xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam a22 u 4500
controlfield tag 008 c19979999azu 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E11-00085
Educational policy analysis archives.
n Vol. 5, no. 16 (July 29, 1997).
Tempe, Ariz. :
b Arizona State University ;
Tampa, Fla. :
University of South Florida.
c July 29, 1997
Staffing up and dropping out : unintended consequences of higher demand for teachers / Mark Fetler.
Arizona State University.
University of South Florida.
t Education Policy Analysis Archives (EPAA)
xml version 1.0 encoding UTF-8 standalone no
mods:mods xmlns:mods http:www.loc.govmodsv3 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govmodsv3mods-3-1.xsd
mods:relatedItem type host
mods:identifier issn 1068-2341mods:part
mods:detail volume mods:number 5issue 16series Year mods:caption 19971997Month July7Day 2929mods:originInfo mods:dateIssued iso8601 1997-07-29
1 of 20 Education Policy Analysis Archives Volume 5 Number 16July 29, 1997ISSN 1068-2341A peer-reviewed scholarly electronic journal. Editor: Gene V Glass Glass@ASU.EDU College of Education Arizona State University,Tempe AZ 85287-2411 Copyright 1997, the EDUCATION POLICY ANALYSIS ARCHIVES.Permission is hereby granted to copy any a rticle provided that EDUCATION POLICY ANALYSIS ARCHIVES is credited and copies are not sold.Staffing Up and Dropping Out: Unintended Consequences of High Demand for Teachers Mark Fetler Abstract Growing public school enrollment and the need to ma intain or improve service to students has increased the demand for te achers, perhaps more rapidly than existing sources can accommodate. While some s chools recruit well qualified teachers by offering higher salaries or better work ing conditions, others may satisfy their need for staff by relaxing hiring standards o r assigning novice teachers to difficult classrooms. Schools' hiring policies have consequences for student success. Dropout rates tend to be higher where facu lties include a greater percentage of minimally educated teachers or teache rs with little experience. The relationship between dropout rate and teacher quali fications is independent of student poverty, school size, and location. A propo sed strategy to reduce dropout rates is to encourage higher preparation and employ ment standards, and to provide appropriate classroom assignments, mentoring, and s upport for new teachers. A systemic view of public schools, while looking fo r stable recurring processes, also recognizes the law of unintended co nsequences. For example, public school enrollment growth has stimulated the need to prepare and hire more teachers. Given compulsory school attendance laws, parent expectations, desired
2 of 20student-teacher ratios, and contractual limits on w ork loads, schools must hire enough teachers to keep pace with growing enrollmen t. Schools may respond to increased need either by offering incentives or by relaxing standards. Of course, schools that successfully entice more desirable can didates with attractive salaries or good working conditions make hiring more difficu lt for others. Similarly, state agencies face pressure to ease the standards and re gulations for teacher preparation and credentialing when confronted with lobbying fro m school boards, administrator groups, and teacher organizations. St ate agencies, schools, colleges, and universities interact, in a loosely defined sup ply system which operates to staff schools. One goal of the system is to maintain trad itionally expected or legally required student-teacher ratios. Although the quali ty of preparation programs is an important aspect of the supply system, meeting scho ol demands for an adequate number of classroom teachers, arguably is a primary goal. The price of preparation tends to constrain the num ber and quality of teachers produced by the supply system. Significant increases in the number of teachers prepared entail greater outlays for traini ng, facilities, and administration, particularly in public universities and schools whe re the bulk of teacher training takes place. The additional coursework and field ex perience associated with more rigorous and thorough teacher preparation generally requires increased costs. Understandably, there is resistance from prospectiv e teachers and policymakers to paying more for teacher preparation if it requires spending less on competing priorities. Educators have long debated systems for instruction in relation to indicators of instructional context, processes, and outcomes. (Levin, 1974; Murnane, 1987; Office of Educational Research and Improvement, 198 8; Shavelson, McDonnell and Oakes, 1989; and Porter, 1991) These systems so metimes underlie proposals for structural reforms of public education which ai m at particular goals. Conventional goals for student success include acad emic achievement, citizenship, or job preparation. Persistence in school to a dipl oma is a minimum goal for at-risk students. The teacher is an essential component of these instructional systems. If it is true that more skilled teachers contribute more effectively to student success, then the education and experience of teachers avail able for hire is an important aspect of instructional systems. While relaxing employment standards at the state or local level in response to high demand addresses reduces tension caused by shortages, it raises the likelihood of hiring less well qualified teachers, with possibly less desirable consequences for student success. This paper examin es teacher education and experience in relation to student dropout rates. It reviews the relationship between student enrollment and teacher demand. It briefly s urveys existing research on the influence of teacher experience and education on st udent achievement, and on factors influencing student decisions to drop out o f school. Actual California high school dropout rates are analyzed and discussed in relation to measures of school size, location, growth, student poverty, teacher ed ucation, and experience. Teacher Supply and Demand Burgeoning student enrollment is an issue for publi c schools, influencing not only the need for adequate facilities, but perhaps even more importantly the need for well-qualified teachers. Nationally, NCES (1996 ) estimates that total K-12 enrollment will grow about 10 percent from 49.8 mil lion in 1994 to 54.6 million by
3 of 202006. In California public school enrollment will r ise over 18 percent from 5.4 million students in 1995 to 6.4 million ten years l ater. (See Note 1 in the Appendix.) California's enrollment growth, coupled with statewide efforts to reduce class sizes, is spurring the demand for teac hers. An estimated 300,000 teachers will be working in 2005, compared to 232,0 00 actually employed in 1996, an expansion of 29 percent. One facet of teacher supply and demand relates to t he skills and abilities expected of teachers. (Darling-Hammond and Hudson,1 990; Reynolds,1991; National Commission on Teaching and America's Futur e,1996; Ashton,1996; Education Week, 1997) The institutional and persona l resources needed to develop those skills influence the rate at which teachers c an be prepared. While teaching expertise is a goal of preparation, usually a crede ntial requires an academic degree and coursework. Although satisfying the requirement s may not guarantee competent performance, it is intended to provide as surance that a teacher is prepared for the classroom. Virtually all public sc hool teachers in the United States have at least a bachelor's degree, and a majority p ossess an advanced degree. (NCES 