Examining the “Race to the Test” Guidelines (part I)

What’s to like and what’s to fear in the administration’s guidelines for the $350 million “Race to the Test” competition?

The Big Picture

There are really two competitions here: the $320 million fund for comprehensive assessment systems (current NCLB tests) and a $30 million fund for high school end-of-course tests. The administration projects two $160 million comprehensive assessment winners and one end-of-course testing winner–a total of three winners. State consortia are the only eligible applicants for these funds and consortia applying for comprehensive assessment system grants must have at least 15 states represented.

However, this “Race” may be more akin to your elementary school’s annual field day–everybody’s a winner! Right now, there appear to be only three consortia competing for three awards.

Much in Common

Advocates of comparability across states will be pleased. Each of the two comprehensive assessment consortia are said to have 30, perhaps 40 members. Member states must agree to implement at least the summative component of the system by the 2014-15 school year. And, each consortia must propose a plan for not only common standards, but also common performance-level descriptors, achievement standards, definitions of English learners, and accommodations policies (see comprehensive assessment systems selection criteria, part A, section V). These would go a long way towards ending many of the games that states play with their testing systems.

All this commonality will cut two ways. Importantly, states will be somewhat insulated from a variety of pressures to lower their cut scores (think grade inflation). Yet, at the same time, will the need for commonality and agreement among a large number of states work against ambitious, innovative approaches? For instance, there appears to be wide agreement that technology must play a large role in any future system — not only to allow for more innovative item types beyond multiple choice, but also to control costs. If all consortia members aren’t ready to implement online testing, the consortia will be held back by what’s possible in a paper-based delivery format. And, costs will explode in a system that must operate in both modes.

We’ve already seen a number of different consortia meld into one. Given that there will only be two awards, the temptation will be to have as many states as possible included in each consortia. That would be a mistake.**

Developing and implementing a new assessment system is much more complicated than adopting common standards. While other countries have elements of the various proposed assessment systems in place (and we’ve done pieces here), we’re going to be trying out a lot of new and different approaches (and if we aren’t it’s a big waste of money). It will already be a monumental feat for any state consortium to actually manage a four-year, $160 million research, development, and deployment process against an aggressive time line. Complicating the governance issues by doubling the number of states involved will increase the level of difficulty exponentially.

Given all that is going on with state policy, few states will have had the opportunity to truly develop a vision and understanding the of the consortia that they are joining. To succeed, a consortia must have a clear and shared vision among its members.

Yes, I want comparability. But not at the expense of mediocrity. Many pundits urged Secretary Duncan to hold a firm line on the quality and number of proposals accepted for “Race to the Top.” This time, it looks like the consortia themselves must hold the line on the quality and coherency of their approaches.

**Yes, consortia could write their rules to manage these issues. But, I have serious doubts as to how practical it will be to enforce these rules.

[This is the first of three posts on the administration's $350 million initiative to improve assessments. See my in-progress grades for the administration's initiative and a look at why college outcomes data is critical for effective assessments and measurements.]

(h/t to Stafford Palmieri for the great “Race to the Test” name).