Programs that Work
Safe Child Program
Healthy and Safe Children
Children not experiencing physical, psychological or emotional abuse
Age of Child
Early Childhood (0-8)
Middle Childhood (9-12)
Type of Setting
Child Care / Preschool
Type of Service
Type of Outcome Addressed
Child Abuse and Neglect
Evidence Level (What does this mean?)
Developed in 1981 and updated in 1994, the Safe Child Program provides children ages three to ten with a set of skills to help them prevent sexual, emotional, and physical abuse. The main idea guiding the program is that there are times when children can and must be responsible for their own well being. At the same time, however, the program emphasizes that a child’s security can be enhanced without creating misunderstanding, fear, or anxiety or reducing a child’s sense of trust.
The program is designed to be used in a group classroom setting. By building on everyday experiences, the program teaches children how to speak up for themselves, how to assess and handle various situations, and to know where to get help. Children actively engage in discussions, games, and role-playing in ten separate sessions, giving them time to practice these skills.
The program has two main focuses. The first is prevention of sexual, emotional, or physical abuse by people known to the child. This is taught in the context of life skills, including communication, choices, asking for help, and decision-making. These skills are developed not as ideas, but as skills through role-play. The other focus of the program is safety around strangers. The classroom session discusses misconceptions the children may have, defines who a stranger is, and then provides four rules for dealing with strangers. These rules are these: keep an arm’s length away, don’t talk to strangers, don’t take anything from strangers, and don’t go anywhere with a stranger.
The program is designed for children in preschool through third grade. Children in the evaluations ranged in age from 3 to 10, attended a variety of rural, suburban, and urban schools, and came from diverse backgrounds.
The program has been evaluated in two separate studies. An evaluation in Denver (Fryer, Kraizer, and Miyoshi, 1987a) involved 48 children in kindergarten through second grade, half randomly assigned to the treatment group and half to a control group. The program was administered over eight consecutive classes in 20-minute sessions. All the children were assessed in two simulations—a pretest simulation and a posttest simulation that was given after the program had been administered to the treatment group. In the simulation, the children individually encountered a researcher who played the role of a stranger. The researcher would meet the child in the hallway and ask the child to help bring in birthday treats from his car. Children “failed” if they agreed to help the stranger. If they agreed to help, the researcher then told the child he would need his or her help later. Someone from the school later informed the child that the stranger no longer needed assistance. After the simulation, each child had a 30-minute meeting with a member of the research team to make sure they were not experiencing any fear or anxiety from the encounter.
The children were also assessed with three other standardized tests to complement the simulation. The Peabody Picture Vocabulary test measured cognitive ability, the Harter Perceived Competence Scale measured self-esteem, and the Children Need to Know Knowledge Attitude Test provided a measure of a child’s awareness and understanding of risk issues. Six months later the program was given to the control children and to the children in the treatment group who had failed the posttest simulation. The follow-up study (Fryer, Kraizer, and Miyoshi, 1987b) administered a third simulation to the 30 children who were still enrolled and present on the day of testing.
A second evaluation (Kraizer Witte, and Fryer, 1989) examined 669 (494 in the treatment group and 175 in a control group of students with similar profiles) students from three states, ranging in age from 3 to 10 years old. The researchers used a scripted role-play to measure behavioral change. In the simulation, each child was asked to help the interviewer by responding to a series of “what if’ stories. The simulation then measured the child’s willingness to speak up about unwanted touching; assessed the child’s behavioral consistency in the face of emotional pressure; evaluated his or her willingness to tell others about the incident; [about the experience? about the touching incident?]; and willingness to keep a secret. Scoring was based on verbal response and body language. Children received one point every time they refused to go along with the examiner, for a total of 14 possible points.
Key Evaluation Findings
In the Denver evaluation, Fryer, Kraizer, and Miyoshi (1987a) found that:
- In the posttest simulation, 78.3 percent of the children in the treatment group passed compared with 52.4 percent of the children in the control group.
- Children who had initially higher self-esteem scores were more successful in the second simulation.
