al. (1990, cited in CEC, 2010) compellingly concluded that LBS indicates that class size does have an impact on students’ achievement since its results provide a largely unambiguous case for the significance of class-size effect. According to Molner et al. (1998, cited in CEC, 2010), who reviewed the Student Achievement Guarantee in Education (SAGE) Program, smaller class sizes have a positive effect on students’ achievement. SAGE was a class-size reduction program performed in Wisconsin with the objective of reducing the student-teacher ratio to 15:1 from kindergarten to 3rd grade. The impact on students’ achievement was to be determined through a comparison of students from these schools with students from selected background schools that had not implemented SAGE. The comparison indicated that first grade students in SAGE schools performed significantly better than their counterparts in comparison schools. The difference in achievement between minority and majority students in SAGE schools was observed to decrease while it grew bigger in the comparison schools in 1st grade. It was also observed that the 2nd grade students in SAGE schools maintained their advantage in terms of achievement over 2nd grade students in the comparison schools. In later work, Molner et al. (1999, cited in CEC, 2010) indicated that SAGE significantly improved minority students’ achievement. Egelson et al. (1996) step in with their analysis of the pilot study and phase-in of class size reduction programs that were introduced in Burke County. A total of 2,318 students from 1st and 2nd grades were involved in the program. The performance of students drawn from schools with small size classes in reading and mathematics were then compared to a comparison group made up of students from schools that had not adopted small size classes. It was shown that the students from small-size classes performed better in the two areas than did their counterparts in large-size classes across the 1st, 2nd and 3rd grades. Another observation made from the study indicated that students from small-size classes dedicated more time to class instruction (a rise from 80% to 86%) than their counterparts in schools that were not part of the class size reduction program. Small-class students were also found to be more disciplined, as indicated by the reduction in time spent for disciplinary cases from 20% to 14%. Reduction of class size brings about improved reading achievement. Cohen et al. (2000) reckon that the implementation of programs aimed at reduction in class sizes across the United States has led to an improvement in terms of individualization of instruction and management of classes. In addition, there have been observed improvements of teacher morale, reductions in the number of disciplinary cases in schools, changes in focus from remediation to prevention and, finally, increased participation of students in classroom activities. Small-size classes have also brought along improved classroom efficiency, positivity in terms of classroom atmosphere and an expansion of learning opportunities, which have all resulted in raised enthusiasm and increased achievement not only among students but also in teachers as they are able to meet their goals. A review of reduction of class sizes in elementary schools by Cohen et al. (2000) indicated that 3rd grade students in small-size classes outperformed their counterparts in large-size classes for two years running; the effect was sustained even after these students rejoined regular-size classes. Rivkin et al. (2005) also support the view that class size does have a significant effect on students’ achievements. They found out that class size has statistically significant effects on students’ performance in two areas, mathematics and reading skills, although the gains made in these two markedly declined as the students progressed from lower grades to higher ones. Their results indicated that significant improvements occurred among students in 4th and 5th grades while the effects on 6th grade students were significantly smaller. However, there was lack of consistency with minority students; having improved results in 4th grade and actually smaller ones in 5th grade. Hence, this particular study came to the conclusion that class-size effects were not significantly larger for minority students. Ferguson’s (1991) work also indicates that there is significant improvement in students’ achievement through decreasing the student-teacher ratio. In a Texas study covering about 800 school districts containing over 2.4 million students in 1063 schools, and using the data on the TEAMS and TECAT students and teachers test results, Ferguson found out that student achievement declined with increasing the student-teacher ratio above 18:1. Test score gains are significant with reduction up to 18 students per class, but there is no significant improvement when the student number per class is below 18. There is also a lack of significant gains from reduction in class sizes after elementary school. Wenglinsky (1997) also weighed in on the impact of class sizes on achievement through a study of three national databases. He analyzed data on students in 4th grade and 8th grade and their performance in mathematics using class size as a means of gauging spending in the education sector. The 4th grade students were from schools in 203 districts, while the 8th grade students were from schools in 182 districts in