Colorado Learning Attitudes about Science Survey (CLASS)

Developed by W. K. Adams, K. K. Perkins, N. S. Podolefsky, M. Dubson, N. D. Finkelstein, and C. E. Wieman

Purpose To measure students’ self-reported beliefs about physics and their physics courses and how closely these beliefs about physics align with experts’ beliefs.
Format Pre/post, Multiple-choice, Agree/disagree
Duration 8-10 min
Focus Beliefs / Attitudes (epistemological beliefs)
Level Upper-level, Intermediate, Intro college, High school

Sample questions from the CLASS:

A significant problem in learning physics is being able to memorize all the information I need to know.

  Strongly Disagree   1  2   3   4   5   Strongly Agree

Knowledge in physics consists of many disconnected topics.

  Strongly Disagree   1  2   3   4   5   Strongly Agree


CLASS Implementation Guide

Everything you need to know about implementing the CLASS in your class.

W. Adams, K. Perkins, N. Podolefsky, M. Dubson, N. Finkelstein, and C. Wieman, New instrument for measuring student beliefs about physics and learning physics: The Colorado Learning Attitudes about Science Survey, Phys. Rev. ST Phys. Educ. Res. 2 (1), (2006).
more details
Gold Star Validation
This is the highest level of research validation, corresponding to all seven of the validation categories below.

Research Validation Summary

Based on Research Into:

  • Student thinking

Studied Using:

  • Student interviews
  • Expert review
  • Appropriate statistical analysis

Research Conducted:

  • At multiple institutions
  • By multiple research groups
  • Peer-reviewed publication

Questions from the MPEX and VASS were taken as the starting point for the CLASS and modified then tested in student interviews. Questions were further revised with expert interviews. The “expert” answer to each question was determined by 16 physicists with extensive teaching experience who agreed to the answers for nearly all questions. Categories were created using reduced-basis factor analysis, where raw statistical categories and categories predetermined by researchers were combined iteratively. The CLASS was given to thousands of students and those with more experience in physics, had more expert-like beliefs. The CLASS has high reliability. CLASS scores were also correlated with other measures of learning. The CLASS has been administered at over 20 institutions with over 9000 students enrolled in many different course levels taught with differing teaching methods. Results have been published in over 45 peer-reviewed publications.


PhysPort provides translations of assessments as a service to our users, but does not endorse the accuracy or validity of translations. Assessments validated for one language and culture may not be valid for other languages and cultures.

Language Translator(s)  
Arabic H. Alhadlaq, F. Alshaya, and S. Alabdulkareem
Chinese Lin Ding and Ping Zhang
Finnish Mervi Asikainen
French Vincent Sicotte
German Christian Kautz, Hanno Holzhüter, and Felix Lehmann (This version is designed for an intro engineering class. Questions that don't make sense in a German engineering context have been removed.)
Indonesian Mutmainna Kadir
Japanese Michi Ishimoto and Hideo Nitta
Portuguese Eduardo Gama and Marta F. Barroso, Federal University of Rio de Janeiro
Spanish Genaro Zavala, Hugo Alarcon, and Angeles Domínguez
Swedish Johan Henriksson
Turkish Derya Kaltakci

If you know of a translation that we don't have yet, or if you would like to translate this assessment, please contact us!

Download the CLASS answer key.

Score the CLASS on the PhysPort Data Explorer

With one click, you get a comprehensive analysis of your results. You can:

  • Examine your most recent results
  • Chart your progress over time
  • Breakdown any assessment by question or cluster
  • Compare between courses

Download the CLASS scoring tool.

Typical Results

In typical physics classes, students’ beliefs usually deteriorate or at best stay the same. There are a few types of interventions, including an explicit focus on model-building and/or developing expert-like beliefs that appear to lead to significant improvements in beliefs. Small courses and those for elementary education and non-science majors also result in improved beliefs. However, because the available data oversamples certain types of classes, it is unclear what leads to these improvements. This figure from Madsen et. al 2015 shows CLASS (n=9296) and MPEX (n=1316) pre- and post-test scores and shifts for a variety of teaching methods. The CLASS and MPEX are similar in the way they measure students' beliefs about physics and learning physics, so the scores for these tests have been combined.

The latest version of the CLASS for Physics, released in 2004, is version 3. There are also variations of the CLASS for chemistry, biology, astronomy and math, which are available at The German translation of the CLASS is missing questions that don't make sense in a German engineering context, so there has 6 fewer questions than the English CLASS.