Time for school? When the snow falls, test scores also drop.

Time for school? When the snow falls, test scores also drop. Students in the United States spend much less time in school thando students in most other industrialized nations, and the school yearhas been essentially unchanged for more than a century. This is not tosay that there is no interest in extending the school year. While therehas been little solid evidence that doing so will improve learningoutcomes, the idea is often endorsed. U.S. Secretary of Education ArneDuncan has made clear his view that "our school day is too short,our week is too short, our year is too short." [ILLUSTRATION OMITTED] Researchers have recently begun to learn more about the effects oftime spent on learning from natural experiments around the country. Thisnew body of evidence, to which we have separately contributed, suggeststhat extending time in school would in fact likely raise studentachievement. Below we review past research on this issue and thendescribe the new evidence and the additional insights it provides intothe wisdom of increasing instructional time for American students. We also discuss the importance of recognizing the role ofinstructional time, explicitly, in accountability systems. Whether ornot policymakers change the length of the school year for the averageAmerican student, differences in instructional time can and do affectschool performance as measured by No Child Left Behind. Ignoring thisfact results in less-informative accountability systems and lostopportunities for improving learning outcomes. Emerging Evidence More than a century ago, William T. Harris in his 1894 Report ofthe Commissioner [of the U.S. Bureau of Education] lamented, The boy of today must attend school 11.1 years in order to receive as much instruction, quantitatively, as the boy of fifty years ago received in 8 years. ... It is scarcely necessary to look further than this for the explanation for the greater amount of work accomplished ... in the German and French than in the American schools. The National Education Commission on Time and Learning would echohis complaint one hundred years later. But the research summary issuedby that same commission in 1994 included not one study on the impact ofadditional instruction on learning. Researchers at that time simply hadlittle direct evidence to offer. The general problem researchers confront here is that length of theschool year is a choice variable. Because longer school years requiregreater resources, comparing a district with a long school year to onewith a shorter year historically often amounted to comparing a richschool district to a poor one, thereby introducing many confoundingfactors. A further problem in the American context is that there islittle recent variation in the length of school year. Nationwide,districts generally adhere to (and seldom exceed) a school calendar of180 instructional days. And while there was some variation in the firsthalf of the 20th century, other policies and practices changedsimultaneously, making it difficult to uncover the separate effect ofchanges in instructional time. Among the first researchers to try to identify the impact ofvariation in instructional time were economists studying the effect ofschooling on labor market outcomes such as earnings. Robert Margo in1994 found evidence suggesting that historical differences inschool-year length accounted for a large fraction of differences inearnings between black workers and white workers. Using differences in the length of the school year acrosscountries, researchers Jong-Wha Lee and Robert Barro reported in 2001that more time in school improves math and science test scores. Oddly,though, their results also suggested that it lowers reading scores. In2007, Ozkan Eren and Daniel Millimet examined the limited variation thatdoes exist across American states and found weak evidence that longerschool years improve math and reading test scores. Work we conducted separately in 2007 and 2008 provides muchstronger evidence of effects on test scores from year-to-year changes inthe length of the school year due to bad weather. In a nutshell, wecompared how specific Maryland and Colorado schools fared on slateassessments in years when there were frequent cancellations due tosnowfall to the performance of the very same schools in relatively mildwinters. Because the severity of winter weather is inarguably outsidethe control of schools, this research design addresses the concern thatschools with longer school years differ from those with shorter years(see research design sidebar). While our studies use data from different states and years, andemploy somewhat different statistical methods, they yield very similarresults on the value of additional instructional days for studentperformance. We estimate that an additional 10 days of instructionresults in an increase in student performance on state math assessmentsof just under 0.2 standard deviations. To put that in perspective, thepercentage of students passing math assessments falls by about one-thirdto one-half a percentage point for each day school is closed. Other researchers have examined impacts of instructional time onlearning outcomes in other states, with similar results. For example,University of Virginia researcher Sarah Hastedt has shown that closuresthat eliminated 10 school days reduced math and reading performance onthe Virginia Standards of Learning exams by 0.2 standard deviations, thesame magnitude we estimate for the neighboring state of Maryland.Economist David Sims of Brigham Young University in 2008 took advantageof a 2001 law change in Wisconsin that required all school districts inthat state to start after September 1. Because some districts wereaffected while others were not, he was also able to provide unusuallyconvincing evidence on the effect of changes in the number ofinstructional days. He found