1995b) The trend is toward higher levels of e ducation. In 1971 28 percent of public school teachers possessed a master's, spe cialist, or doctoral degree. Twenty years later 53 percent of teachers had an ad vanced degree. California's degree and coursework requirements for a preliminary teaching credential generally resemble those of many other s tates. (National Association of State Directors of Teacher Education and Certificat ion, 1996). Unlike a number of states, California requires a Bachelor's degree in a subject other than professional education. Additionally, California teachers comple te a one year preparation program, which provides training in educational pri nciples and teaching strategies. Those who seek a clear credential must fulfill addi tional course requirements and a fifth year of educationally related study after the Bachelor's. California's requirements allow several routes to a credential. Some candidates complete the Bachelor's degree first, then complete the preparat ion program as a graduate student. Others work the preparation program and co urse requirements into their Bachelor's degree in order to receive a preliminary credential. These teachers complete the fifth year of study and remaining requ irements within the next five years. Most teachers who transfer from outside of C alifornia receive a temporary credential based on completion of a Bachelor's degr ee and a professional preparation program. Career changers with at least a Bachelor's degree and competence in their subject of instruction may work as paid teaching interns while they receive support and training in pedagogy from school districts or universities. A second facet of an analysis of teacher supply and demand is the flow of people into and out of public school employment. So urces of credentialed teachers include college and university preparation programs and re entrants from the reserve pool of previously employed teachers. Other sources are school district and university programs to facilitate the mid career tr ansition of people into teaching from jobs in other industries or the military. Nati onally, schools are filling an increasing proportion of vacancies with inexperienc ed applicants. (NCES 1995b) From 1988 to 1991 public schools hired more first-t ime teachers and fewer reentrants or transfers. Reentrants comprised 33 pe rcent of hires in 1988, compared to 24 percent in 1991. First-time teachers made up 31 percent of hires in 1988, compared to 42 percent in 1991. Teachers who transf er from other schools or return to a school have more experience, but receiv e higher salaries than first-time teachers. First-time teachers earn less, but are mo re likely to leave the profession.
4 of 20 Drains on the pool of employed teachers include ret irement and migration into other occupations. Nationwide, between 1990-91 and 1991-92 about 5 percent of teachers left teaching, including retirees. (NCE S 1995a) Teachers with less full-time teaching experience were more likely to l eave. Some 17 percent of those with less than one year of full-time left teaching, compared to 8 percent of those with one year of experience, 7 percent of those wit h two years of experience, and 6 percent of those with three years of experience. Some schools have more turnover than others. (NCES 1995c) Smaller schools experience higher teacher attrition. Based on 1990-91 data, schools with less than 300 enrollment had 10.3 percent turnover, compared to 8.2 percent for those with 300-599 enrollment, and 7.7 percent for those with over 600 enrolled students. Lower salaries and benefits may be a fact or in this relationship. Small schools offer teachers less compensation than large r schools. For example, small public schools paid teachers an average salary of $ 35,317, compared to $42,421 paid by large schools. Student poverty is associate d with teacher turnover. Schools with over 50 percent of students receiving free or reduced price lunches had teacher turnover rate of about 10 percent, compared to an 8 percent rate for schools with lower proportions of such students.Interaction of Preparation and Flow The relationship between teacher preparation and fl ow is complex. Credential requirements restrict access to the teac hing profession. Other conditions remaining equal, higher standards reduce the number of teachers available for work. One way to meet increased demand is to relax the requirements, reducing the time and cost required to become a teacher. For exa mple, when there are too few credentialed applicants, California school district s use emergency permits to hire individuals who lack some requirements for a creden tial, usually proof of competence in their subject(s) of instruction or pe dagogy. (Hart and Burr, 1996) In recent years emergency permits have become more pop ular. A risk of this increased popularity is that less well prepared tea chers may be less effective in their jobs or more prone to attrition. States have sought to increase the supply of teache rs by setting up alternatives to traditional training programs. Zumw alt (1996) describes alternative certification as easing entry requirements, minimiz ing preparation needed prior to paid teaching, and emphasizing on-the-job training. Proponents portray these programs as attracting higher-ability, more diverse experienced people with subject matter majors. (Ashton, 1991; Dill 1996; Fe istritzer, 1994; Haberman, 1992) Zumwalt cautions that it is difficult to gene ralize about the success of alternative programs. The recruitment, preparation and retention of teachers is complex. The underlying assumptions are debatable t h at the knowledge base of teaching is minimal, that schools can supply the ne eded mentoring, and that teaching is a craft best learned on the job. Altern ative approaches assume that school staffs, already criticized for not meeting t he needs of students, have the time, energy, and resources, to support unprepared novice teachers. Plausibly, the success of alternative approaches depends on the ex tent to which novice teachers actually receive needed support and obtain classroo m assignments appropriate to their abilities.Student Performance
5 of 20 Teacher skills and ability influence student achiev ement. Greenwald, Hedges and Lane (1996) reviewed a number of studies of the relationship between school inputs and student outcomes. School resource variab les which described teacher ability, teacher education, and teacher experience were strongly related to student achievement. On the other hand, Hanushek's (1996) s ynthesis of research studies found mixed support for a relationship between scho ol resources and achievement. Although Hanushek did not detect a clear pattern, m easures of teacher experience were more consistently related to achievement than measures of teacher education. Ashton (1996) notes that teachers with regular stat e certification receive higher supervisor ratings and student achievement than tea chers who do not meet standards. Teachers without preparation have troubl e anticipating and overcoming barriers to student learning, and are likely to hol d low expectations for low-income children. Ashton suggests that states which reduce certification requirements or permit the hiring of teachers who do not meet certi fication standards, worsen inequities in the quality of education offered to l ow income children. For example, some alternative certification programs have minima l training requirements with most teachers placed in economically disadvantaged urban schools. Student decisions to drop out, thereby