Giuliani's Spin City

May 15, 2007— -- As he runs for the Republican nomination for president based on his record as mayor of New York City, Rudolph Giuliani loves to go by the numbers.

To hear Giuliani tell it, the number of adoptions in New York jumped nearly 70 percent during his time as mayor -- the result, he claims, of new policies he enacted to reduce the number of abortions.

Calling himself the most fiscally conservative candidate in the race, he has bragged repeatedly about his management of the city budget, saying he turned a $2.3 billion deficit into a "multi-billion-dollar surplus," all while cutting taxes by $9 billion.

And he cut crime so much that New York went from being "the crime capital of America" to "the safest large city in America" by the time he left office in 2002, he said on "Fox News Sunday."

But a close review of Giuliani's statistical claims reveals that spin is interfering with some of his substance.

Giuliani's claim on the number of adoptions, for instance, is based on a highly selective reading of the data, according to the independent watchdog group Annenberg Political Fact Check, which runs the Web site Factcheck.org.

Read another way, the same figures could even suggest that Giuliani actually presided over a decline in the number of annual adoptions in New York during his second term as mayor, the group concluded in a report released last week.

"This is a classic case of using statistical data in a misleading way," Brooks Jackson, the organization's director, said in an interview. "He's used the data selectively to create an exaggerated impression."

The Giuliani campaign cites city statistics that show the total number of adoptions in New York was 66.5 percent higher in the last six fiscal years he was mayor -- 1996 through 2002 -- than in the preceding six years, from 1990 through 1995.

That comparison is valid, Giuliani aides argue, because he created a new office to protect children and encourage adoption in 1996, meaning the former mayor can claim credit for the uptick.