the United States. The results indicated that at 4th grade, a low student-teacher ratio is associated with higher achievements in mathematics, and similarly, at 8th grade, a low student-teacher ratio has the effect of improving a school's social environment and, as a result, improving students’ achievement. From about 80 empirical studies investigating the effects of class size on achievement by students, Smith and Glass (1978) performed two meta-analyses; first for the impact of class size on students’ learning and secondly for any other outcomes of class size on students. The study involved approximately 700 comparisons of students’ achievements in small classes and in large classes. They found out that for students who had spent more than 100 hours in small classes and undertaken carefully controlled assignments, a small class size was significantly associated with better performance at all grade levels. This was especially so for classes with fewer than 20 students, where there was an advantage of a 6-percentile rank as compared to a doubled class size of 40 students. The other outcomes that were the basis of the second meta-analysis were found out to be better learning environments, improved morale levels among teachers and better reactions from students, all resulting from smaller class sizes. A related cluster analysis was used to review over 100 research studies that had been conducted to investigate the relationship between class size and students’ achievement. This involved grouping similar studies together, for example in terms of grades and subjects. It was found out that the most significant improvement of students’ achievement was among students from elementary school, especially among those from kindergarten up to 3rd grade. The same study also found out that small class sizes were especially advantageous to minority students in terms of improving performance (Robinson & Wittebols, 1986). On the other side of the divide are studies indicating that no significant gains in student achievement are made from having students study in small classes. Hoxby (2000) argues that class-size reduction is a program pursued by several stakeholders in the education sector out of self-interest and not for said gains. For example, teachers may seek class-size reduction in order to relieve themselves of heavy work, while teacher unions eye the increased recruitment of teachers that would result from implementation of small-class policies. Hoxby criticizes literature supporting the view that class-size reduction achieves significant gains by attacking the assumption that schools run like production lines where there are input-output systems; she terms this a fallacy, claiming that although class reduction does offer increased opportunities for investment in learning, not all schools may take up such opportunities. For experimental studies that offset many questions of bias, Hoxby states that they are few and far between due to their inherent complexity; thus, their results are extrapolated to a large extent and create misconception. The fact that the participants in such studies, for instance the teachers, are aware of the experiment interferes with their natural input; hence, the results may not be reliable. Hoxby's study uses natural population randomness to avoid influencing incentives in a manner dissimilar to policy experiments. The findings are that there is little or no significant gain from reduction of class size in terms of students’ achievement. Hanushek (1998) is also of the view that class size reduction programs enjoy particular popular appeal, while scientific discussion of them is limited and to a large extent selective. The evidence for gains made is scant and unconvincing; in addition, student-teacher ratios have been decreasing dramatically in recent years without accompanying improvement in students’ achievement. The expensive class size reduction policies implemented across the states are mostly ineffective. The enormous efforts put in place to demonstrate improvement in students’ achievements resulting from class-size reduction should have sufficiently settled the debate, but they have surprisingly failed to make a convincing case. Through an evaluation of previous experimental results, econometric studies and international data, Hanushek demonstrates that there is a lack of discernible gains in achievement by students. He undertakes a review of the STAR project dwelling on the findings that significant gains are observed at kindergarten level only and thus concluding that there is no point of adopting small size classes for the rest of the grades. The STAR results indicate that gains made at kindergarten are sustained throughout the other classes without an improvement in case the students remain in small size classes or a decline in case they are returned to normal or large size classes. Hanushek argues that the STAR’s discussions are to a great level extrapolated and theoretical as opposed to being based on the experimental findings. While not disputing the gains from small size classes made at kindergarten level, he attributes these to an initial profound introduction to the school environment- small size classes make it easier for the child to adapt to school faster- rather than improvement in the cognitive skills of a child. Cho et al. (2010) weigh in with a view that the general consensus that reduction in class size has the effect of improving student’s