Over the past few weeks, I have been focusing on assessment measures that are used by both researchers and teachers to collect information. Part of this exploration focused on intelligence tests, which have always intrigued me while at the same time leaving me with an uneasy feeling. How can we judge a person’s intelligence based off of one test? How do we make sure that this test is an accurate measure of what the person knows, and will not provide a skewed view? When researchers are looking at the accuracy of these types of tests, they are looking at the reliability (aka consistency) and the validity (aka accuracy) of the test. In acting as a researcher, I focused on one type of intelligence test called the Stanford-Binet Intelligence Scales. Take a look at the image below for a quick summary of this test, along with some of the criteria that researchers look for in both validity and reliability: I was then tasked with coming up with my own idea for measuring intelligence, taking into account the issues of validity, reliability, and cultural bias in testing. If I were creating my own assessment, I actually would model it after a test like the Stanford-Binet Intelligence Scales because it already measures a wide variety of skills and it is shown to be both reliable and valid.
However, a common critique of this test and other like it is that they fail to address any type of student motivation or perspective. If I were developing a test to measure intelligence, I would want to include a scale that measures the participant’s mindset (fixed vs growth), motivation, attitudes towards learning – not school, but learning in general – and ability to consider multiple perspectives. In order to maintain validity, I would need to make sure that the test was measure factors that will determine success. This goes beyond problem solving and critical thinking skills. I would also want to take into account the participant’s background. Are they coming into the assessment with trauma? And if so, what impact does that have on their learning? What is their cultural background? How will that impact their understanding of the test questions? In order to make this test free of cultural bias, the questions would need to be culturally neutral. The test should also provide background information, as needed, so that the participant is able to fully understand the context of each question. To maintain reliability, the test would need to be given multiple times throughout a person’s life and produce results that are comparable each time. The test would also need a standard scoring system, especially for sections that can be considered qualitative, like mindset and attitudes. To ensure that the results are truly reflective of the participant, questions could show up several times throughout the test with the same answer choices. If the participant answered the same each time, we would know that the results were a more accurate reflection of the participant. Assessment methods are not perfect, and lots of work has to go into them before they are readily accepted by both researchers and educators. I think that if schools are using intelligence tests as a means to place students on a particular track, the tests need to be more robust and well-rounded than simple measuring quantitative data. What are your thoughts on intelligence tests, standardized tests, or assessments in general? I would love to hear your thoughts! Leave a comment below, and follow me on Twitter @haley_fraeyman Resources: McDevitt, T. & Ormrod, J. (2019). Child development and education (6th ed.). New York: Pearson.
0 Comments
Leave a Reply. |
AuthorI am a graduate student at Michigan State University and a first-year Kindergarten teacher. Archives
April 2021
Categories |