Programs that Work
Parents as Teachers
Healthy and Safe Children
Children Ready for School
Children Succeeding in School
Students performing at grade level or meeting state curriculum standards
Children ages 0 to 5 exhibiting age-appropriate mental and physical development
Babies born weighing more than 5.5 pounds and improving outcomes for low birth weight babies
Age of Child
Early Childhood (0-8)
Type of Setting
Community-Based Service Provider
Type of Service
Health Care Services
Type of Outcome Addressed
Child Abuse and Neglect
Cognitive Development / School Performance
Evidence Level (What does this mean?)
The overarching philosophy of Parents as Teachers (PAT) is to provide parents with child development knowledge and parenting support. Through a four-part intervention model known as the Parents as Teachers Born to Learn® model, trained and certified parent educators offer support to families from pregnancy to the time the children entered kindergarten. The goals of PAT are to increase parent knowledge of early childhood development, improve parenting practices, detect developmental delays and health issues early, prevent child abuse and neglect, and increase children's school readiness and success. PAT program services include home visits to families; health, hearing, vision, and developmental screenings of children; parent group meetings; and a resource network that links families with needed community resources.
The PAT Born to Learn model is a universal-access model, adaptable for families of all configurations, in all life circumstances, and from all types of communities. With funding from the Missouri Department of Elementary and Secondary Education and the Danforth Foundation, New Parents as Teachers (NPAT) began in 1981 as a pilot project for first-time parents of newborns. The program was expanded statewide and renamed Parents as Teachers in 1984. Since 1985, programs offering PAT services have expanded to all 50 states and to other countries. In the 2006-2007 program year, approximately 3,000 PAT programs served close to a quarter of a million families and over 300,000 children.
Families from pregnancy through age 5
Pfannenstiel and Seltzer (1985) evaluated the effects of NPAT in the four original program sites in Missouri, where program services were provided to families from pregnancy to the time the children were three years old. The NPAT sample consisted of 75 families randomly selected from among participating families in each of the four sites, and these families were matched with 75 comparison families randomly selected from a governmental agency list of first-born children who were born during a specified time frame in the same geographic area as the PAT participants. Despite matching, the NPAT and control groups differed significantly on several characteristics, including parents' ages and educational levels at the time their children were born. The evaluation assessed the children's cognitive achievement, language ability, and social/developmental skills at age three. No pretest measures were available.
A 1989 follow-up study to Pfannenstiel and Seltzer (1985) evaluated child outcomes at the completion of first grade (Pfannenstiel, 1989). Ninety-two percent of the original NPAT group (69 children) and 65 percent of the original comparison group (49 children) were located for the follow-up evaluation. Analyses comparing the follow-up samples to those in the original evaluation did not find any significant differences between groups on family-related variables, suggesting that the follow-up sample was representative of the original study sample of NPAT and comparison children. Student outcomes were compared on reading and math achievement, teacher assessments of academic progress and social/behavioral adjustment in the classroom, and parent perceptions of their child's academic and social performance.
Drazen and Haust (1993) assessed a replication of the PAT program, Parents and Children Together (PACT), which was implemented in Binghamton, New York. PACT was initiated in 1987 and served 509 primarily Caucasian families. In the study, the cognitive, language, and physical development of the 20 oldest, most at-risk program participants at ages four to five were compared to control children (who had not participated in PACT) matched on various risk factors such as low income. Analyses confirmed no significant differences between the two groups on these risk factor variables. Outcomes were compared on assessments of physical development, cognitive achievement, language skills, report of suspected abuse or neglect, and family income in each group.
Drazen and Haust (1995), in a follow-up to Drazen and Haust (1993), studied a larger sample of PACT participants. The study sample involved all children born in 1987 who entered kindergarten in 1992 in the Binghamton district and were still enrolled in the system in 1996. Of this group, close to 10 percent (47 children) had participated in PACT until age three, while 412 had not (and formed the comparison group). Only 41 of the 47 PACT children completed outcome measures. Children in both groups were similar on all demographic measures, except that the PACT children were more likely to have at least one parent employed, were more likely to have had mothers with complications of pregnancy or delivery, and were significantly more likely than control children to have attended prekindergarten and/or other preschool programs. Child outcomes included a measure of kindergarten readiness (given to students at the beginning of kindergarten), tests of reading and mathematics readiness at the end of kindergarten, grade point average, and referrals for special education services.
