5 Alternatives to standardized testing

When my younger son was in school, he absolutely hated the yearly California standardized testing. He figured out that he got very little benefit from the week-long tests and would declare “we are only taking these tests for the teachers.” And since he felt most of his teachers didn’t understand him, he saw no reason to work hard to get good scores on these yearly, intensive tests, so he didn’t. What he didn’t realize then is how much the teachers hate those tests too. He was correct, teachers could benefit from the tests, but just as often they are punished. Teachers have so little control over how an individual class performs on standardized tests, that schools which have implemented pay tied to student performance find teachers salaries going up or down seemingly randomly.

Higher standardized test scores benefit schools and districts more than teachers and handcuff the teacher from teaching anything that isn’t on the test. Those kinds of tests only measure test taking skills, not knowledge. They don’t predict future success for the student, and they are often sexist, racist, and classist. And if my son is any indication, they probably aren’t very accurate.

No one really likes these tests, but what is the alternative? How can we measure student progress in a way that is actually useful?

Here are 5 methods currently being used to assess student progress. 

Portfolio-based Assessment

When I asked my now 30 year old son what he wished had been done as an alternative to those hated standardized tests, he responded with portfolio assessment. He is now a part time professor and a full time technical artist and can see the importance of this as a life skill.

A portfolio serves as a compilation of student work meant to show growth over time.  Portfolios are collections of information relating to each child’s developmental progress in an educational setting. Sometimes students give a presentation in front of a panel of teachers at the end of the school year. In this way, they can create their own portfolio and have agency over their own work. PBA’s are usually graded with a rubric, not a percentile. They address skills like presentation, communication, and teamwork that are common in the workplace. PBA’s are very common in art and writing classes but they are now used more generally. Oakland Unified school district and the New York Performance Standards Consortium, a group of schools in New York, uses portfolio-based assessment. 

Problems with PBA include the amount of prep time required from both students and teachers, and difficulty in ensuring fairness in grading a portfolio. Grade inflation/deflation is already a problem in schools so it is unclear what would happen if a student’s grades were based only on teachers’ subjective interpretation.

Sampling 

Research in many areas requires some form of sampling. This is a dramatic departure from our testing framework while still retaining a test. The idea is that if you draw an accurate sample of the group you are studying you are able to draw meaningful conclusions. There are many ways to do this including administering traditional standardized tests to a statistically representative sampling of students, rather than to every student every year. That is how both the National Report Card and PISA gather important data about how students are doing. Sampling is useful because it gives a big picture view of what is happening in education.

This also completely changes the motivation. Sampling is useful for looking at how education is doing as a whole, but it would be even more nonsensical to teach to a “sampling” test. The tests would be shorter and a little more random for each student. Students couldn’t be judged on the results of a sampling test, and it is unlikely that teachers could be either. In fact it is only schools or districts where you could get enough data to form a pattern and since that is the group that cares about the test data anyways, it could provide the desired data without the negative consequences.

However, it would be a hard, hard sell to stakeholders. People who get all the data will continue to want as much data as they can get and for many reformers, any sort of testing has a black mark on it.

Stealth / Game Assessment

There are many programs, websites and apps that allow students to practice subjects like math and English. These kinds of programs already passively collect data on student’s progress and are being used regularly by many teachers. And a big advantage is that students are able to replay the games until they master the concepts.

Better educational games that can assess more than just rote memorization, but can also measure the way students solve problems and move through the world could provide exactly this kind of stealth assessment. It is a little beyond where the EdTech games are now, but it could be coming up the pipeline. This could be coupled with on demand assessments where the child decides when they’re going to take a test which actually counts towards a grade.

The MacArthur foundation did an excellent study on stealth assessments and gaming as a tool to show competence in students. It showed that stealth assessment doesn’t just show which skills a student has mastered at a given moment, but the pattern of answers offers insights into how quickly students learn, how persistent they are, and how well they work with others.

There is still some technical progress that needs to be made here before it can be rolled out wide, but it shows a lot of promise.

SEL Assessments 

Social and emotional learning  skills are now seen as vital to long term student success in school. Researchers are continuing to find that nonacademic qualities play a huge role in determining one’s success. Characteristics like grit and perseverance are important and educational institutions are beginning to realize that these skills should be a priority when it comes to teaching and measurement.

Many school districts, like LA Unified, have already incorporated SEL into their curriculum.

However, like sampling, the goal for testing this is completely different from the extant standardized tests. It seems more likely that SEL assessments will be added to the regular set of tests than replace them, which could simply add another thing for teachers to try to “teach to”.

Multiple Measures

Multiple Measures includes all of the above alternatives as well as smaller, more frequent versions of standardized testing. As the name describes, multiple measures is a way of using multiple forms of data to track and better understand how a student is performing. This data is collected to get a more detailed picture of how each of our students are doing and how schools are doing collectively. We could then incorporate more, and different, kinds of data on student progress and school performance into accountability measures. Data systems now track students in most states from pre-K all the way through high school. We could take a big-data approach, combining information from many different sources to help districts, teachers and students.

Some states, including California, have a wealth of data freely available on the internet that could be used.

These are just some possibilities, but we need a more fair and accurate method to assess our students and give our teachers the freedom to teach as their experience and instincts tell them to, not just as the test dictates.