delaying or precluding a high school diploma, represent a facet of performance distingui shable from achievement. While high school graduation is a student attainmen t, it is not a measure of learning. Similarly, while dropping out represents a kind of failure, neither is it a direct measure of achievement. Educators have specu lated on the relationship between instructional systems and student dropout r ates. During the 1980s California and other states implemented broad curri cular and structural reforms aimed at more rigorous academic standards. McDill, Natriello, and Pallas (1985) and Hamilton (1986), among others, speculated that higher standards might result in higher achievement for some students, at the cos t of a narrower curriculum and increased chances of dropping out for at-risk stude nts. Although individuals make decisions to leave school in response to particular circumstances (Ekstrom, Goertz, Pollack, and Rock, 1986; Rumberger, 1987; Venezky, Kaestle, and Sum, 1987), they do so under the influence of a school environm ent. For example, school size and poverty are correlated with dropout rates. (Cib ulka, 1986; Toles, Schulz, and Rice, 1986; Pittman and Haughwout, 1987; Fetler, 19 89) Overall, the national dropout rate is declining, but is higher in large u rban districts. (Schwartz,1995; Coley,1995) In California Guthrie and Kirst (1988) found that s chool reforms resulted in a narrowing of the curriculum without an increased risk of dropping out. Between 1981 and 1986 there were statewide increases in aca demic enrollments, balanced by declines in remedial courses and electives. Scho ols that successfully implemented the reforms tended to focus on an impro ved learning environment, heightened concern for all students, teacher colleg iality, and teacher and site administrator participation in designing reform imp lementation activities. Most of these schools took steps to help at-risk students, and did not experience increased dropout rates. There is large variation in educational attainment in California's adult population. According to the 1990 U.S. Census, 25 p ercent of Californians 18 or older did not graduate from high school, and 23 per cent have just a high school degree. (California Postsecondary Education Commiss ion, 1996) This figure is likely related to California's high rates of immigr ation compared to many other states, as well as school dropout rates. Annual dro pout rates for grades 9 through
6 of 2012 declined in California from 5.2 percent in 1991 to 4.4 percent in 1994. A comparable national dropout rate was 5.3 percent in 1994. Method The unit of analysis was the school. The three type s of information analyzed related to schools, teachers, and students. The dat a were collected from mandated annual surveys administered by the California Depar tment of Education. (See Note 2 in the Appendix for details.) Even in large high schools, employment needs and actions can vary considerably from year to year. In order to obtain relatively stable estimates of teacher education and employment, four year averages of the study measures were computed using data collected from 19 93 through 1996. All analyses were weighted by the average number of tea chers employed in the school in order to accommodate variation in the size. Cali fornia had 805 regular high schools serving an average 1.3 million students per year. The majority (N = 749) of these schools offered instruction in grades 9 throu gh 12, although various other grade configurations were represented, most commonl y 10-12, or 7-12. Approximately 600 alternative high schools serving about 100,000 students per year were excluded from the study. Typically, t hese alternative schools have small enrollments and do not offer the academic cur riculum needed to attend California's public universities. Reasons for refer ral to an alternative school could include an unstable home environment, emotional dif ficulties, pregnancy, etc. Alternative schools diverge from regular schools in serving a population of students with different needs and providing differe nt kinds of services. Measured school characteristics included enrollment and location. The average student enrollment was an indicator of scho ol size. Federally derived categories of school location provided the basis fo r categorizing school location as Large City, Medium City, Urban Fringe, or Rural. Th e research literature links both of these measures with school dropout rates, a nd they are included primarily as controls for the teacher education and employmen t variables. The school level measures of teacher characteristic s included annual growth in the number of employed teachers, the percent of new first-time teachers, the percent of teachers with only a Bachelor's degree, the average number of years of education, and the estimated number of years of exp erience of the teaching staff. Schools with increasing or declining student enroll ment adjust the number of teachers they employ in order to meet the actual ne ed for instruction. High growth schools experience relatively high demand for teach ers both through the need to replace teachers who might ordinarily leave or reti re, and the need to augment their teaching staff to accommodate added students. The a verage net annual growth in the size of the teaching staff is an indicator of t he stability of the faculty. The percent of new first-time teachers and the average years of experience are indicators of teacher experience. The percent of te achers with only a Bachelor's degree and the estimated number of years of educati on are indicators of teacher educational background. The two measures of student characteristics are the percent of students covered by the federal Aid to Families with Depende nt Children (AFDC) program, and the annual dropout rate. AFDC is the percentage of students in the school's attendance area who are enrolled in either public o r private schools and who are from families receiving aid. As an indicator of pov erty AFDC often correlates with student achievement (White, 1982), and functions in this study as a control
7 of 20variable. The annual dropout rate is an indicator o f student performance. The annual dropout rate estimates the percent of studen ts who leave during the course of a year, and is smaller than a cumulative rate wh ich estimates the percent of a cohort leaving over a period of years. Student achi evement uniformly measured by an objective test is another possible measure of st udent performance, but is not currently available in California. One other perfor mance measure was investigated, the percent of students completing the academic cou rse sequence needed to attend a public university. Unfortunately the coursework m easure lacked sufficient reliability to warrant further analysis. One reason for the unreliability may be that the course content associated with a specific title can vary from school to school. The descriptive and correlational techniques used i n this study permit informed speculation about relationships among the phenomena measured by the study variables. Of course, these techniques by the mselves do not justify conclusions regarding cause and effect. Although th e data describe a span of four years, the analyses are cross-sectional, and do not permit the examination of change over time, which is often needed to support causal inference. Results Table 1 illustrates the differences among schools i n various locations. Perhaps the most striking result is the contrast be tween schools located in large cities and those in rural areas. Rural