White House Plan For Head Start Test Draws Critics
Identifying letters, holding a book the right way, and recognizing that a word is a unit of print are some of the skills that all 4-year-olds in Head Start could be assessed on beginning next fall, under a new program planned by the Bush administration that some early-childhood educators contend will not work.
Officials of the Administration for Children and Families, which is part of the Department of Health and Human Services, were expected to elaborate on their plan this week in Washington at a meeting of Head Start directors from around the country.
But the idea of giving preschoolers any kind of standardized test has some experts worrying that federal officials are going too far in their drive to hold local programs accountable for children's performance.
"It will not work," said Samuel J. Meisels, an expert on assessment of young children and the president of the Erikson Institute, a graduate school for child development in Chicago. "It won't be meaningful data."
According to President Bush's early-childhood-education initiative, called Good Start, Grow Smart, data from the new "national reporting system" will be used to determine whether local Head Start programs are successfully teaching the "standards of learning." Those results "will be used in HHS evaluations of local Head Start agency contracts," the document says.
In Congress, Sen. Edward M. Kennedy of Massachusetts—the ranking Democrat on the Senate Health, Education, Labor, and Pensions Committee—has expressed concerns over the use of Head Start assessments.
"He is opposed to anything that approaches high stakes," said his spokesman, Jim Manley.
Mr. Meisels and most early-childhood experts argue that preschool assessments are often unreliable because children's development is still so fluid at such a young age.
"You have to be really careful about how you apply standardized testing to young children," said Jerlean E. Daniel, a psychology professor at the University of Pittsburgh and an expert on early literacy. "The results you get are so dependent on how the child is feeling that day. And will he get a second chance?"
Ms. Daniel recently spent a year working on literacy initiatives at the Head Start Bureau in the HHS Department, which administers the federal preschool program for disadvantaged youngsters. While she was not aware of plans to institute a new assessment, implementing one would certainly be in keeping with the Bush administration's "outcomes oriented" approach, she said.
But Windy M. Hill, the associate commissioner of the Head Start Bureau, said last week that the new reporting system was simply an effort to evaluate a relatively new focus for Head Start that already was passed as part of the program's 1998 reauthorization: language development and literacy. She added that the results of the assessment would be just one piece of information used to determine whether local programs are meeting expectations.
"I regret that folks, especially those in Head Start, believe that one measure could be used to determine the fate [of a program]," said Ms. Hill, who oversaw several Head Start centers in Texas before being appointed to lead the bureau last year. "The intent of a national reporting system is not to create some kind of pass-or-fail test for grantees or children."
The reason federal officials want to assess every child instead of just taking a random sample, as is done currently, is that they "want to ensure that every child is receiving the same quality of services," Ms. Hill said.
While Ms. Hill said she didn't know if the results would be shared with parents, she said that they would be used to help teachers and program leaders better monitor their efforts and make corrections.
Head Start serves more than 900,000 children and has a budget of $6.5 billion, which pays for nearly 19,000 programs.
As it is, children in Head Start are already assessed at least three times a year to determine whether they are progressing, and the new assessment will likely be integrated into those assessment periods, Ms. Hill said.
Experts say that assessments should be done in ways that make young children feel comfortable, such as through observation or in the context of a conversation. They also say that while assessments of young children can be used appropriately to improve curricula and instruction, they should not be used to hold programs or individuals accountable.
'It Is High Stakes'
But Mr. Meisels of the Erikson Institute, who is on the technical-review panel for the new reporting system, said the assessment "looks like high stakes. It sounds like high stakes. It is high stakes."
He added that "teaching to the test" and even changing children's answers could become a problem if Head Start teachers believe their jobs depend on how children respond.
"We know that this is the case," he said.
While Head Start, which is due for another reauthorization by Congress this year, serves both 3- and 4-year-olds, many children don't enter until they are 4.
"Those children are going to walk in and within days are going to be assessed," Mr. Meisels said. "I think it's going to be relatively difficult for children to achieve well on this."
He added that measuring what children know at age 4 could also affect 3-year-olds in the program. "Everyone will say, 'Get those 3-year-olds ready for the test,'" he said.
The assessment, he argued, is also too narrow and does not consider children's social and emotional development. "There is this gigantic world of preschool operation that is not touched by this," he said.
Ms. Hill disagreed with those criticisms. She said that local programs would continue to be evaluated on all of the standards they are expected to meet, not just literacy.
Some observers also want to know who will be administering the new assessment and how it will be financed.
"No one wants programs that aren't doing the best job possible," said Helen Blank, the director of the child-care and -development division at the Children's Defense Fund, an advocacy organization based in Washington. "But there is the question of whether this is the best way to do it."
Vol. 22, Issue 18, Pages 20,22