Testing TCKs

This is the second of two blogs focused on testing by Melissa Shipman, Executive Director of PACE and a teacher of children with special needs for over 10 years. Melissa lives in Georgia, USA with her husband and two daughters.

Last week, I provided a simple explanation of the types of educational testing, going into more detail about norm-referenced testing. Norm-referenced testing provides information to parents about how their child is mastering academic skills relative to their peers.

In testing-speak, a “peer” is a child very similar to the child being tested. In national, norm-referenced tests, the performances of thousands of children all over the country are combined to yield statistical information that helps us score these tests. This allows us to pinpoint that Janie is performing, for example, on a third grade, sixth month level.

Here is the challenge to the validity of the scores on these tests for our TCKs. We know that many factors affect the performance of a child when they take standardized tests. These factors include:

• Background knowledge

• School experience

• School curriculum

• Family factors

• General health and well-being

• Day-of-test factors (sleep, breakfast, testing environment)

Let’s talk about our fictional TCK named Janie, who has already lived in two different countries besides her parent’s passport country (the US) and attends a French school in northern Africa. Considering the bulleted list above, how does she come into a test different than her “peer,” or the average US schoolchild?

• Background knowledge: Janie has never lived in the US for a significant period of time. She is less familiar with cultural norms, folk and fairy tales, and money in the US than her testing peers.

• School experience: Janie has never attended an American school. For most of her education, she has attended a French national school in Africa. Even with the supplementing in English that her parents do with her nearly every day, she is less familiar with the mathematical symbols and literature taught in American schools, and has had less practice reading and writing in English.

• School curriculum: This is vastly different for Janie in her French school.

• Family factors: If anything, Janie’s parents may be slightly better educated than the average parents of her US peers, and she has been raised in a safe and loving environment.

• General health and well-being: If Janie has not recently made any sort of transition or experienced any stress related to cross-cultural living and education, this may not be a factor. But for many of our TCKs, we know that they are often experiencing significant challenges of one variety or another.

• Day-of-test factors (sleep, breakfast, testing environment): Janie takes this test at her organization’s annual conference. She has been staying up later, eating differently (less at meals because the food is unfamiliar, but more snacks that the volunteer team from the US brought with them – so less protein and more Oreos!), and wishes that she didn’t have to miss time in the kids program to take this test.

So the big question is this: how valid is Janie’s test score?

Here’s what I tell parents when I’m involved with testing: the less similar a child is from their “peer,” or an average American (if it’s a US test) school child of the same age, the less valid the test score.

What is “validity”? The validity of a test indicates that it is truly measuring what it claims to measure. If a child in the UK scores a year behind in a particular area on a UK-normed test, we can safely assume that there is a problem that needs to be addressed – perhaps this child has difficulties in processing certain types of tasks, or perhaps they have missed a lot of school and need remediation. Whatever the cause, we can assume that they are, indeed, performing below expectations.

When a TCK scores below grade or age-level on a test, I do not make the same assumption. I always assume first that low test scores are clues to how a TCK differs from the norming sample, not that they indicate a learning problem.

As education consultants and leaders in working with our families, we will all develop different tolerances related to testing. Personally, I have a problem when children who are not from the US (and by that I mean that their parents are not from the US) are administered standardized tests from the US. I just feel that there are too many areas in which they significantly differ from the norming sample, and that the sample doesn’t represent their “peers” in any way except that they are human children of the same age.

However, I have colleagues who I highly respect with different tolerances and who have no problem administering tests cross-culturally. These colleagues believe that if a child of any nationality is attending an international school using, for example, an American curriculum, a US-normed test is acceptable. While I may not agree, I can certainly understand their logic.

The important point here is not that we develop absolute opinions. The important thing is that we serve our families well, and that we’re able to help them understand the information that is yielded from test scores.

I enjoy the process of problem-solving low test scores. So, considering again our fictional TCK Janie, let’s say that she had some low test scores (a grade level behind) in spelling and math computation. After discussing this with her parents, we would likely conclude that her spelling is impacted because she operates academically mostly in French. And, after considering a sample of her work from school, we would realize that the differences in math notation between French and US schools were probably what gave Janie problems on the test.

I would ask how Janie is performing in spelling and math in school. If her parents reported that her performance in those areas was good, I would encourage them to not worry at all about these isolated scores on one test. My goal would be to help them see the “big picture” of how Janie was doing academically. They could just continue to supplement Janie’s education as they are already doing, perhaps taking a little extra time with spelling and math. I would encourage them not to believe or act as though Janie is a “bad speller” or that she “struggles with math.” Those words and beliefs are powerful, and not true based on an isolated test score with compromised validity.

I hope that this has helped you as you frame your own practices and beliefs related to educational testing.

Please comment below if you have any further questions or if you disagree with me. I’d love to talk!

© 2012-2023 PACE
All rights reserved