Student Learning Objectives (SLOs): What (Little) We Know about Them Besides We Are to Use Them

ShareTweet about this on TwitterShare on Facebook7Email this to someoneShare on Google+0Share on LinkedIn0Share on Reddit0

Following up on a recent post, a VAMboozled! follower – Laura Chapman – wrote the comment below about Student Learning Objectives (SLOs) that I found important to share with you all. SLOs are objectives that are teacher-developed and administrator-approved to help hold teachers accountable for their students’ growth, although growth in this case is individually and loosely defined, which makes SLOs about as subjective as it gets. Ironically, SLOs serve as alternatives to VAMs when teachers who are VAM-ineligible need to be held accountable for “growth.”

Laura commented about how I need to write more about SLOs as states are increasingly adopting these, but states are doing this without really any research evidence in support of the concept, much less the practice. That might seem more surprising than it really is, but there is not a lot of research being conducted on SLOs, yet. One research document of which I am aware I reviewed here, with the actual document written by Mathematica and published by the US Department of Education here: “Alternative student growth measures for teacher evaluation: Profiles of early-adopting districts.

Conducting a search on ERIC, I found only two additional pieces also contracted out and published by the US Department of Education, although the first piece is more about describing what states are doing in terms of SLOs versus researching the actual properties of the SLOs. The second piece better illustrates the fact that “very little of the literature on SLOs addresses their statistical properties.”

What little we do know about SLOs at this point, however, is two-fold: (1) “no studies have looked at SLO reliability” and (2) “[l]ittle is known about whether SLOs can yield ratings that correlate with other measures of teacher performance” (i.e., one indicator of validity). The very few studies in which researchers have examined this found “small but positive correlations” between SLOs and VAM-based ratings (i.e., not a strong indicator of validity).

With that being said, if any of you are aware of research I should review or if any of you have anything to say or write about SLOs in your states, districts, or schools, feel free to email me at

In the meantime, do also read what Laura Wrote about SLOs here:

I appreciate your work on the VAM problem. Equal attention needs to be given to the use of SLOs for evaluating teacher education in so-called untested and non-tested subjects. It has been estimated that about 65-69% of teachers have job assignments for which there are not state-wide tests. SLOs (and variants) are the proxy of choice for VAM. This writing exercise is required in at least 27 states, with pretest-posttest and/or baseline to post-test reports on student growth. Four reports from USDE (2014) [I found three] show that there is no empirical research to support the use of the SLO process (and associated district-devised tests and cut-off scores) for teacher evaluation.

The template for SLOs originated in Denver in 1999. It has been widely copied and promoted via publications from USDE’s “Reform Support Network,” which operates free of any need for evidence and few constraints other than marketing a deeply flawed product. SLO templates in wide use have no peer reviewed evidence to support their use for teacher evaluation…not one reliability study, not one study addressing their validity for teacher evaluation.

SLO templates in Ohio and other states are designed to fit the teacher-student data link project (funded by Gates and USDE since 2005). This means that USDE’s proposed evaluations of specific teacher education programs ( e.g., art education at Ohio State University) will be aided by the use of extensive “teacher of record” data routinely gathered by schools and districts, including personnel files that typically require the teacher’s college transcripts, degree earned, certifications, scores on tests for any teacher license and so on.

There are technical questions galore, but a big chunk of the data of interest to the promoters of this latest extension of the Gates/USDE’s rating game are in place.
I have written about the use of SLOs as a proxy for VAM in an unpublished paper titled The Marketing of Student Learning Objectives (SLOs): 1999-2014. A pdf with references can be obtained by request at

ShareTweet about this on TwitterShare on Facebook7Email this to someoneShare on Google+0Share on LinkedIn0Share on Reddit0

One thought on “Student Learning Objectives (SLOs): What (Little) We Know about Them Besides We Are to Use Them

Leave a Reply

Your email address will not be published. Required fields are marked *