Cornell University Ergonomics Web
The following Cornell Musculoskeletal Discomfort Questionnaires (CMDQ) have been developed by Dr. Alan Hedge and ergonomics graduate students at Cornell University. The questionnaires are based on previous published research studies of musculoskeletal discomfort among office workers. Scoring of the questionnaires should be self-evident to anyone familiar with this type of research. If you need further assistance please contact Alan Hedge at email@example.com.
The following questionnaire files contain graphics and therefore have been organized as downloadable '.pdf' files, and these can be viewed and printed with Adobe Acrobat.
If you use any of these questionnaires in your research we ask that you acknowledge the Human Factors and Ergonomics Laboratory at Cornell University as the source of the instruments.
Scoring Guidelines - Cornell Musculoskeletal and Hand Discomfort Questionnaires
These questionnaires are for research screening purposes and not for diagnostic purposes. Scores can be analyzed in 4 ways:
In the computational analyses missing values can be coded as 0. If the missing value is for the frequency score then use this as a zero in multiplying, i.e. all combinations of Frequency, Discomfort and Interference become 0. However, if the missing value is in the Discomfort or Frequency score then treat it as missing so that the multiplied score will be at least the value of the Frequency score.
The individual items should also be analyzed to determine where there may be a postural problem for the person.
1) What is the importance or benefit of multiplying frequency score by discomfort score by the interference score?
This is just a way of spreading the scores so that you can more easily find the most severe cases. For example, if you have someone who has right shoulder pain every day (score of 5), and this is very uncomfortable (score of 3) and it substantially interferes with their work (score of 3), if we added their score it would only be 5+3+3 = 11, but if we multiply their score for the right shoulder it is 5x3x3 = 45. So let’s compare this to someone who says they had right shoulder pain 3-4 times in the last week (score of 3.5) that is was moderately uncomfortable (score of 2) and that it slightly interfered with their work (score of 2), then if we added their score it would be 3.5+2+2 = 7.5, which doesn’t seem that much different from 11 for the previous person, but if we multiple their scores for the right shoulder it is 3.5x2x2 = 14, which is almost 1/3 of that for the previous. So by multiplying out the scores it really stretches the scales and lets us more easily see those people with the greatest problems.
2) How can I classify the severity of discomfort into mild, moderate, or more.
The discomfort score is a severity of discomfort. Originally we had it as ‘mild’, ‘moderate’ and ‘severe discomfort’, but we found that people more easily understood the categories ‘slightly’, ‘moderately’ and ‘very uncomfortable’ . if you like you can replace the scale point titles with ‘mild’, ‘moderate’ and ‘severe discomfort’.
Validity and Reliability
The survey is derived from previous postural discomfort surveys and has high face validity. The survey is a screening tool and not a diagnostic instrument. If you are using it in a research study in any other way you should undertake your own test-retest reliability study. Given the nature of musculoskeletal discomfort (i.e. mostly intermittently experienced) differences in responses can be expected over longer test-retest intervals. The diagnostic validity of this survey can be tested in any study be comparing survey responses with clinical reports. The validity of the CMDQ has not been formally tested in the US but it has been extensively tested by Dr. Oguzhan Erdinc in Turkey with good results (see validity testing and evaluation report).
This survey questionnaire was developed and used in the following research study:
Hedge, A., Morimoto, S. And McCrobie, D. (1999) Effects of keyboard tray geometry on upper body posture and comfort, Ergonomics, 42 (10), 1333-1349.