Credit: Alison Yin for EdSource TodayCalifornia officials have identified many schools that will have difficulty offering online statewide tests scheduled in the spring unless their Internet capacity is improved.
In response to surveys asking schools to identify their Internet connection problems during the Smarter Balanced field tests earlier this year, a group of technology experts winnowed down the initial list of schools eligible for Internet improvement upgrades from more than 600 to 304.
California lawmakers set aside $26.7 million in June to help school sites shore up their Internet connections so their students would be able to take the Smarter Balanced tests, which will measure their critical thinking and other skills aligned with the Common Core State Standards.
But government employees contracted to survey the state’s schools to determine whether they may be eligible for broadband upgrades say the Broadband Infrastructure Improvement Grant fund will fall far short of the need.
“We could probably spend all of the $27 million on eight or nine counties in the northern part of the state,” said Teri Sanders, the senior director of education technology at the Imperial County Office of Education/K-12 High Speed Network, which works with the majority of the state’s schools to help improve their routing and connection to the Internet.
Those areas alone would need the entire state grant fund just to make it possible to do the computerized testing with any degree of efficiency, Sanders said. “These sites that have zero to very low connectivity are in areas that are so sparsely populated” that commercial Internet providers “have never invested in building to the areas,” Sanders added.
The California Department of Education said last month that most California schools were able to give the field tests last spring without Internet problems. But it is attempting to shore up Internet access, especially for small schools in more remote areas.
The Northern California counties Sanders referred to, some of which include Humboldt, Sonoma, Shasta, Siskiyou, Tehama and Trinity, account for 67 of the 304 school sites in the state that are eligible for broadband upgrades, according to data collected by the high-speed network group. In total, the schools represent about 1 percent of California’s 6.2 million public school students and 3 percent of the state’s 10,366 schools, including schools in rural and urban areas.
Credit: John C. Osborn/EdSource Today
The cost of getting schools in gear for the Smarter Balanced test varies dramatically, depending on how remote an area is and whether there are enough potential customers to lure Internet and cable TV providers there. If an area is remote, the cost of building infrastructure, such as laying down miles of fiber, could be prohibitive for a school site, said Edward Avelar, an outreach specialist for technological services for the high-speed network group.
Smarter Balanced recommended that each student taking the online test would need access to a basic connection of 20 kilobits per second. But Avelar said that estimate did not include other traffic on a school site’s bandwidth, such as other online instruction or online work by the school’s administration.
Even schools in the county that houses technology giants like Apple and Google have unresolved Internet problems. Two schools in Santa Clara County – Loma Prieta Elementary and CT English – are finalists for the Broadband Infrastructure upgrades.
The schools, which are a stone’s throw apart in a rural area just outside of Los Gatos, are a mere 10 feet away from underground fiber cable that would give the district the bandwidth it needs for the online tests, according to Corey Kidwell, superintendent of the Loma Prieta Joint Union School District. Other fiber cable sits about a mile away.
To have reliable Internet access for her test-takers at the two schools, Kidwell says the district needs a 1 gigabit connection, which is far faster than the 50 megabit bandwidth it used for online testing last year. Comcast, the schools’ provider, has offered to increase the district’s Internet capacity to about one-third of the 1 gigabit speed for $3,500 a month. Another company, Sunesys, offered to connect the schools to the underground cable a mile away for $1 million, with an ongoing monthly maintenance fee of $3,500.
“Our total annual budget is $4 million; I couldn’t do that,” Kidwell said. “When Smarter Balanced came, it sucked all the bandwidth out of the school and the shared resources in the community. I am sure it slowed everyone around us down to a crawl,” she said, adding, “We’re on the edge of Silicon Valley. This is silly.”
If the district doesn’t get the broadband grant, Kidwell said, it will do whatever is needed to make sure students take the test, but it will be at a cost. “Last year we basically had to disrupt instruction for at least three weeks,” she added, “because nobody could use computers while the 365 students took the test.”
“I am sure it slowed everyone around us down to a crawl,” said Loma Prieta Joint Union School District Superintendent Corey Kidwell, referring to poor Internet service during online student testing. “We’re on the edge of Silicon Valley. This is silly.”
Internet connection problems also affect schools in urban areas. Thirty-seven schools in the Los Angeles Unified School District are among the finalists for the state grants. Typical issues during the Smarter Balanced test last year included being bumped off the test site, screen freezes and problems logging in, according to district data.
