I am writing a series about normed tests and drawing on my graduate school classes and my years of experience as an educator. When I was in graduate school, my psychological test construction classes were all in the Education Department of my university. Hence, the examples I learned about were data sets from educational testing.
So you wanna give a normed test…
So there you are, giving a test that is ”normed.” There are adorable little tables that reveal insights about the test takers. Let’s explore an example. Let’s say that you want to know about traveling preferences because they matter for something you’re doing. Scores can be from 0 to 45. The lower the score, the less someone enjoys traveling. The higher the score, the more they like traveling. The norms are divided into three groups: homebodies (0-11), meh about traveling (12-20), and loves travel (21-45). Each group has known characteristics. One of your respondents scored between 12 and 20, allowing you to make tailored comments about that respondent. While the norms were being developed, the norming sample was given other questionnaires so that many things were learned about each group. The travel preference test, as compared to answers from the other questionnaires, included questions about how much they spend on gardening. Homebodies spent the most in this imaginary scenario. I made up the scoring groups for this travel preference test.
See [1] for reference
What is a normed test?
A normed test is a tool that has been administered to defined groups of people so a systematic interpretation method can be created. The norms are the groupings of scores, with each group having descriptions of those groups, such as characteristics of the people who are in the group and possible interpretation information for each group. So you’d say based on this score, you can say that the person who took the test has mild depression and this means that… whatever it is that the test was trying to measure depression for. Maybe it means that they should be able to respond to therapy that’s given once a week, or something like that. Or if you’re testing for academic achievement, it might mean that the group the student falls into would be a good candidate for a particular curriculum.
Why give a normed test?
You give normed tests to find out about individuals so that you can predict other things about those individuals. That’s simple enough. However, there is a glaring problem that I kind of hinted at: the results and subsequent interpretations of the results may not be accurate.
Key Terms in Normal People English
Anchoring—where your guideposts are for your groups and how you define those groupings.
Central Tendencies—where the most people are
Invariance—testing looks the same no matter what population is being tested
Normal—what most people are doing
Norming—process of defining groupings for the different levels of scores
Population—a group of people that have something in common, like age, nationality, or whatever, but all the people in that group.
References for the series
Disclaimer:
The content of this blog post is for informational purposes only and does not constitute professional advice.