Wagner, Cameto, and Gerlach-Downie (1996) studied the effects of a two-year PAT intervention among a sample of teen parents in California. The study involved four groups: a PAT group, a group that was given comprehensive case management, a group that combined PAT services with comprehensive case management, and a control group that received no direct services, but did receive regular mailings of child toys. A total of 717 teen mothers were recruited, more than half of whom were Latina, 20 percent were African American, and 20 percent were Caucasian. Almost one-third of the participants received Aid to Families with Dependent Children, two-thirds were enrolled in a high school program, and 30 percent were high school dropouts. Child outcomes were assessed at birth (for weight), and at one and two years on developmental scales (including physical, self-help, social, academic/cognitive, and communication) and Children's Protective Services data on child abuse/neglect reports. In addition, school dropout rates and pregnancy rates of teen mothers were compared. Overall, 402 teens dropped out of the demonstration between enrollment and the child's second birthday, for an overall attrition rate of 57 percent. The dropout rate was fairly evenly distributed among the groups, and the total sample of participants remaining in the study was 315. When the characteristics of program dropouts and those who stayed in the study (participants) were compared, a significant difference was found for rate of high school dropout (35 percent among program dropouts, and 24 percent among program participants).
The effects of PAT were studied by Coleman, Rowland, and Hutchins (1997) in a sample of three groups from a rural Southeastern county. The first group included 21 families who had entered the PAT program within four months of their child's birth, who had remained in the program until their child was 30 to 36 months of age, and who had enrolled their child in public school kindergarten at age five. These 21 families were from an original starting sample of 97 families (i.e., a five-year retention rate of 22 percent). The second group consisted of 22 children who did not receive PAT services, but whose parents requested and received quarterly educational newsletters during the first year following their child's birth. The newsletters were written by PAT staff and included identical information as that provided to the PAT families. The third group of 22 children served as the no-treatment control group, receiving neither PAT services nor newsletters. Demographic variables were used to match PAT families as closely as possible to newsletter-only families and control families. Child outcomes included a cognitive profile, language profile, gross motor profile, and a self help profile assessed at kindergarten entry.
Another study by Wagner and colleagues assessed the effects of a four-year PAT intervention in a sample of parents from Northern California (Wagner et al., 1999). Families with infants up to six months old were randomly assigned to PAT or to a control group that did not receive PAT services but did receive age-appropriate toys through the mail at regular intervals. A total of 497 families [does PAT deal with "families" or does it sometimes specifically target "mothers?"] enrolled in the study (199 in the PAT group and 298 in the control group), of which 81 percent were Latina. Both groups received annual child assessments at or around the children's birthdays until the children turned three. Of the 497 who enrolled, a total of 315 families participated in the age-one assessment (175 PAT group and 140 control). One year later, a total of 375 families participated in the age-two assessment (220 PAT and 155 control), and after another year 363 participated in the age-three assessment (210 PAT and 153 control). The PAT and control groups did not differ significantly from each other on any dimension that was measured at enrollment, however at the one-year assessment the control group had significantly fewer Latina mothers than the PAT group (74 percent versus 84 percent). Child outcomes were assessed on the Peabody Picture Vocabulary Test, the Bayley Scales of Infant Development, and a child developmental profile that included physical, self-help, social, academic/cognitive, and communication dimensions.
Wagner et al. (2001a) studied the effects of PAT in a randomized control trial of 667 children from three geographically dispersed sites, including one on the Eastern Seaboard (Site 1), one in a midsize Southern city (Site 2), and one in a large Western city (Site 3). Selected sites had been in operation for at least two years, served a high proportion of low-income families, and offered home visits at least monthly. Families with infants up to eight months old were recruited and randomly assigned to PAT or to a control group. Most of the background characteristics of treatment and control children were similar. However, in Site 1, significantly more PAT children than control children had Caucasian mothers, while in Site 2 significantly more PAT children than control children had siblings. In Site 3, PAT parents reported much more knowledge about infants than control parents did. Outcomes were assessed at or around the children's second birthdays, while participants were still receiving PAT services. Sixty-four percent of families in both treatment and control groups participated in at least one of the assessments—a child developmental profile that included physical, self-help, social, academic/cognitive, and communication dimensions.