The district applied for the state grant to help offset the construction costs of improving their fiber cable connections at those 37 schools, according to Samuel Gilstrap, a district spokesman. If the district does not get the service upgrade, it will use a device that controls traffic on its bandwidth for the spring test, until it secures more funding, Gilstrap said.
Backup plans for other districts include spreading their testing of students over time. Other schools may have to test fewer students at the same time or transport them elsewhere to take the test.
The Smarter Balanced field test last spring revealed a host of problems for other school sites that made the list for broadband upgrades as a result of the severity of their online problems. By far the biggest problem was that students could not take the test at the same time. One-hundred and sixty-nine schools had to “limit simultaneous test takers,” while 52 schools had to shut down online activity and 24 schools had to transport their students to another site, according to the high-speed network group.
Mountainous Trinity County is considering using microwave devices on towers to transmit computer data to schools and allow about 150 students in 3rd through 8th and 11th grades to take the required test in the spring, said Robert Jackson, the Trinity County Office of Education’s technical director.
But if the state selects Trinity’s schools for service upgrades, engineers tasked with deploying the upgrade may find other methods, Jackson added.
To be a finalist for an upgrade, schools must show that they do not have Internet connections or have connectivity problems that will interfere with their ability to offer the test online.
In rural Tulare County, “connecting to the high-speed fiber network was not an option, because the cost would have been in the hundreds of thousands,” said Daniel Huecker, director and superintendent of the Eleanor Roosevelt Community Learning Center, a K-12 charter school. If his school is not selected for an upgrade, his backup plan for the 162 students required to take the spring test is to schedule some students earlier in the school year and others later, in order to work with the wireless connection the school uses.
In the coming months, industry experts working with the California Department of Education and statewide organizations will review the school sites on the list to rank their need, department spokeswoman Tina Jung said.
“If there are insufficient funds available to award all applications that receive a qualifying score,” Jung said, “a priority ranking will be applied until all funds are exhausted.” The department has asked for a Dec. 18 deadline for deciding which schools will get service upgrades through the state Internet grant program.
The timeline raises concerns about whether the school sites chosen for upgrades will be able to get them in time for the spring test. “Once you put an order in to a provider, it takes 90 days from that date for service,” said Sanders of the high-speed network group.
If students are unable to take the test online for any reason, they’ll have the option of taking a hard copy version, Jung said. The high-speed network group will also work to secure discounted rates for schools not awarded service upgrades, according to Avelar.
At the same time, other administrators are weighing their options if their schools aren’t selected for an upgrade.
The Trinity County Office of Education will have to resort to measures it used last year during the field test if its schools are not awarded a service upgrade, said Sarah Supahan, assistant superintendent of curriculum and instruction.
It will schedule fewer students taking the test at one time, she said, “which in one school meant one student at a time – or transport students an hour or more away to another district to take the test, possibly on multiple days.”
“Until other funds can be found and improvements can be made,” Supahan added, “I don’t see any other option.”
To get more reports like this one, click here to sign up for EdSource’s no-cost daily email on latest developments in education.
Comments (15)
Comments Policy
We welcome your comments. All comments are moderated for civility, relevance and other considerations. Click here for EdSource's Comments Policy.
jskdn 9 years ago9 years ago
Why not just locate testing servers on these campuses without sufficient Internet bandwidth? Couldn’t that create the same testing environment as campuses connecting to remote testing servers by way of the Internet?
Replies
el 9 years ago9 years ago
Interesting idea, truly. I don't know enough about the system to know if it will work if not attached to the mother database as is, but there's no reason you couldn't design a standalone server that could travel. Real bandwidth would be more beneficial in the long run, but the standalone server seems like a better solution, and probably no more costly, than a set of paper and pencil tests. You could pair it with 100 … Read More
Interesting idea, truly. I don’t know enough about the system to know if it will work if not attached to the mother database as is, but there’s no reason you couldn’t design a standalone server that could travel.
Real bandwidth would be more beneficial in the long run, but the standalone server seems like a better solution, and probably no more costly, than a set of paper and pencil tests. You could pair it with 100 laptops so that schools could test more kids at once and get the server and tech on to his next destination more quickly.
navigio 9 years ago9 years ago
Depends how they implement authentication and how they retrieve and populate results.