school dropo ut rates are less than a third of large city schools, while student enrollment, the p ercent of teachers with only a Bachelor's degree, and the percent of students in p overty is less than a half of large city schools. Compared to the others, large city sc hools have higher dropout rates, larger student enrollments, a higher percent of stu dents in AFDC families, and a larger percentage of teachers with only a Bachelor' s degree. Rural schools differed from the others, although not so consistently or gr eatly as the large city schools. Rural schools had teachers with fewer years of expe rience, and smaller student enrollments. Rural schools also had teachers with f ewer years of education than the medium city and urban fringe schools. The findings with regard to teacher experience and education in rural schools could pro duce an expectation that rural schools would also have a larger percentage of teac hers with only a Bachelor's degree. Contrary to this expectation, teachers who possess only a Bachelor's degree make up a smaller percentage of the faculty in rura l locations than in other areas.Table 1 Profile of Schools by LocationMeasureLarge City MediumCity UrbanFringe Rural Number of Schools 145106418136 Dropout Rate 7.6(b)3.3(a)2.5(a)1.9(a) Enrollment 2,347(b)2,025(a,b)1,945(a,b)912(a) Faculty Growth 22.214.171.124.9 Percent New Teachers 5.75.05.0(a)5.4 Years of Education 5.55.6(a,b)5.7(a,b)5.4
8 of 20Percent B.A. 18.6(b)9.4(a)9.4(a)8.0(a) Years of Experience 16.2(b)16.5(b)16.5(b)15.0(a) AFDC 23.2(b)15.6(a)12.4(a)11.5(a) Note: Values denoted by (a) differ from those of la rge city schools, and values denoted by (b) differ from those of rural schools, (p < .05), using the Tukey-Kramer HSD comparison method. Table 2 displays statewide means, medians, and stan dard deviations. Evidence that three measures are skewed is the difference be tween the mean and median of the percent of faculty with only a Bachelor's, the perc ent AFDC, and the dropout rate. These differences indicate that the distributions o f schools are skewed so that more schools have lower values of these measures than hi gher values. The lopsided shape of these distributions probably contributes to lowe ring the related correlations and regression coefficients obtained in the following a nalyses. The positive mean and median of the faculty growth indicator is consisten t with the overall growth in California's enrollments. Even so, about one third of high schools reduced the size of their faculty over the four years covered by this s tudy.Table 2 Means, medians, and standard deviations of student, school and teacher measures.MeasureMedianMeanStandard Deviation Student Enrollment (ENR) 2,1121,983853 Urban Location (URB) 00.233.4 Faculty Growth (FGR) 0.61.03.2 Percent New Teachers (PNT) 126.96.36.199 Years of Education (YED) 188.8.131.52 Percent Bachelor's (PBA) 6.611.411.5 Years of Experience (YEX) 184.108.40.206 Percent AFDC 11.415.312.3 Dropout Rate (DOR) 220.127.116.11 Table 3 shows that school average dropout rates are moderately correlated with all study variables, except the annual growth of th e number of faculty. A variable reflecting the urban location of a school was coded with a value of one if the school was situated in an urban setting and coded zero oth erwise. Traditionally, school size
9 of 20and poverty are correlated with dropout rates, whic h is replicated in this study. The average number of years of teacher education and ex perience are negatively correlated with the dropout rate, so that schools w ith more highly educated and experienced teachers tend to have fewer dropouts. T he percent of teachers with only a Bachelor's degree and the percent of new teachers a re positively correlated with the dropout rate, suggesting that schools with minimall y educated, novice teachers tend to have more dropouts. Consistent with expectation, years of teacher experience and education are positively associated with one anothe r, and negatively associated with the percent of new teachers and percent of teachers with only a Bachelor's. That is, schools with more highly educated and experienced t eachers tend to have fewer novice and minimally educated faculty.Table 3 Correlations of student, school and teacher measure s.ENRFGRPNTYEDPBAYEXAFDCDOR URB .25.07.10-.21.34-.03.34.51 ENR .16.02.04.21-.10.26.39 FGR .26.08.13-.18-.09.10 PNT -.23.33-.43.09.26 YED -.57.36-.14-.25 PBA -.21.19.44 YEX -.21-.20 AFDC .51 Note: Correlations with an absolute value greater t han .10 are statistically significant, (p < .01). Even though the indicators of teacher education and experience are significantly correlated with the dropout rate, it is conceivable that school size, location, and AFDC account for both the teacher cha racteristics and the dropout rate. For example, larger schools appear both to hire mor e teachers with only a Bachelor's degree and to have higher dropout rates. Perhaps sc hool size mediates the relationship between teacher education level and the dropout rat e. AFDC is correlated with the teacher education and experience variables, except for percent of new teachers, and could mediate those relationships. If poverty and s chool size can explain teacher education and experience as well as dropout rates, it may be that teacher characteristics have little effect of their own. The multiple regression analysis in Table 4 helps t o assess the influence of each indicator on the dropout rate independently of the influence of other measures. A stepwise regression analysis identified four vari ables which contributed significantly to an explanation of dropout rate. In order of entry these variables were: AFDC, urban location, the percent of teachers with only a Bachelor's, school enrollment, and the percent of new teachers. R-squa re for this analysis was .50, which represents the proportion of variance in school dro pout rates that can be accounted for by the four predictors. While this value of R-squar e is statistically significant, it is possible that greater values could be obtained by u sing additional or more precise information about students, teachers, or schools in the analysis. The standardized
10 of 20betas permit a comparison of the importance of thes e variables in predicting dropouts. AFDC appears to have the greatest impact, with a ch ange of one standard deviation in AFDC related to a change of 2.6 standard deviations in the predicted dropout rate. Urban location of the school trailed AFDC as a pred ictor of dropout rates. Percent of teachers with only a Bachelor's degree and enrollme nt were about equally important, followed by the percent of new teachers.Table 4 Regression Analysis of Dropout RateMeasure Parameter Estimate Standardized Beta Intercept -2.21*0 AFDC 0.11*2.6 Urban Location 2.64*2.2 Percent B.A. 0.07*1.6 Enrollment 0.001*1.6 Percent New Teachers 0.19*1.1 *Significant (p < .001). R-square = 0.50 Table 5 displays a profile of two groups of schools identified as in either the top or bottom ten percent with regard to dropout ra tes. There were 80 schools in each group. The dropout rate of schools in the low group was about one-fortieth of those in the high group. The two groups differed markedly in terms of enrollment and AFDC, consistent with earlier results. The profiles show statistically significant differences between the two groups for the percent of teachers with only a Bachelor's and the percent of new teachers. The percent of teachers wi th only a Bachelor's was about three times larger in the high dropout group compar ed to the low dropout group.Table 5 Profile of Schools with High and Low Dropout RatesMeasureLow Ten Percent High Ten Percent Dropout Rate* 0.312.6 Enrollment* 15792733 Teacher Annual Growth 2.12.0 Percent New Teachers* 4.67.0 Years of Education* 5.75.4 Percent B.A.* 8.524.4 Years of Experience 17.115.5 AFDC* 6.126.9 Differences between means are significant, (p < 001).
