placebo controlled study methods to education-FINALLY

[dwolen]

Here's a link to a NYT article about the need to applying placebo controlled trials to actually determine the effectiveness of various teaching techniques. Hallelujah! Hooray, finally! Hopefully this will end the "testimonials," hunches, all other non-evidence based decision making in education.

nyti.ms/17Ft0D5

Guesses and Hype Give Way to Data in Study of Education
BY GINA KOLATA September 2, 2013.
I don't know if this link will work. If not, Healthy11 will usually save the day!

[healthy11]

The link works!

I know there have been a number of previous studies about teaching reading, but far fewer about math programs, so I found the following to be of particular interest:

Evaluation of Early Elementary Math Curricula

Contractor: Mathematica Policy Research; SRI International

Background/Research Questions:

The Title I, Part A program is intended to help ensure that all children have the opportunity to obtain a high-quality education and reach proficiency on challenging state standards and assessments. As the largest federal program supporting elementary and secondary education, these resources are targeted primarily to high-poverty districts and schools.

There has been very little reliable information available to educators and policy makers about which curricula are most likely to improve math performance. The evaluation is focused on early elementary grades since disadvantaged children are behind their more advantaged peers even before entering elementary school in basic math competencies. The study is examining:

What is the relative effectiveness of different math curricula on student achievement in early elementary schools?
Under what conditions is each math curriculum most effective?
What is the relationship between teacher knowledge of math content and pedagogy and the effectiveness of the math curricula?

Design:

The evaluation questions are being addressed through an experimental design in which schools were randomly assigned to selected math curricula; there is no control group. Math curricula were selected for the evaluation through a competitive process. The math curricula being evaluated are widely-used and representative of different instructional approaches, and are appropriate for funding under Title I. The curricula are Investigations in Number, Data, and Space (Pearson Scott Foresman), Math Expressions (Houghton Mifflin), Saxon Math (Harcourt Achieve), and Scott Foresman-Addison Wesley Mathematics (Pearson Scott Foresman). Implementation of the math curricula and their impact on first-grade student achievement was measured in 4 districts and 39 schools during the 2006–07 school year. Programs were implemented in an additional 71 schools and also in second grade during the 2007–08 school year. The impact is the difference in math achievement between the groups using the selected math curricula.

Cost/Duration: $21,039,437 over 5 years (September 2005–September 2013

Current Status:

A report on first grade results during the 2006–07 school year was released in February 2009 (http://ies.ed.gov/pubsearch/pubsinfo.asp?pubid=NCEE20094052). The second report on a larger sample of first and second graders was released in October 2010 (http://ies.ed.gov/pubsearch/pubsinfo.asp?pubid=NCEE20114001). A third report is expected in 2013.

Key Findings:

The second report looked at the effects of the curricula on the total sample after the first year of implementation in both the first and second grades:

In first-grade classrooms, students taught using Math Expressions scored an average of 0.11 standard deviations higher on the ECLS-K math test than students taught using either Investigations or SFAW. This difference in test scores is equivalent to moving a student from the 50th to the 54 percentile.
In second grade classrooms, students taught using Math Expressions and Saxon scored an average of 0.12 and 0.17 standard deviations higher respectively on the ECLS-K math test than students taught using SFAW. These differences are equivalent to moving a student from the j50th to the 55th or 57th percentile respectively. Saxon teachers reported spending an average of one hour more per week on math instruction than teachers using the other curricula.

[dwolen]

$21 million+ for this math study? Wow!

[healthy11]

It does sound like a LOT of money, but I suppose if it's a 5 year study, that's like $4 million per year, and they said 110 schools were participating, so that's around $40,000 per school per year, and if they're evaluating 4 different math programs, it's $10,000 per program per school. It could equate to a full-time person at each school, I suppose...