- Joined
- Mar 26, 2019
- Messages
- 5
- Reaction score
- 1
This was provided by Altius in response to a question by a pre-health advisor. It's more detailed than the FAQ on the Altius website.
How is CASPer scored?
Altus has a trained team of external raters who rate all of our Casper tests. We strive for a diverse population of raters in terms of age, ethnicity, gender, education and professional background. Our raters are located in Canada, the United States, and Australia. The raters score the test native to their country, as tests are specifically curated to each country in terms of language and culture. The Casper test is built on 10 different constructs. Casper is not a pass or fail test, but rather a way of ranking a large number of applicants on personal characteristics and professionalism. All questions are designed to be open-ended to assess if the applicant possesses any or all of the 10 constructs. A single question may lead to a response that demonstrates empathy, professionalism, and self-awareness. Due to the nature of the questions, there is no single right answer. As long as an applicant articulates the motivation for their response and displays the constructs that are probed for, they have the opportunity to perform well.
Every section is rated by a different person to create less bias, and to have more diverse perspectives. Additionally, raters are blinded to the applicant and cannot see any personal information or defining traits about the applicant (race, gender, age, etc.). Each section of an applicant's test is scored by a unique rater. Raters assign a score to each of the 12 sections, based on all three questions, on a Likert-type 1-9 scale, scoring them relative to all other applicants. Once every question has been graded, we take the mean of the 12 responses and come to a Casper raw score. We then convert this raw score into a z-score. Z-scores transform the data to have a mean of 0 and a standard deviation of 1. In other words, they are a measure of how different an applicant is from the average.
How is CASPer scored?
Altus has a trained team of external raters who rate all of our Casper tests. We strive for a diverse population of raters in terms of age, ethnicity, gender, education and professional background. Our raters are located in Canada, the United States, and Australia. The raters score the test native to their country, as tests are specifically curated to each country in terms of language and culture. The Casper test is built on 10 different constructs. Casper is not a pass or fail test, but rather a way of ranking a large number of applicants on personal characteristics and professionalism. All questions are designed to be open-ended to assess if the applicant possesses any or all of the 10 constructs. A single question may lead to a response that demonstrates empathy, professionalism, and self-awareness. Due to the nature of the questions, there is no single right answer. As long as an applicant articulates the motivation for their response and displays the constructs that are probed for, they have the opportunity to perform well.
Every section is rated by a different person to create less bias, and to have more diverse perspectives. Additionally, raters are blinded to the applicant and cannot see any personal information or defining traits about the applicant (race, gender, age, etc.). Each section of an applicant's test is scored by a unique rater. Raters assign a score to each of the 12 sections, based on all three questions, on a Likert-type 1-9 scale, scoring them relative to all other applicants. Once every question has been graded, we take the mean of the 12 responses and come to a Casper raw score. We then convert this raw score into a z-score. Z-scores transform the data to have a mean of 0 and a standard deviation of 1. In other words, they are a measure of how different an applicant is from the average.