11 of 20Discussion This study replicates traditional findings that hig her dropout rates are more likely in larger schools and in poor or urban areas The relationship between poverty and student performance has long been documented. ( White, 1982) The correlates of poverty, perhaps including childhood neglect, lack of family or peer support for education, neighborhood crime, etc. work to decreas e the chances of success in school. People who are struggling to meet more basi c, short term needs of food, shelter, and physical safety probably attend less t o academic development, however much it is in their long term interest. Despite cas es of exceptionally effective schools and teachers, or determined parents and students, a substantial performance gap persists between schools with disadvantaged and tho se with more affluent students. An argument for larger schools is that economies of scale permit more extensive curricula, as well as more efficient use of facilities and equipment than would be possible with smaller enrollments. On the other hand schools are complex institutions and greater size may tend to make scho ols seem more intimidating, less welcoming, and less supportive to students. Larger organizations may face greater administrative challenges to deliver student servic es, provide teacher support, insure discipline, and maintain facilities. Large school s ize may affect more and less advantaged students differently. Those students who are at-risk of failure and dropping out may become discouraged in larger, more impersonal situations, without sufficient guidance and support. High achieving, mo re advantaged students may be more able to benefit on their own from the offering s of larger schools. One facet of research on effective schools is to id entify factors which help students overcome disadvantages. This study confirm s prior findings that teacher experience and education are two such factors. Whil e earlier research emphasized student academic achievement, this study goes furth er by looking at students who are at-risk of dropping out. Teacher education and expe rience appear to influence dropout rates. The smaller the proportion of inexperienced teachers who are new to a school, the lower the dropout rate. The smaller the percent of teachers with only a Bachelor's degree, the lower the dropout rate. This influence appears to hold independently of poverty, and school size, and location. The stepwise procedure did not identify faculty gro wth, years of teaching, or overall years of education, so that these variables were excluded from the regression analysis. The results of a regression analysis can be sensitive to the inclusion of highly correlated pairs of variables, a situation d escribed as multicollinearity. For example, the percent of teachers with only a Bachel or's and overall years of education are correlated with each other. However, preliminar y analyses which included overall years of education in the model did not appreciably change the results for the other variables, suggesting that multicollinearity is not an issue. Conceptually, it appears that these two resource variables, the overall year s of education and the percent of teachers with only a Bachelor's, represent distingu ishable characteristics of a school faculty. Notably, years of teaching and years of education a re less strongly associated with the dropout rate than the percent of new teach ers or the percent with only a Bachelor's degree. One possible reason for the diff erence may exist in the assignments that some schools typically give new te achers. It is commonly thought th at more experienced teachers with seniority usually obtain more desirable classroom assignments with well behaved, higher achieving stu dents. Novice teachers lacking seniority receive less desirable, more difficult cl assrooms with lower achieving,
12 of 20at-risk students. Novice teachers are more likely t o have minimum levels of education, particularly if the school district has lowered its hiring standards to maintain staffing levels. Arguably, at-risk student s are most in need of intensive, skilled instruction, counseling, and support. Less well educated novice teachers are the least capable members of faculty able to provid e the services and support needed to forestall student dropout. Ironically, as novice teachers gain experience and education, they also gain the seniority which enabl es them to opt out of the difficult assignments. While this pattern of assigning classr ooms could exist in many schools, the effects are probably more severe in hard to sta ff schools with a high proportion of disadvantaged, at-risk students. Such unpopular sch ools probably have difficulty in retaining their more experienced and educated staff who move on to more attractive work sites. With lower levels of education and expe rience at a school there is less capability to support and train new teachers. The r esults of the analyses suggest that this assignment pattern and negative consequences m ay be more prevalent in urban settings.Conclusion Schools are complex institutions comprised of disti nct yet interdependent systems. These systems can work in alignment to sup port student success. To the extent the systems are out of alignment they will b e less supportive. Of course, some parts of these systems are beyond the control of sc hools so that perfect alignment is an unrealistic goal. The system which supplies teac hers, depends not only on school administration and personnel offices, but also on e mployee organizations, state regulatory agencies, college and university teacher training programs, and on the labor market decisions of individuals who choose jo bs according to their own interests. The instructional system includes not on ly teachers, curriculum, classrooms, textbooks, and materials, but also the community so cial and economic context. Parents and peers may support academic values and a ctivities, or undermine them. Despite the unruly nature of these systems, educato rs and policy makers at schools, colleges and in government can strive to w ork together toward a common vision of student success. Ideally, new teachers wh o are hired to maintain or improve student teacher ratios will have the experience and education needed to support the success of all students, including the disadvantage d and those who are at-risk of dropping out. If new teachers lack some these skill s, they should receive appropriate classroom assignments, mentoring, and support. Main taining a student teacher ratio by accepting lower standards for experience and edu cation does not optimally support student success. Assigning difficult classrooms to unprepared, unsupported, novice teachers additionally threatens at-risk students. W hile accepting lower standards may be expedient in the short term, there are indirect costs including the personal costs to an individual of a decision to drop out and the bur den imposed by dropouts on the public. In the long run, students, educators, and t he public are better served by insisting on higher standards, and by providing the resources and the will to implement them.ReferencesAshton, P. (Ed.) (1991). Alternative approaches to teacher education. Journal of Teacher Education, 42 (2), 82.
