Schools turn to online testing for Standards of Learning

Schools turn to online testing for Standards of Learning

EVA RUSSO/TIMES-DISPATCH

Frankie Wilson, a third-grader at Harrison Elementary School, in Prince George County, used a computer to take a benchmark test last week.

» 6 Comments | Post a Comment

Melissa Mitchem, a second-year teacher at Falling Creek Elementary School, has been eager for her fifth-graders to take their Standards of Learning tests online.

This spring, she'll get her wish.

Falling Creek will start online testing as the Chesterfield County school system continues its implementation of Web-based testing at the elementary school level.

Last year, five of the county's 38 elementary schools administered some testing online. This year, 29 elementary schools will administer at least one subject test online. All high and middle school SOL testing will be conducted online. Testing begins Wednesday and runs through August.

Chesterfield is ahead of Richmond and the counties of Hanover and Henrico -- and ahead of the state -- in moving from pencil-and-paper testing to online assessments.

And Virginia is leading the way in online testing. The Department of Education estimates that 1.7 million online tests will be administered this spring, Shelley Loving-Ryder, assistant superintendent for the Division of Student Assessment and School Improvement, said in a recent report to the Virginia Board of Education. That total is up from 1.4 million last spring.

"That's an extraordinary number," she said. "Virginia has become a national leader in online testing."

. . .

All SOL tests are available online, except the English writing tests. The state Department of Education is taking steps to move them to the Web.

Some school systems are limiting online testing to SOLs for state accreditation. Others also are administering student-progress assessments at all levels in the core subjects of science, math and history.

When the state Department of Education launched the Web-based-testing initiative in 2000, 15 school systems participated. By 2005, all school systems were administering tests online.

Every year since the department began the initiative, it has provided $50,000 per school system, plus $26,000 for each school, to improve technology infrastructure.

The online testing program began in response to the need to get results back quickly for end-of-course tests for high schools, Loving-Ryder said. About 98 percent of those assessments are administered online, she said.

"We love the online testing because we get the SOLs results back so quickly," said Eric Jones, director of high school education for Henrico's schools. "It allows us to re-teach the material to students so they get another opportunity to test before they leave for the summer."

Because so few end-of-course tests in high schools still are administered in the paper-and-pencil format, the Department of Education is considering a new policy in which traditional versions of tests would be available only as an accommodation, Loving-Ryder said.

"As we move toward more and more online testing, and that is the expectation, this will become an issue in the lower grades as well," she said.

Statewide, about 61 percent of middle school tests are conducted online, as are 12 percent at elementary schools, Loving-Ryder said. This spring, those percentages will increase up to 7 percent at the elementary school level and up to 21 percent at the middle school level, she said.

. . .

Online testing doesn't come without its technical problems.

Two years ago, major glitches kept about 9,000 Virginia students from finishing their online assessments in separate testing days because the company administering the tests, Pearson Educational Measurement, experienced system shutdowns.

Last year, tests were administered without major problems, said Charles Pyle, spokesman for the Virginia Department of Education.

Chesterfield's schools have had a case or two of individual students being bumped from the system midtest and having to restart, Technology Director Lynda V. Gillespie said.

In 2005 in Henrico, about 200 students taking SOL tests were affected when the word "pimp" was created as a randomly generated security code and the testing software froze.

"We haven't had any technical issues since then," Jones said.

Other challenges for school systems include not having enough computers and having to interrupt instruction during testing.

Renee Williams, assistant superintendent for instruction for the Prince George County school system, said schools must use all computers for testing.

"Since classes are taught in some of the computer labs, depending on the testing schedule, these classes must be held in another classroom," she said.

Carole O'Brien, director of guidance, testing and research for Hanover's schools, said online testing involves a lot of planning.

Schools need back-up plans for system failures, and they also need to be able to handle students who require more time to take a test or need to make up a test; those who finish early; and those who are absent.

"If the plan is in place months prior to testing," O'Brien said, "the benefits to students far outweigh the challenge of time and effort it takes to make the plan."



Contact Juan Antonio Lizama at (804) 649-6513 or .