additional instruction days to beassociated with increased scores in math for 4th-grade students, thoughnot in reading. Collectively, this emerging body of research suggests thatexpanding instructional time is as effective as other commonly discussededucational interventions intended to boost learning. Figure 1 comparesthe magnitude of the effect of instructional days on standardized mathscores to estimates drawn from other high-quality studies of the impactof changing class size, teacher quality, and retaining students ingrade. The effect of additional instructional days is quite similar tothat of increasing teacher quality and reducing class size. The impactof grade retention is comparable, too, though that intervention ispertinent only for low-achieving students. [FIGURE 1 OMITTED] Although the evidence is mounting that expanding instructional timewill result in real learning gains, evidence on the costs of extendingthe school year is much scarcer and involves a good deal of conjecture.Perhaps the best evidence comes from a recent study in Minnesota, whichestimated that increasing the number of instructional days from 175 to200 would cost close to $1,000 per student, in a state where the medianper-pupil expenditure is about $9,000. The total annual cost wasestimated at $750 million, an expense that proved politically andfinancially infeasible when the proposal was recently considered in thatstate. Comparing costs of expanding instructional days with the costs ofother policy interventions will be an analytic and policy exercise ofreal importance if the call for expanded instructional time is to resultin real change. Complicating this analytic task are differences in costs that existacross schools and states. Utilities, transportation, and teachersummer-labor markets vary widely across geographic areas, and all affectthe cost of extending the school year. So, while the benefits ofextending the school year may exceed the costs in some states or schooldistricts, they may not in others. A further complication is thepossibility of diminishing returns to additional instructional time. Ourresearch has studied the effect of additional instructional days priorto testing, typically after approximately 120 school days. The effect ofextending instructional time into the summer is unknown. Also, ourresearch has focused on the variation in instructional days prior toexams, or accountable days. The effect of adding days after exams couldbe quite different. Costs of extending school years are as much political as economic.Teachers have come to expect time off in the summer and have been amongthe most vocal opponents of extending school years in several locations.Additional compensation could likely overcome this obstacle, but howmuch is an unresolved and difficult question. Teachers are not the only ones who have grown accustomed to asummer lasting from June through August. Students and families havecamps, vacations, and work schedules set up around summer vacation."Save Our Summers" movements have for years decried thebenefits of additional instructional days and proclaimed the benefits ofsummer vacation, and the movements have grown as states have consideredextending the school year and individual school districts have moved uptheir start dates. Longer school years might reduce tourism and itsaccompanying tax revenue. These additional costs likely vary by stateand district, but are clearly part of the analytic and politicalcalculus. Time and Accountability As education policymakers consider lengthening the school year andface tradeoffs and uncertainties, it is important to recognize thatexpanding instructional time offers both opportunities and hazards foranother reform that is well established, the accountability movement.Educators, policymakers, parents, and economists are sure to agree thatif students in one school learn content in half the time it takescomparable students at another school to learn the same content, thefirst school is doing a better job. How students would rank theseschools is equally obvious. Yet state and federal accountability systemsdo not account for the time students actually spent in school whenmeasuring gains, and so far have no way of determining how efficientlyschools educate their students. One implication of this oversight is that accountability systemsare ignoring information relevant to understanding schools'performance. Year-to-year improvements in the share of studentsperforming well on state assessments can be accomplished by changes inschool practices, or by increases in students' exposure to school.Depending on the financial or political costs of extending school years,those with a stake in education might think differently about gainsattributable to the quality of instruction provided and gainsattributable to the quantity. To see how the contributions of these inputs might be separated,consider data from Minnesota. Between 2002 and 2005, 3rd graders in thatstate exhibited substantial improvements in performance on mathassessments, a fact clearly reflected by Minnesota's accountabilitysystem. But during that period, there was substantial year-to-yearvariation in the number of instructional days students had prior to thetest date. In Figure 2, we plot both the reported test scores forMinnesota 3rd graders (the solid line) and the number of days ofinstruction those students received (the bars). Useful, and readilycalculated, is the time series of test scores, adjusting for differencesin the number of instructional days (the dotted line). [FIGURE 2 OMITTED] Comparing the reported and adjusted scores is useful for at leasttwo reasons. First, it illustrates the role of time as a component oftest gains. Overall, scale scores increased by 0.4 standard deviationsfrom 2001-02 to 2004-05. Of this increase, a large portion wasattributable to expansion in instructional time prior to the test date.Adjusting for the effect