The research presented by Diana Baumrind, who co-authored the study with Elizabeth Owens, both research psychologists at UC Berkeley's Institute of Human Development, calls into question a current claim that any physical punishment is harmful to a child.

The study separates out parents who use spanking frequently and severely - resulting in evidence of harm - and focuses on those families who occasionally spank their children, a practice that Baumrind calls normal for the population sampled.

By "spanking," Baumrind refers to striking the child on the buttocks, hands or legs with an open hand without inflicting physical injury and with the intention of modifying the child's behavior.

Baumrind's study also compares spanking with another kind of discipline, namely verbal punishment.

"We found no evidence for unique detrimental effects of normative physical punishment," Baumrind said in an invited address to the American Psychological Association annual meeting today in San Francisco.

"I am not an advocate of spanking," said Baumrind, "but a blanket injunction against its use is not warranted by the evidence. It is reliance on physical punishment, not whether or not it is used at all, that is associated with harm to the child."

She said that, in the absence of compelling evidence of harm, parental autonomy and family privacy should be protected.

Her study of spanking in middle-class, white families was undertaken in response to anti-spanking advocates who have claimed that physical punishment, by itself, has harmful psychological effects on children and hurts society as a whole.

These claims, Baumrind said, have not distinguished the effects of occasional mild-to-moderate spanking from more severe punishment, or taken into account such confounding factors as earlier child misbehavior and the effects of total child rearing patterns - from rejection, on one hand, to warmth and explanation, on the other.

The UC Berkeley study, however, was able to account for all of these factors and others, due to its unique data base. The data were drawn from longitudinal records of child rearing and child outcome in California East Bay families collected at the Institute of Human Development. Families in the Family Socialization and Developmental Competence Project (FSP) were first studied in 1968 when their children were preschoolers, and then in 1972-73 and 1978-80, when the children were early primary schoolers and early adolescents.

In addition to the rich archival material on parental styles and discipline, combined with independent observations and interviews with the children, Baumrind's team created a new instrument for the spanking study. Called the Parent Disciplinary Rating Scale, this instrument rated parents on their strategies for using discipline.

Few of the families, only 4 percent, never used physical punishment when their children were preschoolers, but there was a wide range in the frequency and severity of spanking throughout the whole sample, said Baumrind.

A small minority of parents, from 4 to 7 percent depending on the time period, used physical punishment often and with some intensity. Although these parents were not legally abusive, they were overly severe and used spanking impulsively. Hitting occurred frequently, but it was the intensity that really identified this group, said Baumrind.

She said intensity was rated high if the parent said he or she used a paddle or other instrument to strike the child, or hit on the face or torso, or lifted to throw or shake the child.

This group of parents, identified in the "red zone" for "stop" was removed from the sample at the first stage of analysis. With them went most of the correlations initially found between spanking and long-term harm to children, said Baumrind.

"When we removed this 'red zone' group of parents," said Baumrind, "we were left with very few small but significant correlations between normative physical punishment and later misbehavior among the children at age 8 to 9.

"Red zone parents are rejecting, exploitative and impulsive. They are parents who punish beyond the norm. You have very little to explain after you remove this small group."

She said the few links that remained were explained by the child's prior misbehavior. In other words, when researchers controlled for the tendency of the child to be uncooperative or defiant as preschoolers, all correlations between spanking and harmful effects were close to zero.

In addition to a "red zone," parents were classified into orange, yellow and green zones.

"There were no significant differences between children of parents who spanked seldom (green zone) and those who spanked moderately (yellow zone)," Baumrind said.

Families in the orange zone could have used spanking often, but with little or no intensity. Those in the "yellow zone" used physical punishment only occasionally, with little or no intensity, while those in the "green zone" used little or no physical punishment with no intensity.

The children of parents in the green zone who never spanked were not better adjusted than those, also in the green zone, who were spanked very seldomly, Baumrind said.

Studies of verbal punishment yielded similar results, in that researchers found correlations just as high, and sometimes higher, for total verbal punishment and harm to the child, as for total physical punishment and harm.

"What really matters," said Baumrind, "is the child rearing context. When parents are loving and firm and communicate well with the child (a pattern Baumrind calls authoritative) the children are exceptionally competent and well adjusted, whether or not their parents spanked them as preschoolers."