Tressy Capps 9 years ago9 years ago
Bill Gates to the rescue! Hey, here’s an idea….scrap common core.
Replies
FloydThursby1941 9 years ago9 years ago
Would you be willing to be a part of the committees that spend thousands upon thousands of hours to put in place the new national system to test children in 50 states and get us out of the bottom of the pack in terms of international education and test scores? If we scrap it, it will mean an amazing amount of wasted time and difficulty in starting over. If we do follow this … Read More
Would you be willing to be a part of the committees that spend thousands upon thousands of hours to put in place the new national system to test children in 50 states and get us out of the bottom of the pack in terms of international education and test scores? If we scrap it, it will mean an amazing amount of wasted time and difficulty in starting over. If we do follow this unprecedented step, we really need to make sure we as a nation are unified behind whatever we replace it with. If we come up with something else and everyone is on blogs wanting to scrap it, it would be better to stick with this whatever the imperfections. If you and all the critics would agree to be a part of fixing the problems and agree that they may be outvoted on certain technicalities, it might be worth it. What we have to avoid at all costs is a situation where we replace this and don’t end up with anything better or anything we can work on as a nation together, as the highest priority.
Zeev Wurman 9 years ago9 years ago
One wonders if that will also be the excuse not to scrap the high speed rail in a few years, after we will sink a few billions into this and realize that it will become a money sinkhole forever.
Something about not throwing good money after bad comes to mind.
el 9 years ago9 years ago
Yes, how terrible it will be when you will be able get to so many of California’s public universities via public transportation as a day trip, from nearly anywhere in the state.
FloydThursby1941 9 years ago9 years ago
Zeev, you avoided the question. Everyone is saying scrap a huge undertaking because there are a couple flaws, but they look for flaws and secretly oppose reform at all. That's why I call their bluff. If the flaws are the issue, why not get all the critics to work on the committees to develop a new system, with the agreement that once it goes through, they all try to make it work? … Read More
Zeev, you avoided the question. Everyone is saying scrap a huge undertaking because there are a couple flaws, but they look for flaws and secretly oppose reform at all. That’s why I call their bluff. If the flaws are the issue, why not get all the critics to work on the committees to develop a new system, with the agreement that once it goes through, they all try to make it work? Because they were opposed no matter what and are looking for mistakes to be negative.
Be constructive. If you don’t like a small thing, work to fix it. We were not moving well before this. In fact, in my view, NCLB is the only thing W did right. We need Common Core. I wish it were not so computerized, but I will do my best to be a part of the solution. Going to meetings and saying we should change this or that is constructive, but just randomly suggesting we scrap it every time a mistake is make just makes the people trying to make this work dismiss you as a nut and not listen to what you have to say. If you’re not part of the solution, then you are part of the problem. If you are part of the solution, propose fixes, but not wholesale elimination which would be disastrous. Do your best to find constructive solutions.
el 9 years ago9 years ago
Thank you for this article, reaching into the details of bandwidth and infrastructure throughout our state. Many people will read this and think this is a waste, to bring this connectivity 'just for the test.' In real life, this is about rural broadband in general, not just for one test a year, but to bring these kids access to great internet resources year round and to bring data services to their communities - data services that … Read More
Thank you for this article, reaching into the details of bandwidth and infrastructure throughout our state.
Many people will read this and think this is a waste, to bring this connectivity ‘just for the test.’ In real life, this is about rural broadband in general, not just for one test a year, but to bring these kids access to great internet resources year round and to bring data services to their communities – data services that are essential for improving not only lifelong learning for citizens of all ages, but also for business development and improved access to medical care and many other valuable services that benefit these communities.
It’s a national problem and it’s as important and valuable as universal phone service.
John mockler 9 years ago9 years ago
Well now we start looking at the costly logistical details of on line assessment. Andremember we still have not done a full real test.