of instructional days, we estimate that scoresincreased by roughly 0.25 standard deviations, nearly 40 percent lessthan the reported gains. Second, the comparatively steady gain in adjusted scores over theperiod provides evidence of improvements in instructional quality,independent of changes in the amount of time students were in class. Thefast year-to-year increases in the first and last periods result inlarge part from increases in the amount of time in school, while thenegligible change in overall scores between 2003 and 2004 does not pickup real gains made despite a shortened school year. Adjusted scores pickup increases in learning gains attributable to how schools usedinstructional time, such as through changing personnel, curricula, orleadership. The point here is that time-adjusted scores provideinformation that is just as important as the overall reported scores forunderstanding school improvements. A robust accountability system wouldrecognize that more instructional time can be used to meet goals, butthat more time is neither a perfect substitute for, nor the same thingas, better use of time. The Hazards of Ignoring Time Failing to account for the role of time in student learning notonly means missed opportunity, it also creates potential problems.First, it can allow districts to game accountability systems byrearranging school calendars so that students have more time in schoolprior to the exam, even as the overall length of the school year remainsconstant. Beginning in the 1990s, districts in a number of states beganmoving start dates earlier, with many starting just after the first ofAugust. The question arose whether these changes might be linked topressures on districts to improve performance on state assessments.David Sims showed that Wisconsin schools with low test scores in oneyear acted strategically by starting the next school year a bit earlierto raise scores. Evidence of gaming soon emerged in other states aswell. Wisconsin passed its 2001 law requiring schools to begin alterSeptember 1 to prevent such gaming; similar laws were recently passed inTexas and Florida. The motives driving earlier start dates could spill over into otherinstructional policies. Minnesota moved its testing regimen fromFebruary to April in the wake of accountability standards, whileColorado legislators have proposed moving their testing window fromMarch into April, with advocates suggesting that the increased time forinstruction would make meeting performance requirements under No ChildLeft Behind more feasible for struggling schools. While administeringthe test later in the year has potential benefits in measuredperformance, grading the tests over a shorter time frame costs more,estimated at some $3.9 million annually in Colorado. Schools thussacrifice educational inputs (such as smaller classes or higher teachersalaries) to pay for the later test date. A second hazard involves fairness to schools at risk of beingsanctioned for poor performance: these schools can face longer odds ifweather or other schedule disruptions limit school days. The impact ofinstructional time on learning means that one factor determining theability of schools to meet performance goals is not under the control ofadministrators and teachers. We illustrate the effects of time on makingadequate yearly progress (AYP) as defined by No Child Left Behind bycomparing the performance of Maryland schools the law identified asunderperforming to estimates of what the performance would have been hadthe schools been given a few more days for instruction. We begin with data from all elementary schools in Maryland that didnot make AYP in math and reading during the 2002-03 to 2004-05 schoolyears. We adjust actual performance by the number of days lost in agiven year multiplied by the marginal effect of an additional day ontest performance as reported in Marcotte and Hemelt's study ofMaryland schools. This allows us to estimate what the proficiency ratesin each subject would have been had those schools been open for allscheduled instructional days prior to the assessment. We then comparethe predicted proficiency rate to the AYP threshold. We summarize the results of this exercise in Figure 3. The lightbars represent the number of schools failing to make AYP in math andreading in various years. The dark bars are the number of those schoolsthat we predict would have failed to make AYP if the schools had beenable to meet on all scheduled days. We make these estimates assumingthat low-performing schools would have made average gains with eachadditional day of instruction. [FIGURE 3 OMITTED] The average number of days lost to unscheduled school closingsvaried substantially over the period, from more than 10 to fewer thanfour and a half. Many schools that did not make AYP likely would havehad they not lost so many school days. For example, we estimate that 35of the 56 elementary schools that did not make AYP in math in 2002-03would have met the AYP criterion if they had been open during allscheduled school days. Even if these schools were only half asproductive as the typical school, 24 of the 56 flagged schools wouldlikely have made AYP if they had been open for all scheduled days. There is, however, a way to reduce risks like these for schools andto limit incentives for administrators to move start or test dates atthe same time: that is to recognize and report time as an input ineducation. A simple and transparent way to do this is for state reportcards, which inform parents about school outcomes and summarize theinformation on AYP status, to include information about the number ofinstructional days at test date as well as the total number ofinstructional days for the year. This information is readily availableand already monitored by schools, districts, and states. Local and stateeducation authorities could use it when assessing performance, forexample, in hearing an appeal from