achievement conflicts with academic studies of the subject; these produce conflicting results. In their investigation, they exploited natural population randomness as used by Hoxby (2000) in their work and found that although there were gains in reading and mathematics from reduction of class size in schools in Minnesota, the impact was quite small; a reduction by ten students would result in test score increases of only about 0.04-0.05 standard deviations. This implies that class-size reductions may not result in tangible increase in students’ learning. Their conclusion is that even though class-size reduction might improve students’ achievement, it may not be wise to implement it as a policy since it is quite expensive, especially considering that there are other cost-effective means of achieving similar results. Implications The literature reviewed above indicates that there is still a great deal of conflicting evidence regarding the impact of class-size reduction on students’ achievement. Even for studies that support the view that class-size reduction produces gains in learning, the extent, significance and persistence of said gains in later grades remains in contention. Data analyses from the STAR program indicate that the gains made in kindergarten and the first three grades of elementary school are sustained but do not improve with advancement into higher grades (Pritchard, 1999; Mostella, 1995). Rivkin et al. (2005), on the other hand, argue that the gains made from class-size reduction depreciate with ascension to higher grades. In a different study (Finn et al., 1990, cited in CEC, 2010) found out that the gains made from class-size reduction continue to increase as the students proceeds from grade to grade. Another point of difference is the impact of class-size reduction on minority students. While several studies indicate that reduction in class size has a positive effect on learning for minority students (Robinson & Wittebols, 1986; Mostella, 1995; Krueger, 1999), Rivkin et al. (2005) noted a lack of significant minority students’ improvement resulting from class-size reduction. The main disconnect, however, is in the overall effect of class-size reduction on students’ achievement in elementary school. A good number of studies are of the view that reducing the number of students per class has significant impact in improving students’ learning as evidenced by tests addressing mathematical, reading and science prowess (among several other tests) (Robinson & Wittebols, 1986; Finn et al., 1990; Mostella, 1995; Krueger, 1999; Rivkin et al., 2005; CEC, 2010). Other studies had different results, indicating that there were no substantial gains in terms of students’ achievement resulting from class-size reduction (Hanushek, 1998; Hoxby, 2000; Cho et al., 2010). Applications The implications of the literature review go a long way in determining the recommendations of this report in terms of application of the findings. The majority of the work done, both experimental and analytical, indicates that reduction in class size has a significance effect on improving students’ achievement. Since most of the evidence present points to a positive correlation between class-size reduction and students’ learning, it is recommended that class-size reduction be implemented in elementary schools. Implementation of the STAR program in Tennessee has shown enough proof that gains are to be made, and it provides a template (and sets a benchmark) that can be adopted in other states, or even at county levels. There is, however, a major concern regarding the implementation of class size reduction programs; it should be accompanied by support systems to optimize it and to ensure that tangible gains are made from the large budgets that are involved. A case in point is the class size reduction program that was implemented rather haphazardly in California (Bohrnstedt & Stecher, 2002). To accommodate the reduction in class sizes, new classrooms had to be developed and many teachers recruited. This resulted in a scenario where teachers with little experience and few qualifications were among those recruited, which ultimately led to declines in instructional quality. As a result, the gains observed in students’ achievement cannot be directly attributed to the class size reduction program implemented. This means that before class-size reduction is undertaken, several other considerations have to be factored in; for example, are there sufficient numbers of qualified teachers to meet the demand created by numerous classrooms? Conclusion The state of the education sector is a concern to society, and one of the most debated areas is class-size reduction. This study comprehensively analyses both primary and secondary material that either supports or disputes the view that students’ achievement is significantly affected by class size. It reveals that most literature on the subject leans towards the former case. The information analyzed is evidence from experimental investigations, reviews and analyses of these experimental studies by other authors, meta-analyses and related cluster analyses, and finally analyses of class size policy implementations in several states. The conclusion from this study is that although several debates and questions still persist, results from research point toward significant improvements in students’ achievements when class size is reduced. There is a lack of conclusive evidence