13 of 20Ashton, P. (1996). Improving the preparation of tea chers. Educational Researcher. Vol. 25, No. 9, pp. 21-22.California Department of Education. (1996). Adminis trative Manual for CBEDS Coordinators and School Principals. Sacramento: Aut hor. California Postsecondary Education Commission. (199 6). Performance Indicators of Higher Education. Sacramento: Author, Commission Re port 96-10. Cibulka, M. (1986). State level policy options for dropout prevention. Metropolitan Education. 2, 30-38.Coley, R. (1995). Dreams deferred: High school drop outs in the United States. Princeton: Educational Testing Service, Policy Info rmation Center. Darling-Hammond, L., and Hudson, L. (1990). Precoll ege science and mathematics teachers: Supply, Demand, and Quality. In Review of Research in Education (Vol. 16). Washington, DC: American Educational Research Association. Dill, V. (1996). Alternative teacher certification. In J. Sikula (Ed.), Handbook of Research on Teacher Education (2nd edition, pp. 932 -957). New York: Macmillan. Education Week. (1997). Quality Counts: A Report Ca rd on the Condition of Public Education in the 50 States. Author, Vol. XVI, Janua ry 22, 1997. Ekstrom, R., Geortz, M., Pollack, J., and Rock, D. (1986). Who drops out of school and why? Findings from a national study. Teachers C ollege Record. 87 (6), 356-373. Feistritzer. C. (1994). The evolution of alternativ e teacher certification. The Educational Forum. 58 (2), 132-138.Fetler, M. (1989). School dropout rates, academic p erformance, size and poverty: Correlates of educational reform. Educational Evalu ation and Policy Analysis. 11 (2), 109-116.Fetler, M. (1997). Where have all the teachers gone ? Education Policy Analysis Archives. Vol. 5 No. 2. A peer-reviewed scholarly e lectronic journal available at http://olam.ed.asu.edu/epaa/Grossman, P. (1990). The making of a teacher. New Y ork: Teachers College. Guthrie, J. And Kirst M. (1988). Conditions of Educ ation in California. Berkeley, CA: University of California Policy Analysis for California Education. Policy paper No. PP88-3-2, pp. 125 140.Haberman, M. 1992). Alternative certification: Can the real problems of urban education be resolved by traditional teacher educat ion? Teacher Education and Practice. 8 (1), 13-27.Hamilton, S. (1986). Raising standards and reducing dropout rates. Teachers College record. 87(3), 410-429.
14 of 20Hammack, F. (1986). Large school systems/ dropout r eports: An analysis of definitions, procedures and findings. Teachers Coll ege Record. 87 (3), 324 341. Hanushek, E. (1996). A more complete picture of sch ool resource policies. Review of Educational Research. Vol. 66, No. 3, pp. 397-409.Hart, G. And Burr, S. (1996). A State of Emergency ... In a State of Emergency Teachers. Sacramento: California State University I nstitute for Education Reform. Levin, H. (1974). A conceptual framework for accoun tability in education. School Review, 82, (3). pp. 363-392.McDill, E., Natriello, F., and Pallas, A. (1985). R aising standards and retaining students: The impact of reform and recommendations on potential dropouts. Review of Educational Research. 55, 415-433.Murnane, R. (1987). Improving education indicators and economic indicators: The same problems? Educational Evaluation and Policy An alysis, 9, (2), pp. 101-116. National Association of State Directors of Teacher Education and Certification (1996). Manual on Certification and Preparation of Educational Personnel in the United States. Dubuque, Iowa: Kendall Hunt.National Commission on Teaching and America's Futur e (1996). What Matters Most: Teaching for America's Future. New York, Teachers C ollege, Columbia University. National Center for Educational Statistics. (1996). Projections of Education Statistics to 2006. Washington DC: Author.National Center for Education Statistics. (1995a). Digest of Education Statistics. Washington, D.C.: U. S. Department of Education Off ice of Educational Research and Improvement.National Center for Education Statistics. (1995b). America's Teachers Ten Years After "A Nation at Risk." Washington, D.C.: U. S. D epartment of Education Office of Educational Research and Improvement, NCES 95-76 6. National Center for Education Statistics. (1995c). Which Types of Schools Have the Highest Teacher Turnover? Washington, D.C.: U. S. D epartment of Education Office of Educational Research and Improvement, NCES 95-77 8. Office of Educational Research and Improvement. (19 88). Creating Responsible and Responsive Accountability Systems: Report of the OE RI State Accountability Study Group. Washington, DC.: U. S. Department of Educati on. Pittman, R. And Houghwout, P. (1987). Influence of high school size on dropout rate. Educational Evaluation and Policy Analysis. 9 337 343. Porter, A. (1991). Creating a system of school proc ess indicators. Educational Evaluation and Policy Analysis, 13, (1), pp. 13-29.
