Monday, May 23, 2016

Assessment validity

Over the past few weeks I have been part of several interesting discussions on the topic of cultural validity in assessment. There is so much to talk about here but I will offer just a short few thoughts now and aim to follow up with greater insight as I construct my own understanding over next few weeks.

The experience that prompted me to write this post occurred yesterday at my school. I was helping out with the assessment of some prospective students that have applied to join our school. I was asked to conduct a reading diagnostic with a student who happened to come from China. The goal of this task is to ascertain a broad range of the student's literacy and language development. This information is useful to handover to classroom teachers so that the can develop a profile of their new students as efficiently as possible. This particular diagnostic hails from North America and, as such, contains several references to North American culture and life in general. I was asked whether this particular student demonstrated comprehension of the reading material at a D level. My response was that she could read this text with fluency and was able to answer the related comprehension. This was true, she could read it with comprehension, however I explained to her the context of the details. In the story a skunk causes an array of animals to run away. I know that this is because skunks use their odour as a defensive weapon against potential threats. The reason that I know this is that I've grown up with several cultural influences via the media and in print that have allowed me to establish this in my own mind. For a 6 year old girl from rural China this is not relevant or significant for her life. I don't believe that its fair to judge a child's ability to comprehend something on content that is completely foreign to them.

Benign as the intention may be, the data gained is not useful. Had I not explained the relevance of the skunk to the story, the student would not have got the question correct and could possibly have been labelled something that she is not. I wonder how often this happens - particularly in international schools. If we take diagnostic reading programs, for example. Although teachers may have an educated, research-based opinion on what an appropriate process is to learn how to read, if the measure that is being used to assess achievement within the process is faulty then the process itself is invalid.

To quote Guillermo Solano-Flores (pg. 3):

"Tests are cultural artifacts. They are part of a complex set of  culturally established instructional and accountability practices; they are created with the intent to meet certain social needs or to comply with the mandates and legislation established in a society; they are written in the language used by those who develop them; their content is a reflection of the skills, competencies, forms of knowledge, and communication styles valued by a society - or the influential groups of that society; and they assume among test-takes full familiarity with the contexts used to frame problems, the ways in which questions are worded, and the expected ways to answer those questions."

Assessing the Cultural Validity of Assessment Practices: An introduction. In: Cultural Validity in Assessment: Addressing Linguistic and Cultural Diversity. 2011. Routledge, NY, USA.

Photo Credit: Damian Gadal via Compfight cc