on whether the gains are observed at higher grades within elementary school. The gains made, however, do persist after the students from small-size classes are integrated back into regular-size classes. In view of the evidence leaning towards class size reduction as being advantageous, its implementation is recommended. More randomized experimental research on this subject is required in order to settle the debate once and for all. References Bohrnstedt, G. W., & Stecher, B. M., (2002). What we have learned about class size reduction in California. Capstone Report. Sacramento, CA: California Department of Education. Retrieved September 2, 2011 from http://www.scclc.net/attachments/070_Class%20Size%20Redeuction%20Research%20Findings.pdf CEC, (2010). How smaller class size affects student achievement: A literature review of the STAR project and other studies. Retrieved September 2, 2011 from http://www.coffinseducationcenter.com/?p=218 Chetty, R., et al. (2011). How does your kindergarten classroom affect your earnings? Evidence from Project STAR. Retrieved September 2, 2011 from http://obs.rc.fas.harvard.edu/chetty/STAR_slides.pdf Cho, H., et al. (2010). Do Reductions in Class Size Raise Students’ Test Scores? Evidence from Population Variation in Minnesota’s Elementary Schools, Department of Applied Economics University of Minnesota. Retrieved September 2, 2011 from http://faculty.apec.umn.edu/pglewwe/documents/MNclasz3.pdf Cohen, G., et al. (2000). The class size reduction program; Boosting Student Achievement in schools across the nation: A first-year report. U.S. Department of Education. Retrieved September 2, 2011 http://www.myshortpencil.com/schooltalk/_images/classsize.pdf Egelson, P., Harman, P., & Achilles, C. M. (1996). Does class size make a difference? Recent findings from state and district initiatives. Washington, DC: ERIC Clearinghouse. ED 398644. Retrieved September 2, 2011 from http://catalogue.nla.gov.au/Record/5597040 Ferguson, R. F. (1991). Paying for public education: New evidence on how and why money matters. Harvard Journal on Legislation, 28 (2): 465-498. Finn, J.D., et al. (1990). Three Years in a Small Class. Teaching & Teacher Education, 6(2): 127-136. Hanushek, E. A, (1998). The evidence on class size. University of Rochester. Occasional Paper Number 98-1. Retrieved September 2, 2011 from http://edpro.stanford.edu/hanushek/admin/pages/files/uploads/evidence.size.mayer-peterson.pdf Hoxby, C. M. (2000). The effects of class size on student achievement: New evidence from population Variation. Retrieved September 2, 2011 from http://web.missouri.edu/~podgurskym/Econ_4345/syl_articles/Hoxby_Student_Achievement.pdf Krueger, A. B. (1999). Experimental Estimates of Education Production Functions. Quarterly Journal of Economics. 114(2). Retrieved September 2, 2011 from http://www.nber.org/papers/w6051.pdf Kruger, A. B, Hanushek, E. A & Rice, J. K. (2002). The class size debate. Economic Policy Institute. http://edpro.stanford.edu/hanushek/admin/pages/files/uploads/classsizedebate.full%20volume.pdf Molnar, A., et al. (1999). Evaluating the SAGE program: A pilot program in targeted pupil-teacher reduction in Wisconsin. Educational Evaluation and Policy Analysis, Vol. 21, No. 2, pp. 165–177. Molnar, A., Percy, S., Smith, P., & Zahorik, J. (1998). 1997-98 results of the Student Achievement Guarantee in Education (SAGE) Program. Milwaukee, WI: University of Wisconsin-Milwaukee. Mosteller, F., (1995), The Tennessee study of class size in the early school grades. Retrieved September 2, 2011 from https://www.princeton.edu/futureofchildren/publications/docs/05_02_08.pdf Nye, B. A., et al. (1994). The lasting benefits study: A continuing analysis of the effect of small class size in kindergarten through third grade on student achievement test scores in subsequent grade levels: seventh grade (1992-93) technical report. Nashville: Center of Excellence for Research in Basic Skills, Tennessee State University. Pritchard, I., (1999), Reducing class size: What do we know? National Institute on Student Achievement, Curriculum and Assessment, Office of Educational Research and Improvement, U.S. Dept of Education. Retrieved September 2, 2011 from http://www.ed.gov/pubs/ReducingClass/ Rivkin, G. R, et al. (2005). Teachers, Schools, and Academic Achievement. Econometrica, 73(2): 417–458. Retrieved September 2, 2011 from http://www.econ.ucsb.edu/~jon/Econ230C/HanushekRivkin.pdf Robinson, G. E. and Wittebols, J. H. (1986). Class size research: A related cluster analysis for decision-making. Arlington, VA: Education Research Service. Retrieved September 2, 2011 from http://findarticles.com/p/articles/mi_qa3673/is_n4_v114/ai_n28646689/pg_4/ Smith, M. L. & Glass, G. V. (1978). Meta-Analysis of research on the relationship of class-size and achievement. The Class Size and Instruction Project. Retrieved September 2, 2011 from http://www.eric.ed.gov/PDFS/ED168129.pdf Wenglinsky, H. (1997). When money matters: How educational expenditures improve student performance and how they do not. Princeton, NJ: The Educational Testing Service, Policy Information Center. Retrieved September 2, 2011 from http://www.ets.org/Media/Research/pdf/PICWMM.pdf Word, E., et al. (1990). Project STAR: Final executive summary report, kindergarten through third grade (1985-1989). Nashville: Tennessee State Department of Education. Retrieved September 2, 2011 from http://www.heros-inc.org/summary.pdf Read More