Series: Evidence Speaks
Up Front

The Tennessee pre-K study

Dale C. Farran and Mark W. Lipsey

State investments in center-based school readiness programs for preschoolers (pre-K), whether targeted for poor children or universally implemented, have expanded more rapidly than evaluations of their effects. Given the current interest and continuing expansion of state funded pre-K, it is especially important to be clear about the nature of the available evidence for the effectiveness of such programs. Despite widespread claims about proven benefits from pre-K, there is actually strikingly little credible research about the effectiveness of public pre-K programs scaled for statewide implementation.

Like many states that became interested in scaling up a state funded pre-K program in the early 2000’s, voluntary pre-K (TNVPK) was introduced in Tennessee in 1996 as a way to provide academic enhancement to economically disadvantaged children. It expanded in 2005 to an $85 million-plus statewide investment serving 18,000 Tennessee income-eligible children in 935 classrooms across all 95 counties.

Launched in 2009, the TNVPK Effectiveness Study, a coordinated effort between Vanderbilt’s Peabody Research Institute and the Tennessee Department of Education, is the first randomized control trial of a scaled up state funded pre-K program and the first well-controlled comparison group study of the effects of program participation as children progress through elementary school.

In a more detailed report available here we present and defend findings from the full evaluation report.  We summarize the longitudinal effects of TNVPK on pre‐kindergarten through third grade achievement and behavioral outcomes for a sample of 1076 children, of which 773 attended TNVPK classrooms and 303 did not. Both groups have been followed since the beginning of the pre‐k year.  Children in TNVPK classrooms made initial strong gains and were perceived by their teachers at kindergarten entry as being better prepared.  However, the achievement of the control children caught up to that of the pre-K children by the end of kindergarten.  In second and third grades achievement trends crossed over, with academic achievement for the pre-K children becoming worse than for the control children.

Critics of our study have argued that the effects reflect the unusually poor quality of the TNVPK program.  We demonstrate that measured classroom quality in the TNVPK classrooms was virtually identical to that in programs that have been lauded by pre-K advocates.  We demonstrate that gains by children in the TNVPK program during the pre-K year were substantial relative to the control group and in keeping with the effect sizes obtain by programs that advocates have identified as successes and in other scaled up state programs.  The evidence is simply inconsistent with the TNVPK program being unusually weak. 

Our results demonstrate the need for a stronger and more current evidentiary base on scaling up pre-K.  The shift to caring for 4-year-olds in public schools is a relatively recent one based largely on faith that this is beneficial for the participating children.  As these programs come under the administrative control of the public school system and are implemented in far-flung areas of a state, it is necessary to determine what the consequences are and what safeguards might need to be put into place.  A clear, well-articulated vision for how the care of 4- and 5-year-olds differs from that for older children is needed to protect these classrooms from becoming junior kindergartens.  And a specific, perhaps new, definition is needed for “high quality.”



Mark W. Lipsey

Director of the Peabody Research Institute, Vanderbilt University