Wagner et al. (2001b) conducted an age-three follow-up in Site 1 of the original study Wagner et al. (2001a). Site 1 originally included a sample of 206 families, and 181 families were retained in the sample at the age-three assessment. No significant differences in demographic or background variables were seen between the treatment and control groups. Child outcomes were assessed on a child developmental profile that included physical, self-help, social, academic/cognitive, and communication dimensions and a parent report of child pro-social behaviors.
Finally, Wagner et al. (2002) used data from Wagner et al. (2001b) to determine whether the impact of PAT on certain parental and child development measures was different when data was stratified by family income. Family income was divided into two categories, very low income (less than $15,000 per year for the household) and moderate income (more than $15,000 per year for the household).
Key Evaluation Findings
The original study of 75 NPAT children and 69 comparison children (Pfannenstiel and Seltzer, 1985) found the following:
- NPAT children scored significantly higher than comparison children in their ability to organize components to solve problems, intellectual functioning, verbal intelligence, and general knowledge. No significant differences were found between groups on the measure of sequential processing (e.g., word order and number recall).
- NPAT children also significantly outscored comparison children on all three measures of language skills, including auditory comprehension, verbal ability, and language ability.
- In terms of social/developmental skills, NPAT children were rated more positively than comparison children on 8 of the 21 social development items rated by researchers. On 13 of the 44 parental assessment items, NPAT parents rated their children significantly higher than did comparison parents, including, for example, on "separating easily from parents," "describing his/her feelings," and "differentiating present and future social roles." Comparison parents rated their children significantly higher than did NPAT parents on one item measuring how frequently their child enjoyed playing with other children. In all instances the effect size, while significant, was small. On the remaining 30 parental assessment measures there were no significant differences
- NPAT children scored significantly higher than did comparison children on longitudinal math achievement, with an average math score in the 82nd percentile compared with the 74th percentile. No significant differences were found between NPAT and comparison children for reading achievement.
- There were no significant differences between groups in teacher assessments of children's academic progress and social/behavioral adjustment. The one exception was spelling, on which teachers rated NPAT children significantly higher than they did comparison children.
- There were no significant differences between groups in parental assessment of academic and social performance.
- There were no significant differences between groups in average scores on the measure of cognitive abilities (only 13 PACT and 11 control students were tested).
- PACT students scored significantly higher than control students on language skills with significantly fewer PACT graduates (30 percent) than control students (65 percent) scoring below their age level on the test.
- Significantly more control participants than PACT participants had gross motor developmental delays on the total score of the developmental screening test.
- No significant differences were found between groups for the family report of suspicions of child abuse or neglect.
- Welfare dependence (receipt of AFDC) in both groups doubled between the time of the children's birth and one year later. From a child's first to second birthday, marginally significant differences emerged between groups, with welfare dependence declining in the PACT group (from 40 percent to 30 percent), and increasing among control families (from 30 percent to 50 percent).
- Children whose families participated in PACT had significantly higher school readiness scores on the three tests that were used (kindergarten, math, and reading readiness) than those whose families did not participate.
- PACT children also had significantly higher grades in kindergarten than did control children, with computed averages of 95 percent versus 93 percent.
- A significantly lower proportion of PACT participants than control participants were enrolled in remedial special education in first grade (14 percent versus 31 percent).
- There were no significant differences among teen parents in the four groups for high school dropout rates.
- While no significant differences were found among groups for having a subsequent pregnancy during the study period, significantly fewer PAT-only mothers than control group mothers had multiple pregnancies during this period (1.4 percent versus 4.8 percent). No significant differences were found among the PAT-only, case management, or combined intervention (PAT plus case management) groups.
- For those who entered the study while pregnant, mothers in the PAT-only group had marginally lower rates of low birth weight babies than did mothers in the control group (4 percent versus 8 percent). No significant differences were found among the PAT-only, case management, or combined intervention groups.