15 of 20Rumberger, R. (1987). High school dropouts: A revie w of issues and research. Review of Educational Research, 57, 101 121.Schwartz, W. (1995). School dropouts: New informati on about an old problem. Washington, DC: U.S. Department of Education, Offic e of Research and Improvement. (ED 386515)Shavelson, R., McDonnell, L., and Oakes, J. (Eds.) (1989). Indicators for Monitoring Mathematics and Science Education: A Sourcebook. Sa nta Monica, CA: The RAND Corporation.Toles, T., Schulz, E., and Rice, W. (1986). A study of variation in dropout rates attributable to effects of high schools. Metropolit an Education. 2, 30 38. Venezky, R., Kaestle, C., and sum, A. (1987). The s ubtle danger: Reflections on the literacy abilities of America's young adults. Princ eton, NJ: Educational Testing Service.White, K. (1982). The relation between socioeconomi c status and achievement. Psychological Bulletin, 91, 461-481.Zumwalt, K. (1996). Simple Answers: Alternative Tea cher Certification. Educational Researcher, Vol. 25, No. 8, 40-42.AppendixNote 1. Enrollment and Teachers The actual graded K-12 enrollment and numbers of te achers shown in Table 6 are from the California Basic Educational Data Syst em (CBEDS) of the California Department of Education. Enrollments are published annually in a document entitled "California Public Schools Enrollment." Counts of t eachers are published annually in a document entitled "Count of Certificated and Clas sified Staff in California Public School Districts." The teacher counts reflect certi ficated staff with classroom assignments, and exclude administrators and pupil s ervices staff.Table 6 Actual and Projected Enrollment and Teaching StaffSchool YearGraded K-12 EnrollmentTeachersClass Size ReductionProgram 1991-925,001,670219,3531992-935,089,808220,8711993-945,166,261223,9321994-955,242,078228,2041995-965,367,926232,4881996-975,495,075238,951259,000
16 of 201997-985,623,422244,532266,0001998-995,737,874249,509269,0001999-005,841,535254,017274,0002000-015,945,067258,519279,0002001-026,052,242263,179284,0002002-036,160,231267,875289,0002003-046,271,881272,730295,0002004-056,392,367277,969300,000 The projections of graded public school enrollment are published by the California Department of Finance Demographic Resear ch Unit in a document entitled "K-12 Graded Public School Enrollment by Ethnicity, History, and Projection 1995 Series." The projections are based on a grade-progr ession ratio (or cohort survival) projection method and the most recent ten years of historical enrollment data from CBEDS. The actual average ratio of K-12 pupils to classroo m teachers from 1991 92 through 1995-96 is 23 to 1. The projected numbers o f teachers from 1996-97 onward assume continuation of the 23 to 1 student teacher ratio. During the fall of 1996 the California Legislature enacted a program, which giv es incentives to school districts to reduce class size in three elementary grades. Under this program there is a limit of twenty students in a "class." An estimated 20,000 a dditional teachers are needed to fully implement this program, which represents abou t an eight percent increase in the size of the work force. The projected number of tea chers under the Class Size Reduction Program is calculated by applying an eigh t percent increase to the original projections. Over the next ten years, with class si ze reduction, the teaching workforce should increase in size by 68,000, which is about 3 0 percent growth. Fetler (1997) provides a more detailed discussion of enrollment g rowth in relation to the supply of teachers.Note 2. Sources of Data The annual Professional Assignment Information Form (PAIF), administered as a part of California Basic Educational Data Syst em, was the source of teacher descriptive measures for each school, including: nu mber of teachers, years of experience, percent new teachers, teacher education al level, percent of teachers with only a Bachelors degree, and growth in teacher staf fing levels. School level personnel actions can vary from year to year, depending on lo cal policies, resources, and needs. In order to improve the stability of the employment related measures, four year averages (1993 through 1996) of these measures were computed. Years of experience is defined as the total years o f public and/or private educational service. Percent new teachers is computed as the number of t eachers with no previous educational service divided by the total number of teachers in the school. Teacher years of education is a computed index. Tea chers' responses were coded as follows: Bachelors degree = 4, Bachelors p lus 30 semester hours = 5, Masters degree = 6, Masters plus 30 semester hours = 7, and Doctorate = 8. The coded responses were averaged to produce the in dex. The percent of teachers with only a Bachelors degre e is computed as the
17 of 20number of teachers with only a Bachelors degree div ided by the total number of teachers in the school.The growth in the number of teachers was computed a s the value of the slope coefficient of a four year linear trend. This coeff icient is interpreted as the average yearly change in the number of teachers emp loyed at the school. CBEDS high school profile data sets were the source of measures of school characteristics, including the school dropout rate and the proportion of graduates meeting the public university course requirements, and the percent of students receiving Aid to Families with Dependent Children. Four year averages (1993 through 1996) of these measures were computed in or der to improve stability. These school summary files are available at the web site hosted by the California Department of Education: http://goldmine.cde.ca.gov/ftpbranch/retdiv/demogra phics/Demohome.html The school dropout rate is computed as the number o f reported dropouts in grades 9 through 12 divided by the eligible number enrolled. According to CBEDS (1996) a dropout is a person who meets the followin g criteria:was formerly enrolled in grades 7, 8, 9, 10, 11, or 12has left school for 45 consecutive school days and has not enrolled in another public or private educational institution or school programhas not re-enrolled in the schoolhas not received a high school diploma or its equiv alentwas under twenty-one years of agewas formerly enrolled in a school or program leadin g to a high school diploma or its equivalent. The percent