Testing Theory to Explain and Solve the Head Start Fade (The Fade Phenomenon) 285Since the 1970s, reports on achievement in most Head Start programs showed little gain
and a lack of long-term outcomes. Decades of reports nationally on Head Start and Pre–K
programs such as those from the Department of Health and Human Services (DHHS, 2003)
showed results similar to those explained by Barnett (1995, 2003) and others.
Over the years, however, other Pre–K programs like Head Start have been developed from
research and theories and have provided gains to suggest the Fade may not have to exist.
Early childhood research and development has engendered some programs that have
demonstrated significant gains in Pre–K and Head Start-age children. Examples include the
North Carolina Abecedarian program (Ramey & Ramey, 1989), the Perry Child Development
Center (PCDC) program in Michigan (DHHS, 2001, 2003), and the Chicago Child and Parent
Centers or CCPC (Barnett, 1995; Reynolds, Temple, Robertson, & Mann, 2003). These Head
Start and Pre–K programs provide theories supported by significant, empirical, long-term
gains for students where little or no Fade was evident during grades K–3. These positive
effects have been evident for children living in poverty and for minorities.
Research results provided in K–3 early childhood studies such as those in the empirical
research-based 1985–1990 STAR experiment (Word et al., 1990) and other studies of small
classes have exposed an evidence-based, underlying theory that offers a fresh look at viable
options to explore, moderate, and /or eliminate the Fade effect, especially for children living
in poverty and minority students. The STAR experiment and related class-size studies have
shown significant longitudinal gains from a randomized experiment. In STAR, students and
staff were randomly assigned to one of three class conditions: the small class (n=13– 17
students) outperformed peers in the larger classes (n = 22–25 students), as well as those in
larger classes with a full-time teacher aide. STAR small classes showed gains of four to six
months beyond larger classes at each grade level and demonstrated even larger gains for
minority students (Finn & Achilles, 1999), as well as improved performance and student
behavior (Finn, Pannozzo, & Achilles, 2003). STAR researchers have shown that the small
class groups had increased enrollment in high-school advanced classes and higher graduation
rates. Problems in the Head Start research designs and potential solutions to the Fade became
evident when these robust class-size research and evaluation outcomes became available.
A Pilot Test of Reducing the Fade Effect
The STAR experiment’s small-class results highlight the elements of successful transitions
for children from Pre–K to K–3 programs. Reports from the Family Child and Education Study
(FACES, 2001) suggested that these same elements might occur in transitions from Pre–K to
K–3 and might mitigate the Fade effect. Theories drawn from effective Pre–K, Head Start, and
other coherent programs were investigated in a small study (Clarke, 2007). Results suggested a
new direction to take to moderate the Head Start Fade. The study highlighted a small group of
Pre–K (Head Start) through grade three and four students who began early intervention in
Head Start programs similar in design to exemplars like the Abecedarian, PCDC, and CCPC
programs. These students then entered small-class programs in grades K–3 similar to STAR
that provided coherence (Newmann, Smith, Allensworth, & Byrk, 2001, and a psychological
sense of community or PSOC (Sarason, 1974).
The retrospective predictive study (Johnson, 2002) was conducted using data from 1999–
2007 three Michigan districts. Unfortunately, there was considerable subject mortality
because of changes in state testing and reporting formats, gaps in community-based Head
Start records that were lost in transitions to K–4 school-based programs, and uneven use of
small-class models like STAR and small-class case studies such as Burke County, North