- There were no significant differences among groups for Child Protective Services reports of child abuse or neglect, although the combined intervention group was significantly less likely than the control group to have an opened case of child abuse or neglect (0 percent versus 2 percent). No significant differences were found among the PAT- only, case management, or combined intervention groups.
- No significant differences were found among groups at the one-year assessment for child scores on the developmental tests. At the two-year assessment, both the case management and combined intervention groups had significantly higher scores for cognitive development than did PAT-only or control groups. Additionally, the case management, combined intervention, and PAT groups scored significantly higher than the control group on the social development scale at age two. No significant differences were noted among groups regarding the children's physical, self-help, or communication development.
- When the sample was limited to teens who received the expected or recommended level of home visits or direct case management during their child's second year, those in the PAT-only group had more positive outcomes. Effects were again strongest for cognitive development, with both the combined intervention and the case management group (but not the PAT-only group) scoring significantly higher than the control group. The combined intervention, case management, and PAT-only groups also scored significantly higher than the control group on social development. No statistically significant differences were found among groups for physical, self-help, or communication development.
- PAT children scored significantly higher on the average language age equivalencies than children from the newsletter-only families and children from the control group. No significant differences were found between the newsletter-only and control groups.
- PAT children scored significantly higher on the average self-help/social age equivalencies than children from either of the other groups. In addition, the average age equivalency of children from newsletter-only families was significantly higher than the average of children from control families.
- No significant differences were found between groups on the cognitive or the motor development profiles.
- For the total sample (including children of both Latina and non-Latina mothers), scores on the researcher-assessed mental development scale indicated that one-year-olds in the PAT group significantly outscored one-year-olds in the control group. No significant differences were noted for two-year-olds or three-year-olds.
- Analyses comparing subgroups by ethnicity found that while one-year-old PAT children of Latina mothers significantly outscored controls, three-year-old control children of non-Latina mothers significantly outscored their PAT group counterparts.
- For the total sample, results for the self-help scale found that children in the PAT group significantly outscored children in the control group at both the one-year and three-year assessments.
- Further analysis suggests these findings may have been primarily attributable to the children of Latina mothers, as no significant findings were noted for children of non-Latina mothers.
- For the social development scale, children in the control group significantly outscored children in the PAT group at the age-two assessment.
- Subgroup analysis suggests that the above finding was consistent for children of non-Latina mothers only. Conversely, three-year-old children of Latina mothers in the PAT group significantly outscored controls on the measure of social development.
- On the physical development scale, children in PAT significantly outscored children in the control group at the one-year assessment. These results were consistent for both children of Latina mothers and children of non-Latina mothers.
- For the total sample, no significant differences were found between the PAT and control groups on the communication portion of the child development profile, the parent-reported cognitive development scale, or the Peabody Picture Vocabulary Test.
- For children of Latina mothers, significant differences were found on the parent-reported cognitive development scale for three-year-olds.
- No significant differences between PAT and the control groups on either the measure of cognitive and physical development or the parent report of adaptive social behavior inventory at age two.
- When the subgroup of very-low-income children was assessed separately, a significant difference favoring the PAT group was found for adaptive social behavior.
- PAT children showed significant, positive effects relative to control group children in the area of self-help development at the three-year assessment.
- There were no significant differences between PAT and control groups on the measures of cognitive development, physical development, or the parent report of adaptive social behavior inventory.
- Very low income PAT children showed positive effects of PAT on child and parental development outcomes compared to the control group; however these differences were generally not statistically significant.
- "Moderate" income PAT children showed mixed results on child development and parenting outcomes relative to the control group, with those differences generally not being statistically significant.
School districts, family resource centers, nonprofit organizations, public health and social service agencies, government agencies, and child care centers.
The NPAT program was originally funded by the Missouri Department of Elementary and Secondary Education in cooperation with the Danforth Foundation. The PAT expansion was legislated and partially funded by the Missouri General Assembly. Funds for other replications of the program have been generated from a variety of sources, including federal funds from Goals 2000: Educate America Act, Early Head Start, Even Start, and Title I. State and local dollars from school districts, corporate funders, private foundations, and social organizations can also be used.
