If you are in an AP psychology class right now, you will probably take a comprehensive exam at the end of the course. That test will give your teacher an idea of how much you learned in your AP class, but that’s not all it will do. Standardized tests like the AP psychology exam come with average scores to compare with the scores of your class. Your teacher will be able to compare your class scores with those of other AP psychology students!
- What are standardization and norms in psychology?
- What are norms in psychological testing?
- What is the process of standardization for psychological tests?
- What are some examples of standardization and norms in psychological testing?
- Why is standardization important for psychological tests?
Standardization and Norms in Psychology
Did you know that many psychology research studies include tests or assessments? They measure personality, mental disorders, intelligence, abilities, and interests. When a psychologist uses a test in a research study, the test must be reliable and valid. The test results need to be consistent and accurate.
There is another thing that a test must be, though, for a psychologist to use it in research: the test needs to be standardized.
A standardized test includes reliable methods, accurate content, administration instructions, and average population scores to compare with a sample’s scores.
Standardized tests include everything that a psychologist needs. If a test or treatment is standardized, it comes with detailed administration instructions. It also means that it went through lots of research and changes before it was standardized.
Standardization is the process of trying out a new test or treatment, getting feedback and making changes to improve it.
Some psychologists specialize in developing, trying out, and revising tests. They are called psychometricians. They study the field of psychometrics, which is all about how to measure things like intelligence, personality, or individual abilities. These things can be way harder to assess than you might think!
Fg. 1 Standardized Tests, pixabay.com
You are the captain of your sports team. You are responsible for overseeing tryouts and choosing new members for your team. You need to know which players are best for the team and what positions are best for each player. You come up with a series of drills and moves to assess the abilities of each person trying out. At the end of tryouts, you compare the performances of each person and choose the ones who played best to be part of the team.
Process of Standardization of Psychological Test
Standardizing a test or treatment can be a long process. There are two big goals of standardization: collecting the average scores of the population and deciding on test administration details. The population is all the people you want to take the test or go through the treatment. As you can imagine, this could be a big group of people.
Some examples of populations are all 3rd graders in America, all patients in psychiatric hospitals, all clients receiving individual therapy treatment for depression, all people over age 80 in a specific state, or all high-school AP psychology students.
The administration of a test just means the instructions that go with a test. They tell the teacher, therapist, or psychologist how to give the test, how much time to give the test-taker, how to grade the answers, what scores to expect, and what each score might mean. Providing information on what scores to expect requires knowing the scores of other people who have taken the same test. Psychometricians figure this out by getting hundreds or thousands of people to try the new test.
Fg. 2 Online test administration, pixabay.com
Steps in the Test Standardization Process |
| Step 1 | Choose what to measure and how to define it. |
| Step 2 | Decide who will be taking the test (the population). |
| Step 3 | Choose what questions to ask, how to word them, and how many to include. |
| Step 4 | Choose a response format for each question. |
| Step 5 | Get lots of people in your population to take the new test and collect feedback from them. |
| Step 6 | Make changes to the test based on the feedback. |
| Step 7 | Repeat steps 5 and 6 as much as needed. |
| Step 8 | Compare the test to other tests that measure similar or opposite concepts. |
| Step 9 | Perform statistical analyses during the entire process to make sure the test is reliable and valid. |
| Step 10 | Put together instructions for administration. |
Steps 1 and 2 are the most essential in the entire process. Standardizing a test is a lot like any other kind of research. You must know the goal or purpose of the test before you can get started. You also need a good operational definition for whatever you want to measure.
A test you find online is supposed to tell you if you are happy or not. The results the test gives you depend on how the test-takers define happiness. Maybe they believe happiness is about having many friends, so they ask questions about your relationships. Maybe they define happiness as having all your needs met and feeling relaxed, so they ask questions about those things instead. The operational definition determines the content of the test.
Step 4 is about how test-takers can respond to each question. Do they fill in a blank, choose from multiple answer options, rate their response on a scale, indicate either yes or no (true or false), or provide a whole paragraph? The best response format for the test depends on what kind of data the researchers are collecting and what kind of questions they are asking.
A question on a test asks, “On a scale of 1 to 10, how much did you enjoy your experience today?” Below the question, there are 10 dots, with the words “not at all” next to the first dot and “very much” next to the last dot. You draw a checkmark on the 8th dot because you had a good time, but you feel the organization could improve part of the tour.
