Currently, the primary source of information on outcomes for children and families served by Head Start comes from the Family and Child Experiences Survey (FACES).(4) Data from FACES are displayed in the figures below. Figure 1 shows the low scores and limited progress of Head Start children in the key areas of language, pre-reading, and pre-mathematics. These data are from the class of children who entered the program in 1997. The percentile scores show how Head Start children perform compared with the average performer. On a percentile scale, an average performer would be at the 50th percentile, meaning that half of children who take the test score above the average performer and half score below the 50% mark.(5) Head Start children as a group fall far below the 50th percentile in all areas of achievement. Though children are making some progress, clearly few children perform as poorly as children who enter and leave the Head Start program.(6)
* Statistically significant difference between Fall and Spring.
Figure 2 shows a similar picture for children who entered the program in 2000. Though children in Head Start programs performed somewhat better in 2000 than in 1997 in some areas, scores remain far below the average level in all areas of competency.(7)
* Statistically significant difference between Fall and Spring.
Both higher achieving and lower achieving Head Start children have low scores overall and show limited progress. Children who were in the upper 25% of their Head Start class when they entered Head Start in 1997 showed no gains on any measure of cognitive ability over the course of the Head Start program year, and actually experienced losses on some measures in comparison to national norms. Gains over the Head Start year were limited to children who were in the bottom 25% of their class.(8) However, even these gains fell far short of bringing children to levels of skill necessary for school success. For example, children in the bottom 25% of their Head Start class left Head Start with language skill scores at the 5th percentile, meaning that only 5% of all children who take the test score lower than these Head Start children do. Findings for mathematics showed a similar pattern.
The more recent 2000 FACES data show modest improvement in results for children, but overall progress is still too limited. Children continue to lag behind national norms when they exit Head Start. Data from Head Start FACES 2000 shows that:
- The level of childrens achievement in letter-recognition(9) for the 2000 Head Start year is far below the majority of U.S. children who know all letters of the alphabet upon entering kindergarten, according to the Early Childhood Longitudinal Study of the Kindergarten class of 1998.(10)
Spanish-speaking children in Head Start did not gain at all in letter recognition skills in 2000.
- Although writing scores increased 2 points during the 2000 Head Start year, this was a drop from children who entered Head Start in 1997 who increased 3.8 points in writing during the 1997 Head Start year.
- Children entered Head Start in 2000 with scores at about the 16th percentile in vocabulary, or about 34 percentile points below the average. Children entering Head Start scored at about the 31st percentile in letter recognition and at about the 21st percentile in early mathematics.
- Children who entered Head Start in 2000 made progress in early mathematics during the Head Start year that was statistically significant; however the difference was small (from 87.9 to 89.0 on a scale for which 100 is the average). As Figure 1 shows this 1.2-point difference is not a substantial gain toward national averages. Moreover, this amount of progress was no greater than that found for children who attended Head Start from Fall to Spring in 1997.
- Children who entered the program in 2000 with overall lower levels of knowledge and skill showed larger gains during the program year compared to children who entered with higher levels of knowledge. However, they still lagged far behind national averages.(11)
- Head Start children did not start kindergarten with the same social skill levels as their more socio-economically advantaged peers, and they continued to have more emotional and conduct problems.(12)
- A follow-up study of children enrolled in Head Start in 1997 showed that children who attend Head Start make less progress than the average kindergartener. Thirty-four percent of Head Start children showed proficiency in knowing the ending sounds of words, 53% in knowing the beginning sounds of words, and 83% in letter recognition. Data from a nationally representative sample of all first-time kindergartners shows that fifty-two percent demonstrated proficiency in knowing the ending sounds of words, 72% in knowing the beginning sounds of words, and 94% in letter recognition.(13)
- In 2000, the mean standard score for vocabulary increased 3.8 points, from 85.3 to 89.1 on a scale for which the average is 100. This result is similar to the data for 1997 that showed Head Start children scored about 85 at the beginning of the year and gained about 4 points by the end of the year.
- In 2000, the mean standard score for writing increased by 2 points, from 85.1 to 87.1.
- In 2000, children showed gains in book knowledge and print conventions (that is, they can show an adult the front of a storybook and open it to where the adult should start reading). This progress is statistically greater than for the 1997 Head Start year during which no progress was made in this area.
- In 2000, Spanish-speaking children in Head Start showed significant gains in English vocabulary skills without declines in their Spanish vocabulary.
- In 2000, children showed growth in social skills and reduction in hyperactive behavior during the Head Start year. Even children with the highest levels (scoring in the top quarter) of shy, aggressive, or hyperactive behavior showed significant reductions in these problem behaviors. Teachers rated childrens classroom behavior as more cooperative at the end of the Head Start year than when children first entered the program.(14)
- In 2000, children who received higher cooperative behavior ratings and lower problem behavior ratings from Head Start teachers scored better on cognitive assessments at the end of kindergarten, even after controlling for their scores on cognitive tests taken while in Head Start.
- Children who entered Head Start in 1997 showed significant gains in their social skills, such as following directions, joining in activities, and waiting turns in games, and gains in cooperative behaviors, according to ratings by teachers and parents. The quality of childrens social relationships, including relating to peers and social problem solving, also improved.
Head Start program and teacher characteristics show some positive relationships to educational and social outcomes for children. Examples include:
- Teachers educational credentials are linked to greater gains in early writing skills. Children taught by Head Start teachers with bachelors degrees or associates degrees showed gains toward national averages in an assessment of early writing skills, whereas children taught by teachers with lesser credentials merely held their own against national norms.
- Provision of preschool services for a longer period each day is linked to greater cognitive gains. Children in full-day classes in Head Start showed larger fall to spring gains in letter recognition and early writing skills than did children in part-day classes.
Head Start has other positive qualities:
- In 1997, the program received very high ratings of satisfaction from parents, and for the roughly 16% of children in Head Start with a suspected or diagnosed disability, 80% of parents reported that Head Start had helped them obtain special needs resources for the child.
- A follow up study of children who attended Head Start in 1997 showed that children were capable of making some progress during their kindergarten year in vocabulary, writing, and early mathematics, though performance remained significantly below national norms.
How do eligible children fare when they do not receive Head Start services? The FACES study is not designed to answer this question; there is no control group. Eligible children who do not receive services could be falling further behind or could be making gains similar to or greater than those for children in the program. The national Head Start Impact Study was launched in 2002 and is using a randomized design to answer this question. Additional experimental studies are being conducted to assess the effectiveness of specific quality improvement strategies.(15)
A national study of Early Head Start, which is part of the Head Start program serving low-income pregnant women and children from birth through three, was recently conducted using a randomized experimental design. Results show that children receiving Early Head Start have scores that are statistically higher than their peers who did not receive Early Head Start on measures of cognitive, language, and social and emotional competency. Fewer Early Head Start children scored in the at-risk range of functioning in both language and cognitive functioning. However, Early Head Start children continue to perform below the national average.
In summary, there is more work to do. Despite the positive qualities of Head Start programs, children in Head Start are making only very modest progress in only some areas of knowledge and skill, and children in Head Start are leaving the program far behind their peers. More progress must be made and can be made to put Head Start children on par with others by the time they enter kindergarten.