Parents as Teachers National Center (PATNC) estimates program costs of $1,400 to $1,500 per family annually. Costs vary depending on service intensity, as well as on program location and the availability of in-kind contributions. Program start-up costs for implementing the PAT Born to Learn® Curriculum, Prenatal to 3 Years are approximately $595 for each parent educator to attend the one-week initial training, plus an additional $295 for the curriculum materials, not including the cost of the parent educator's time or cost of transportation, lodging, and meals during the training.
Programs offering PAT services are local implementations that may differ somewhat depending on their sponsoring agency, but all PAT programs provide the same four core services:
- Personal Visits: Personal home visits are the major service delivery component of the program. During these visits, parent educators share age-appropriate child development information with parents, help them learn to observe their own children, address their parenting concerns, and engage the family in activities that provide meaningful parent-child interaction. Visits are usually one hour long and are scheduled bimonthly, biweekly, or weekly, depending on family needs and local program budgetary restrictions.
- Group Meetings: Parent group meetings provide opportunities for parents to share information about parenting issues and child development. Parents learn from and support each other, observe their children with other children, and practice parenting skills. Many programs also offer informal drop-in and play times for families.
- Screening: Periodic child screening provides for early identification of developmental delays and health, vision, and hearing problems. Each child’s developmental progress is also reviewed regularly.
- Resource Network: Parent educators help families identify and connect with community resources. Programs take an active role in establishing ongoing collaborative relationships with other organizations that serve families.
The core of the program is the Born to Learn® Curriculum, Prenatal to 3 Years. It is an age-related curriculum that progresses through the mother’s pregnancy and monthly throughout the child’s first three years. The curriculum is designed to promote optimal child development and positive parent-child relationships. The content is adaptable, and parent educators can individualize the curriculum to the particular strengths, weaknesses, and needs of the participating families. It includes parent handouts to reinforce learning between visits.
Programs are implemented by trained parent educators. The majority of parent educators have at least a four-year college degree. Implementers tend to be nurses, teachers, social workers, and, in some communities, paraprofessionals. All parent educators must complete the Born to Learn Institute which requires 33 hours of pre-service training for initial certification, and continuing professional development for annual recertification.
Issues to Consider
This program received a "promising" rating. The three quasi-experimental evaluations find, generally, that PAT participants significantly outscored control group children on measures of language skills, cognitive abilities, physical development, and social development. In addition, one of the randomized controlled trials (Wagner et al., 1999) reports several significant program effects on social, self-help, physical, and cognitive dimensions—although effects varied by child's age and a significant difference was found for only one of three cognitive tests. No significant effects were found in either of the two studies that assessed child abuse and neglect outcomes, although Wagner (1999) did find an effect on these outcomes when PAT was combined with case management among teen mothers.
Despite the generally positive findings for the three quasi-experimental studies and one randomized control trial, it is important to consider that the Wagner et al. (2001a, 2001b) randomized controlled studies did not find any significant differences between children who participated in the PAT group and those in the control group. Further analyses of these studies by Wagner et al. (2002) found significant effects on PAT children as compared to the control group for three of the 45 outcomes measured by the study; however, given the large number of outcomes measured, a finding of only three significant outcomes may in fact be due to random chance rather than program success. Furthermore, although the three quasi-experimental studies show promising results for the PAT program, none used baseline (pretest) data on outcomes of interest to control for group differences at the start, two studies had small sample sizes, and two studies reported baseline demographic characteristics showing PAT participants were more advantaged than control participants.
All school districts in the state of Missouri currently offer PAT services. In addition, the program has been implemented in the 49 other states, as well as Australia, Belize, Canada, China, Germany, Mexico, New Zealand, and the United Kingdom.
Karen Guskin, Ph.D.
Parents as Teachers National Center
2228 Ball Drive
St. Louis, Mo. 63146
Tel (314) 432-4330 or 1-866-PAT-4YOU (1-866-728-4968)
Fax (314) 432-8963
Parents as Teachers National Center (PATNC) is a not-for-profit organization established in 1987 to provide leadership and training for Parents as Teachers programs.