receiving AFDC is the percentage of stu dents in the school's attendance area who are enrolled in either public o r private schools and who are from families receiving Aid to Families with Dependent C hildren (AFDC). This measure was taken from the high school performance data set s available at the same web address. The classification of schools into demographic regi ons was derived from information contained in the 1994-95 Public School Name and Address File disseminated by the National Center for Education S tatistics (1996). http://www.ed.gov/NCES/ccd/index.html This file contains the names and addresses of the 8 9,151 public schools in the 50 states, District of Columbia, and five outlying areas for 1994-95. Also included on each record is the School's enrollment (membership) and various other codes. These codes comprise school type, lowest and highest grad es taught, and school locale. The locale code is a definition of how the school i s situated in a particular location relative to populous areas, based on the s chool's mailing address. The Code translations are as follows:1= Large Central City2= Mid-size Central City3= Urban Fringe of Large City4= Urban Fringe of Mid-Size City5= Large Town
18 of 20 6= Small Town7= Rural The definitions for locales are: Large City: Central city of a Metropolitan Statisti cal Area (MSA) with a population greater than or equal to 400,000 or popu lation density greater than or equal to 6,000 people per square mile. Mid-size City: Central City of an MSA with a popula tion less than 400,000 and a population density less than 6,000 people per squ are mile. Urban Fringe of Large City: Place within an MSA of a Large Central City and defined as urban by the Census Bureau. Urban Fringe of Mid-size City: Place within an MSA of a Mid-size Central City and defined as urban by the Census Bureau. Large Town: Town not within an MSA, with a populati on greater than or equal to 25,000. Small Town: Town not within an MSA and with a popul ation less than 25,000 and greater than or equal to 2,500 people. Rural: A place with less than 2,500 people and code d rural by the Census Bureau. Only a small number of high-schools were in the "Sm all Town" category, which was therefore merged into the "Rural" categor y. The two "Urban Fringe" categories were a lso combined as were the "Large C ity" and Large Town" categories after verifying that they were similar in terms of the other study variables.About the AuthorMark FetlerCalifornia Department of Education721 Capitol MallSacramento, CA 95814(916) 657-4267 firstname.lastname@example.org Employment Experience Consultant. California Department of Education. 199 7 present. Coordinate the development statewide assessment standards and tests administered under the Golden State Examination Program. Consultant. California Commission on Teacher Creden tialing. 1995 1997. Plan, organize and conduct research on teacher cred ential examinations and teacher supply and demand. Director. Planning, Effectiveness and Accountabilit y Unit, Chancellor's Office, California Community Colleges. 1990 1995. Manage accountability task force. Develop federal and state accountabilit y programs. Administrator. Educational Planning and Information Center, California
19 of 20 Department of Education. 1984 1990. Manage and de velop K-12 accountability policies and programs.Consultant. California Assessment Program, Californ ia Department of Education. 1980 1984. Conduct research and assess ment projects. Evaluation Specialist. Northwest Regional Education al Laboratory, Portland, Oregon. 1979 1980. Deliver evaluation technical a ssistance and training to state education agencies. Senior Staff Associate. Western Interstate Commissi on for Higher Education, Boulder, Colorado, 1978 1979. Evaluate research i nstruments. EducationB.A., Pyschology, Colorado College, 1972Ph.D., Psychology, University of Colorado, 1978Study abroad in Germany at the Universities of Gtt ingen and Bielefeld Publications Authored research in peer-reviewed journals, e.g., American Educational Research Journal, Educational Evaluation and Policy Analysis, Applied Measurement in Education, Sex Roles and the Journal of Communication Community Service Volunteer for Elk Grove Unified School District, Bo ard member of the Elk Grove Community Planning Advisory Council, fund rai ser for the Strauss Festival of Elk Grove, and member of the Elk Grove Rotary Club.Copyright 1997 by the Education Policy Analysis ArchivesThe World Wide Web address for the Education Policy Analysis Archives is http://olam.ed.asu.edu/epaa General questions about appropriateness of topics o r particular articles may be addressed to the Editor, Gene V Glass, email@example.com or reach him at College of Education, Arizona State University, Tempe, AZ 8528 7-2411. (602-965-2692). The Book Review Editor is Walter E. Shepherd: firstname.lastname@example.org The Commentary Editor is Casey D. Cobb: email@example.com .EPAA Editorial Board Michael W. Apple University of Wisconsin Greg Camilli Rutgers University John Covaleskie Northern Michigan University Andrew Coulson firstname.lastname@example.org
20 of 20 Alan Davis University of Colorado, Denver Sherman Dorn University of South Florida Mark E. Fetler California Commission on Teacher Credentialing Richard Garlikov email@example.com Thomas F. Green Syracuse University Alison I. Griffith York University Arlen Gullickson Western Michigan University Ernest R. House University of Colorado Aimee Howley Marshall University Craig B. Howley Appalachia Educational Laboratory William Hunter University of Calgary Richard M. Jaeger University of North Carolina--Greensboro Daniel Kalls Ume University Benjamin Levin University of Manitoba Thomas Mauhs-Pugh Green Mountain College Dewayne Matthews Western Interstate Commission for Higher Education William McInerney Purdue University Mary P. McKeown Arizona Board of Regents Les McLean University of Toronto Susan Bobbitt Nolen University of Washington Anne L. Pemberton firstname.lastname@example.org Hugh G. Petrie SUNY Buffalo Richard C. Richardson Arizona State University Anthony G. Rud Jr. Purdue University Dennis Sayers University of California at Davis Jay D. Scribner University of Texas at Austin Michael Scriven email@example.com Robert E. Stake University of Illinois--UC Robert Stonehill U.S. Department of Education Robert T. Stout Arizona State University