Imagine taking a test, and the first question asks, “what did you have for lunch today?” The response options are “true” or “false.” Something is wrong here!
Steps 5 and 6 can take a very long time. The creators of the test may repeat these steps many times. Trying out the test on the population may reveal that the questions on the test are all too hard or too easy. It may indicate that the questions are too similar or worded weirdly. It could even tell the researchers that they developed a good test, but it doesn’t measure what they want to measure!
All this feedback from getting the population to try out the test helps the psychometricians revise and edit the test. Sometimes, the researchers throw the test out and start over from scratch. Steps 8 and 9 happen throughout the entire process. Step 8 helps the developers determine what questions to ask, and step 9 helps them decide whether they need more, less, or different questions.
A test designed to measure anxiety is compared to another test for anxiety and to a test that measures relaxation. The scores on the two anxiety tests should be similar, but the scores from the anxiety and relaxation tests should be the opposite of each other. If someone scores high on anxiety, they should also score low on relaxation. If they score high on relaxation, they should score low on anxiety.
Step 10 may not seem like much, but it is crucial. Test administration manuals tell researchers where to give the test, how to give it (on paper or online), how much time it should take, how to grade it, and how to understand the results. You take a test online, and you get a score of 35. What does that mean? Is it a good or bad score? How does it compare to other people who have taken the same test?
Examples of Standardization and Norms in Psychological Testing
A test designed to determine the most popular ice cream flavor in America went through the standardization process. The option most often selected by those who tried the test was chocolate. If the people who tried out the test accurately represent the population (all Americans), we can say that the population norm for the best ice cream flavor is chocolate.
You and your best friend decide to take the test. Your best friend selects vanilla. You search the list of options for your favorite flavor, but it isn't on the test. You realize that there are only 3 response options on the test: strawberry, vanilla, and chocolate. Wait a minute: there are way more than 3 ice cream flavors! Is this test accurate?
What if the same test listed every ice cream flavor in existence. What would your response mean? If you select chocolate, you know that your favorite flavor is the same as most people in America. If you choose anything else, you know your favorite flavor is a less popular option. This is a simple example, but it illustrates how a population norm helps researchers understand a test result.
A population norm is an estimated average score for a member of a specific population. It includes what scores are considered extremely high or low among that population.
Norms in Psychological Testing
In general, there are 4 types of population norms in psychology. These are age norms, grade norms, percentile norms, and standard score norms. Age norms are used in intelligence testing and measuring height and weight in children. Grade norms are mostly used in educational settings. Grade-level exams like SATs compare student results to others in the same grade and other grades.
Percentile scores are more nuanced. They tell you where your score falls in a large set of scores. If you score at the 50th percentile, half of the people who took the exam scored the same as you. If you score at the 75th percentile, 75% of the people who took the exam scored below your score. If you score at the 25th percentile, 25% of the people who took the exam scored below your score. You can also figure out what percentage of test-takers scored above you based on your percentile score.
Standard scores are even more nuanced than percentile scores. They tell you how close you scored to the mean of all the scores. If your standard score is higher than 0, it means you scored above the mean. If it is a negative number (below 0), it means you scored below the mean. In psychology, these are usually called z-scores.
StudySmarter has a whole explanation about z-scores!
Importance of Standardization in Psychological Testing
We make decisions based on tests every day. Teachers plan what material to teach, doctors decide if a child is growing properly, and therapists try to determine if a treatment method is working. Colleges decide which new students to admit. These decisions are all based partly on the results of tests. If the decision is based on a standardized test, psychologists would call that an evidence-based decision.
Standardized tests and treatments are essential scientific tools. The long, complicated standardization process helps researchers ensure the tests they use are accurate and give consistent results. The process also tells the test developers what changes to make to the test. Once a test is standardized, it continues to go through the standardization process for every new population, language, format, and culture.
Fg. 3 College entrance exams, pixabay.com
Standardization and Norms - Key takeaways
- A standardized test includes reliable methods, accurate content, administration instructions, and average population scores to compare with a sample’s scores.
- Standardization is the process of trying out a new test or treatment, getting feedback, and making changes to improve it.
- A population norm is an estimated average score for a member of a specific population. It includes what scores are considered extremely high or low among that population.
- You need a good operational definition for whatever you want to measure on the test.
- There are two big goals of standardization: collecting the average scores of the population and deciding on test administration details.
Explanations 
Exams 
Magazine 