How they test

Here is a sample of where central Virginia schools stand in administering Standards of Learning tests online:
AMELIA COUNTY
All testing, except fourth-grade math, is online.
CAROLINE
Elementary schools: Fourthand fifth-grade math testing is online.
Middle and high schools: All testing is online.
CHESTERFIELD COUNTY
Elementary schools: This year, 29 of 38 elementary schools plan to test at least one subject online. Last spring, five schools administered online tests in social studies, history and math.
Middle and high schools: All testing is online.
HANOVER COUNTY
Elementary schools: None.
Middle schools: Will conduct end-of-course math tests online, plus a selection of other multiple-choice tests.
High schools: All end-of-course testing is online.
HENRICO COUNTY
Elementary schools: None.
Middle and high schools: All testing is online.
HOPEWELL
Elementary schools: None.
Middle school: Eighth-grade math and science testing is online.
High school: All testing is online.
NEW KENT COUNTY
Elementary schools: Fifth-grade social studies testing is online; may add another subject this year.
Middle and high schools: All testing is online.
RICHMOND
Elementary schools: None.
Middle schools: Will administer end-of-course tests online, excluding sixth-, seventhand eighth-grade tests.
High schools: End-of-course testing is online.
PRINCE GEORGE COUNTY
Elementary schools: Four schools will administer science and history test online, and the fifth school also will test online in math.
Secondary schools: All testing is online.
SOURCE: Local school districts

Advertisement

 
View More: tricities,petersburg,hopewell,henrico county,hanover county,city of richmond,chesterfield county,
Not what you're looking for? Try our quick search:
 

Advertisement

Reader Reactions

Flag Comment Posted by USN_HOO on April 13, 2009 at 10:08 pm

Concerned - The best way I know of to explain what “formative” use of tests and “transparency” are is to contrast them to their opposites: “summative”, and “obtuse”, respectively. If we are focused on improving education should we use testing as a summative tool to evaluate teacher performance only? or also use each test as a formative aid in the classroom and home environment to review each and every mistake made to correct any misunderstandings. The New York tests are transparent; their state officials release all questions to the public after tests are graded, usually within two weeks. Virginia is justifying obtuse and opaque behavior by citing cost of new test questions. Our children deserve the very best we can afford to give them. Transparency in testing should be the standard for schools no matter what the purpose of testing is purported to be; education is a formative process and should not be used to “sum up” learning and give grades without hope of improvement. I hope my expanation is helpful to you.

Flag Comment Posted by MeToo on April 13, 2009 at 7:44 pm

USN-HOO:

Are you suggesting that they get to preview the current test or review more past tests?  Many of the past tests are available on the VDOE website.  Some really can’t be released for review- History or Science for example.  You can come up with millions of math questions and reading passages, but how many ways can you ask Who shot Lincoln?  When it came to test time, you’d never be able to definitively say whether or not the student’s knew the answer or simply memorized the correct answer from past tests.

Same concept with SAT and other standardized tests- you can get past examples, but you aren’t always going to get every detail of every one.

Maybe I misinterpreted your comment all together, but that’s what I took from it.

Flag Comment Posted by MeToo on April 13, 2009 at 7:38 pm

RJT- you grips (valid) should be addressed to the VDOE, not school districts.  Districts have ZERO do to with content and scoring… that’s a state/ test producer problem.  As for the computer testing issues, I agree and think that districts should make sure they have fool proof internet and a good backup plan.

Concerned- you are right in that the “benchmark” assessments are entirely for the purpose of predicting how students will do on the final SOL, aka are the teachers teaching?  However it’s basically like 4 mini mid-terms.  It’s really not weeks they are missing to take this test- about a half a day for each (4) 3 times per year.  That’s about 6 days.  I don’t know about your child(ren)‘s specific teacher, but he/she may be taking more time to do prep (which they really aren’t supposed to do!).  Until high school, SOLs count for nothing more than an indication of whether or not the teacher taught and the students learned something.  In other words, they have no real meaning… daily grades should be considered as much more significant in most cases.

