Published Online: August 2, 2000
Published in Print: August 2, 2000, as Testing

Departments

Testing

Article Tools
  • PrintPrinter-Friendly
  • EmailEmail Article
  • ReprintReprints

Dropouts

Critics have long worried that requiring students to pass a test to graduate from high school could increase the dropout rate. Now, a report from researchers at Boston College suggests that's true.

The study by Marguerite Clarke, Walter Haney, and George Madaus of the National Board on Educational Testing and Public Policy, an independent board set up to monitor assessment in America, draws on evidence from several sources, including state-level data and a national longitudinal survey of 8th graders.

While much of the research relies on correlations and cannot prove cause and effect, Mr. Haney said, "there's a fair amount of evidence that higher rates of dropping out are associated with high-stakes high school graduation tests."

One study found that the 10 states with the lowest dropout rates in 1986 had neither minimum-competency tests nor tests used to determine whether students graduated from high school. Of the 10 states with the highest dropout rates, nine used tests to help make graduation decisions, and four used them to make promotion decisions.

Another study found that dropout rates between grades 8 and 10 tended to be higher—by 4 percentage points to 6 percentage points—in schools serving low-income students that used minimum-competency tests than in similar schools that did not rely on such exams.

In a third piece of evidence, Florida high school students whose grade point averages were between 1.5 and 2.5 on a 4-point scale were more likely to drop out of school if they failed the state's high school exit exam than similar students who had not failed the test. For students with lower grades, there was no relationship between failing the test and dropping out.

Read the report, "High Stakes Testing and High School Completion" online (requires Adobe's Acrobat Reader).


A Layman's Guide

Ever wonder what the difference is between a "mean" and a "median" or why statisticians use terms like "no measure of central tendency without a measure of dispersion"?

A new primer on testing, by independent researcher Gerald W. Bracey, is now available from the American Youth Policy Forum, in cooperation with the National Conference of State Legislatures. The forum published the guide "in response to confusion about—and frequent misuse of—the vocabulary of testing" by both policymakers and the media.

Read "Thinking About Tests and Testing: A Short Primer in 'Assessment Literacy,'" online.

—Lynn Olson

Vol. 19, Issue 43, Page 16

Back to Top Back to Top

Most Popular Stories

Viewed

Emailed

Recommended

Commented