A variety of curricular and training materials are available through PATNC. These materials are for organizations and early childhood professionals; they are not for use by individual parents. Available resources include:
Born to Learn® Curricula — Each contain detailed personal visit plans, activities for parent/child engagement, child development information, professional resources and parent handouts. Some have been translated into other languages, including Spanish, French, German and Mandarin.
- Born to Learn® Curriculum Prenatal to 3 Years
- Born to Learn® Curriculum 2 Years to Kindergarten Entry
- Born to Learn® Curriculum 3 Years to Kindergarten Entry
- A Closer Look: The PAT Standards and Self-Assessment Guide, which provides standards and quality indicators for implementing the PAT Born to Learn® model, as well as assistance for conducting a self-assessment.
- Supervisor's Manual and Program Administration Guide, a desktop resource for PAT program supervisors that details program administration, supervision, advocacy, fund development, marketing, and evaluation strategies.
- Supporting Military Families provides resources and parent materials for programs serving military families. It addresses issues faced by military parents and families including relocation, deployment, reunification, and combat stress.
- Nutrition and Fitness of Young Children focuses on nutritional requirements, feeding behaviors, and fitness issues for children birth to age three. Personal visit plans, educator resources and parent handouts follow the Born to Learn curriculum format. (Parent handouts have been translated for Spanish-speaking families.)
Coleman, Mick, Bobbie Rowland, and Betty Hutchins,
Parents as Teachers: Policy Implications for Early School Intervention,
paper presented at the Annual Conference of the NCFR Fatherhood and Motherhood in a Diverse and Changing World (59th), Arlington, Va., November 7-10, 1997.
Drazen, Shelley M., and Mary Haust, Raising Reading Readiness in Low-Income Children by Parent Education, paper presented at the Annual Convention of the American Psychological Association (101st), Toronto, Ontario, Canada, August 23, 1993.
Drazen, Shelley M., and Mary Haust, The Effects of the Parents And Children Together (PACT) Program on School Achievement, Binghamton, N.Y.: Community Resource Center, 1995.
Pfannenstiel, Judy C., New Parents as Teachers Project: A Follow-Up Investigation, Overland Park, Kan.: Research and Training Associates, 1989.
Pfannenstiel, Judy C., and Dianne A. Seltzer, Evaluation Report: New Parents as Teachers Project, Overland Park, Kan.: Research and Training Associates, 1985.
Pfannenstiel, Judy C., and Dianne A. Seltzer, "New Parents as Teachers: Evaluation of an Early Parent Education Program," Early Childhood Research Quarterly, Vol. 4, 1989, pp. 1-18.
Wagner, Mary, and Serena Clayton, "The Parents as Teachers Program: Results from Two Demonstrations," The Future of Children, Vol. 9, No. 1, 1999, pp. 91-115.
Wagner, Mary, Donna Spiker, and Margaret Inman Linn, "The Effectiveness of the Parents as Teachers Program with Low-Income Parents and Children," Topics in Early Childhood Special Education, Vol. 22, No. 2, 2002, pp. 67-81.
Wagner, Mary, Donna Spiker, Frances Hernandez, Julia Song, and Suzanne Gerlach-Downie, Multisite Parents as Teachers Evaluation: Experiences and Outcomes for Children and Families, Menlo Park, Calif.: SRI International, 2001a.
Wagner, Mary, Elizabeth Iida, Donna Spiker, Frances Hernandez, and Julia Song, The Multisite Evaluation of the Parents as Teachers Home Visiting Program: Three-Year Findings from One Community, Menlo Park, Calif.: SRI International, 2001b.
Wagner, Mary, Renee Cameto, and Suzanne Gerlach-Downie, Intervention in Support of Adolescent Parents and Their Children: A Final Report on the Teen Parents as Teachers Demonstration, Menlo Park, Calif.: SRI International, 1996.
Wagner, Mary, Serena Clayton, Suzanne Gerlach-Downie, and Mary McElroy, An Evaluation of the Northern California Parents as Teachers Demonstration, Menlo Park, Calif.: SRI International, 1999.