a school that failed to meet its AYPgoals. Further, this information could be used to estimate test scoresadjusted for instructional days, to be used alongside unadjusted changesin performance. Distinguishing between gains due to expanded instructiontime and better use of that time can enrich accountability systems andprovide more and better information to analysts and the public alike. Looking Ahead There can be no doubt that expanding the amount of time Americanstudents spend in school is an idea popular with many educationpolicymakers and has long been so. What makes the present different isthat we now have solid evidence that anticipated improvements inlearning will materialize. Practical obstacles to the extension of the school year includesubstantial expense and stakeholder attachment to the current schoolyear and summer schedule. The benefits of additional instructional dayscould diminish as school years are lengthened. Further, it is unknownhow teachers would use additional instructional days if they areprovided after annual testing is already finished. Simply extending theyear well after assessments are given might mean that students andteachers spend more days filling (or killing) time before the end of theyear. This would make improvements in learning unlikely, and presumablymake students unhappy for no good reason. Though the issue has seen little movement in the past and facesreal opposition going forward, the policy climate appears likely to befavorable once the fiscal challenges now facing public school systemsrecede. It is our hope that policymakers and administrators who try totake advantage of this window of opportunity don't harm reformsthat have succeeded in improving learning outcomes and don'timplement reforms in a manner that would fail to do the same. Advocatesfor extended school years have so far said virtually nothing aboutwhether or how accountability systems should accommodate longer schoolyears. Across the country, a small number of schools and districts aremodifying or extending the academic year. The Massachusetts 2020initiative has provided resources for several dozen schools to increasethe number of instructional days they offer from 180 to about 200. Otherexamples include low-performing schools that have lengthened theirschool day in an effort to improve, and the longer school days, weeks,and years in some charter schools. However, such initiatives remainrare, with no systemic change in the instructional time provided toAmerican students. Our work confirms that increasing instructional timecould have large positive effects on learning gains. Encouraging schoolsand districts to view the school calendar as a tool in the effort toimprove learning outcomes should be encouraged in both word and policy. RELATED ARTICLE: Research Design Our studies use variation from one year to the next in snow or thenumber of instructional days cancelled due to bad weather to explainchanges in each school's test scores over time. We also take intoaccount changing characteristics of schools and students, as well astrends in performance over time. The advantage of this approach is thatweather is obviously outside the control of school districts and therebyprovides a source of variation in instructional time that should beotherwise unrelated to school performance. Furthermore, Maryland andColorado are ideal states in which to study weather-relatedcancellations. In addition to having large year-to-year fluctuations insnowfall, annual snowfall in both states typically varies widely acrossin Maryland and Colorado, some districts are exposed to much greatervariation in the severity of their winters than others, which allows usto use the remaining districts to control for common trends shared byall districts in the state. Further, because we have data from manyyears, we can compare students in years with many weather-relatedcancellations to students in the same school in previous or subsequentyears with fewer cancellations. Although cancellations are eventuallymade up, tests are administered in the spring in both states. This ismonths before the makeup days held prior to summer break. In Marcotte (2007) and Hansen (2008), we estimate that eachadditional inch of snow in a winter reduced the percentage of 3rd-,5th-, and 8th-grade students who passed math assessments by betweenone-half and seven-tenths of a percentage point, or just under 0.0025standard deviations. To put that seemingly small impact in context,Marcotte reports that in winters with average levels of snowfall (about17 inches) the share of students testing proficient is about 1 to 2percentage points lower than in winters with little to no snow. Hansenreports comparable impacts from additional days with more than fourinches of snow on 8th-grade students' performance on math tests inColorado. Marcotte and Steven Hemelt (2008) collected data on school closuresfrom all but one school district in Maryland to estimate the impact onachievement. The percentage of students passing math assessments fell byabout one-third to one-half a percentage point for each day school wasclosed, with the effect largest for students in lower grades. Hansen(2008) found effects in Maryland that are nearly identical to thosereported by Marcotte and Hemelt, and larger, though statisticallyinsignificant, results in Colorado. Hansen also took advantage of adifferent source of variation in instructional time in Minnesota.Utilizing the fact that the Minnesota Department of Education moved thedate for its assessments each year for six years, Hansen estimated thatthe percentage of 3rd- and 5th-grade students with proficient scores onthe math assessment increased by one-third to one-half of a percentagepoint for each additional day of schooling. Dave E. Marcotte is professor of public policy at the University ofMaryland, Baltimore County. Benjamin Hansen is a research associate atIMPAQ International, LLC.