In a way SOLs are good- they should by purpose show what students learned and that education is consistent across the state.  The practice doesn’t quite come out that way.  It becomes teaching to the test, unfortunately.  You can think No Child Left Behind for that.  As long as NCLB remains as written, teaching to the test will remain as practice.  It doesn’t really matter what the teacher’s tell you the purpose of SOLs are… it’s all a failing measure to give and take money.  In the end, it measures dollars, not schools, teachers, or students.  Furthermore, the sanctions for failing to “meet the grade” only exasperate the problem and offer very little, if any, opportunity for intensive help and restructuring for the district.

I don’t think they use the benchmark data to “dumb down” the test, but rather prepare themselves to know who’s probably going to fail and put extra pressure on them to do whatever they can to pass.  Afterall, it’s about image and reputation- can’t let the good ol’ Commonwealth down!

Flag Comment Posted by RJT on April 13, 2009 at 6:26 pm

The school districts and department of education continue to gloss over problems with online testing.  Eric Jones forgets the issues with the Biology SOLs in 2006 where students got knocked out and proctors and administrators could not access the Pearson site to restart the tests or end the tests for students who’d completed them.  Nor do they address the repeated issues with mis-scoring of tests.  Educational Testing, which administers the SATs has gone so far as to offer “scoring insurance” where for $100 extra you can have your SAT graded by hand and insure no computer scoring mistakes.  Online SOL testing is a good idea, but has significant problems with how it’s currently being done.

Flag Comment Posted by concerned on April 13, 2009 at 4:47 pm

USN_HOO,  Can you please explain what you mean by “transparency” and “formative” use of tests?

I have real concerns about the excessive use of benchmark SOL testing each quarter.  My children, in 3rd grade this year, have missed the equivalent of weeks of instruction due to repetitive SOL testing.  Schools typically report in the media that SOL testing provides valuable DATA that drives future instruction in order to better prepare students to take the actual SOL tests in the spring. But we do not see that happening in our school. In fact, my girls have failed most of these benchmark tests, and when we have asked for “reteaching” or remediation, we have been told there is no time for this in the schedule. Well, why do the benchmark testing?  This is a waste of time if it will not change instruction or direct remediation efforts!  This is just plain common sense. So what I have come to suspect, having a B.S. in Math with a concentration in statistics, is that the DATA is actually being sent to the VDOE in order to “re-norm” the tests (revise or rather “dumb down” the test questions) in order to ensure the required pass rate which will enable schools to make AYP. Anyone else have any opinions on this?

It is also very pathetic the way school staff deceive parents in meetings to believe that the SOL tests are testing their children.  Parents, the SOL tests are the TEACHERS’ TEST!  The teacher and the school are being tested. It’s called an accountability test to justify your tax dollars going into a big bureaucratic hole. I always told my kids to relax because they are not the ones being tested! This works wonders to prevent kids from worrying over SOL tests.

Flag Comment Posted by USN_HOO on April 13, 2009 at 12:26 pm

Reliable research has shown formative use of tests do significantly improve achievement, but transparency is most important. No matter how often a test is given or how quickly it is graded, the student is disadvantaged when his or her test questions are hidden from view. Virginia is not a leader in the transparency of SOL testing; in fact, we are one of the worst states in the nation. We release only a sample of a test and our children suffer, because their teachers, parents, and peers do not have access to the whole test. It is in fact a worst-case scenario when a child “fails” a test after “failing to pass” the test before, because his or her expectation will be to fail it again. Sampling is not good enough to provide formative use of their “test” and help them; they need to see whole tests and not be short-changed in the cost/benefit analysis. It’s a quality versus quantity issue and we’re not a leader in this area - sorry.

Post a Comment(Requires free registration)

  • Please avoid offensive, vulgar, or hateful language.
  • Respect others.
  • Use the "Flag Comment" link when necessary.
  • See the Terms and Conditions for details.
Click here to post a comment.

Advertisement

Advertisement

Online Features
Blogs
DataCenter
Videos
Weekend
 

Advertisement