In the follow-up study, Fryer, Kraizer, and Miyoshi (1987b) found that:
- All the children in the treatment group who passed the posttest simulation also passed the third simulation six months later, showing retention of skills over time.
- A small percentage of children did not benefit from even repeated exposure to the program.
In the second evaluation, Kraizer, Witte, and Fryer (1989) found that:
- Prior exposure to other prevention materials made no significant difference in the children’s knowledge or attitudes or in the role-playing score on the pretest.
- Children in the treatment group showed significant gains after the program. Out of a possible score of 14, the three treatment groups saw average gains of 3.7, 3.9, and 4.3 points, compared with 0.5 and 0.3 for the control groups.
- The role-playing aspect appears to have achieved the desired effect of changing children’s actual behaviors, rather than a child’s self-esteem or knowledge of and attitudes toward risk and safety.
- Young children (kindergarten, first grade, and preschool) showed the greatest gains from the program.
- 95.5 percent of the children reported having no fear or anxiety after the completion of the program.
Elementary and preschools
The original evaluation was funded through a grant from the National Center on Child Abuse and Neglect.
- A key component of the program is the use of role-playing, which allows children to practice the skills the program is trying to teach.
- The program’s goal is to provide only the information that is absolutely necessary to teach the prevention skills. Thus, the program does not provide explicit information about child abuse, does not define touching--as good or bad, and does not suggest that the adults that children know may abuse them.
- The program can effectively be used with young (preschool-aged) children.
The Safe Child Program is a videotaped curriculum delivered in ten different segments. The videotapes are used to teach the basic concepts and role-playing techniques. The videotapes are then followed up with actual role-playing, discussion, and other classroom activities.
Classroom teachers typically present the program, although school counselors and staff from local child abuse prevention agencies may also teach the sessions.
The professional training component of the program consists of an introduction to the program, an overview of the problems of child abuse, techniques for teaching the classroom program, instruction on how to recognize and report abuse, and information for implementation of the program.
Issues to Consider
This program received a “promising” rating. Neither the Fryer, Kraizer, and Miyoshi study nor the Kraizer, Witte, and Fryer study actually demonstrates a decrease in abuse. Rather, the evaluations show that this program is effective in changing children’s behavior in ways that should reduce the risk of child abuse from strangers. While the Denver evaluation used an experimental design and showed significant and positive effects, the sample was very small. The evaluation by Kraizer, Witte, and Fryer (1989) also used an experimental design with a much larger sample.
Both evaluations emphasized that the positive results are due to the role-playing aspect of the program. The evaluations found that correct answers on the paper tests were not predictive of success in the simulations, nor was previous exposure to educational materials. The researchers concluded that knowledge of facts and even conceptual understanding are not enough to translate into behavioral changes. Rather, children need to practice the skills.
Denver, Colorado; Nashua, New Hampshire; Tulsa, Oklahoma; Pearl River, New York; Dade County, Florida; and sites in Canada, Europe, and Australia.
Sherryll Kraizer, Ph.D.
Coalition for Children
P.O. Box 6304
Denver, CO 80206
Fax: (303) 320-6328
The Safe Child Program package is available for purchase in English, Spanish, French, and Creole and includes a teacher’s guide, implementation guide, resource manual, and parents’ videotape. Also included are separate videotapes and teacher manuals for each grade level.
A description of the program can be found at http://www.safechild.org/
Fryer, G. E., S. K. Kraizer, and T. Miyoshi, "Measuring Actual Reduction of Risk to Child Abuse: A New Approach,"
Child Abuse and Neglect
Vol. 11, pp. 173-179, 1987a.
Fryer, G. E., S. K. Kraizer, T. Miyoshi, "Measuring Children's Retention of Skills to Resis Stranger Abduction: Use of the Simulation," Child Abuse & Neglect Vol. 11, pp. 181-185, 1987b.
Kraizer, S., S. S. Witte, and G. E. Fryer, "Child Sexual Abuse Prevention Programs: What Makes Them Effective in Protecting Children?," Children Today September-October 1989, pp. 23-27, 1989.