Replies
Doug McRae 9 years ago9 years ago
Yup, John, the costs for hardware and bandwidth and professional development to use that hardware and bandwidth for on-line testing are very significant. But, those costs will indeed develop capacity for not only on-line testing but also on-line instruction, and in the long run will be justified simply to get into the 21st century for our K-12 education system. More nuanced than pure dollars are issues concerning equity for installing on-line testing for low-SES … Read More
Yup, John, the costs for hardware and bandwidth and professional development to use that hardware and bandwidth for on-line testing are very significant. But, those costs will indeed develop capacity for not only on-line testing but also on-line instruction, and in the long run will be justified simply to get into the 21st century for our K-12 education system. More nuanced than pure dollars are issues concerning equity for installing on-line testing for low-SES student populations, for ELs, and for students with disabilities, all issues that should be considered and largely addressed before moving the lever forward for 100 percent on-line test administrations. The post also mentions the option to do (at least partial) Smarter Balanced tests in a paper/pencil mode; however, to do that, we have to have comparability studies in hand to insure the results are comparable for both on-line and p/p test administration modes. Those studies have yet to be conducted, as far as I know. There was discussion last May that Smarter Balanced would conduct a small computer-adaptive vs paper/pencil study (I’m not sure it was sufficiently robust to yield dependable comparability adjustments) in June, but I haven’t heard whether or not that study was completed by Smarter Balanced . . . . . or if it was completed, what the results were . . . . .
Zeev Wurman 9 years ago9 years ago
In addition to bandwidth and its associated hardware (routers, local servers) one should also factor in: - costs of extra technical support during the testing window. A school needs more than one technical person in case of problems, otherwise a single service issue can cause a dozen of minor issues to sit un-addressed for long periods, ruining the test for large number of kids. - cost of installing large number of electrical outlets for large number of … Read More
In addition to bandwidth and its associated hardware (routers, local servers) one should also factor in:
– costs of extra technical support during the testing window. A school needs more than one technical person in case of problems, otherwise a single service issue can cause a dozen of minor issues to sit un-addressed for long periods, ruining the test for large number of kids.
– cost of installing large number of electrical outlets for large number of computers, and additional AC where needed.
– making sure that testing computers are in an almost perfect working condition — no sticky keys on the keyboard, no “lazy” mouse, no broken sound card, etc. Otherwise expect complaints and even lawsuits.
The last point in particular points to the fact that over time schools will probably move to dedicated machines for testing only, much like voting machines that are kept aside to be in a good working order when the time comes. Which makes me disagree with Doug (sorry, Doug! 🙂 that much of this hardware will serve dual-duty and enable a lot of on-line instruction.
Doug McRae 9 years ago9 years ago
Ze'ev -- We're probably not that much in disagreement. I'd certainly agree that standby hardware devices and contingency plans for bandwidth are vital for any large scale on-line test administration, due to the costs / consequences for hardware or bandwidth failure during testing [i.e., total loss of any potential data, comparability of data, etc]. And I'd point to the need for sufficient devices and bandwidth to support a reasonable 4-5 week test administration window … Read More
Ze’ev — We’re probably not that much in disagreement. I’d certainly agree that standby hardware devices and contingency plans for bandwidth are vital for any large scale on-line test administration, due to the costs / consequences for hardware or bandwidth failure during testing [i.e., total loss of any potential data, comparability of data, etc]. And I’d point to the need for sufficient devices and bandwidth to support a reasonable 4-5 week test administration window for a school [hopefully not more than one week for each grade level within a school], compared to the 12-week window now in regulations that is poor large scale testing practice in terms of test security issues as well as comparability of results. The current 12-week window has been justified primarily as a concession to current technology shortages, but it is a concession that degrades the quality of the test results that are obtained. It shows how right now technology issues are trumping good educational measurement practices in California, and that is a trade-off we should not have to make.
el 9 years ago9 years ago
Something we were talking about at our school in trying to determine what hardware to purchase is the different experience on different sized screens. Someone who had observed kids taking the tests saw that the kids had the computers in split screen mode often and that the amount of scrolling they had to do or didn't, she postulated, would increase the difficulty and the fatigue in answering the question. So this is another potential factor in … Read More
Something we were talking about at our school in trying to determine what hardware to purchase is the different experience on different sized screens. Someone who had observed kids taking the tests saw that the kids had the computers in split screen mode often and that the amount of scrolling they had to do or didn’t, she postulated, would increase the difficulty and the fatigue in answering the question.
So this is another potential factor in equity, when comparing test scores. Will the kids who have 20″ screens have a different experience than kids using small chromebooks? Or iPads?
John mockler 9 years ago9 years ago
The amount of interesting response indicates a good initial report. EdSource should dig even deeper on this issue. John