Is there an equivalent scale in UX as the Cooper-Harper Scale or the Cooper-Harper Ratings are in Handling Qualities Engineering in Aerospace? The Cooper-Harper scale associates a quantitative value to the ability and ease of completing a task in prefer and acceptable ranges such as time to complete task or the precision of the task as well as how easy was it to meet or exceed those criteria?
-
There may be such a scale, but I think you'll find that unlike in engineering (and further more in aerospace engineering) there isn't really a standardized process for completing specific tasks, and so the utility of creating a scale like this is going to create as much problem as it helps to solve.– Michael Lai ♦Mar 26, 2016 at 9:04
-
@MichaelLai, usability testing is often about completing a specific task and it isn't uncommon to have a single (or limited number of) way to complete it. Cooper Harper helps to measure the workload of that task completion. Are you referring to a different process where Cooper Harper is less appropriate?– Nicholas PappasMar 27, 2016 at 0:30
-
@EvilClosetMonkey Yes, I think there are many interactions on websites (especially social networking types) that don't necessarily have a specific start and end point (plus the interactions/flows in-between). So the use of the Cooper-Harper scale or its equivalent for UX would most likely be limited to those linear or defined tasks/use cases. In which case you are better off looking at completion rate analytics anyway.– Michael Lai ♦Mar 28, 2016 at 21:40
-
1@MichaelLai, good points about non-linear tasks. Cooper-Harper doesn't care about how you get from point-A to point-B, it is simply a subjective assessment of how difficult it was to get there. In combination with observational data, it is still a very good scale on workload. Certainly, if your testing doesn't reveal how the user got from A-to-B the data is near useless -- even for a linear task, you have to see where the user goes off the tracks. It would actually be fascinating to see a non-linear task take more steps than designed and still be rated with a low workload.– Nicholas PappasMar 28, 2016 at 22:02
Add a comment
|
2 Answers
Yes, there are modified Cooper-Harper Scales that measure operator workload. It involves some basic tweaking from "pilot" to "operator".
Here is one such representation of the Modified Cooper-Harper Subjective scale:
I used these modified scales many times to measure acceptable operator workload for multiple defense department and aerospace programs, which didn't target a pilot.
One of the advantages to this simple modification is that many are already familiar with the source material. Especially if you are dealing with aerospace or defense department programs. The ability to demonstrate workload on a scale that is already familiar makes telling your story easier.
answered Mar 26, 2016 at 15:33
-
Where you have applied the Modified Cooper-Harper Subjective scale, have they been for linear tasks? How does it compare to usual forms of usability testing using analytics and completion rate?– Michael Lai ♦Mar 28, 2016 at 22:40
-
1Cooper-Harper isn't something done instead of other analytic tools, it's an option to add to the overall suite. When combined with completion rate, for example - not being able to complete the task is an automatic 10 on Cooper-Harper, but completing the task can be 1-to-9 on the scale. Cooper-Harper is an option within any "usual form" of usability testing. Mar 29, 2016 at 15:33
-
1As to linear tasks - are you referring to tasks that have one, and only one, path? While that would be rare, Cooper-Harper doesn't care. It's a measure of perceived workload - so one user may take 10 steps to complete and score the process a 8; while another might take 20 steps and score it a 3. By having observational techniques in place you can review why Person-A thought it was so hard while Person-B thought it was pretty straight forward, but took twice as many steps. It may have been they took different paths and Person-B just found their path very intuitive. Mar 29, 2016 at 15:38
-
I've applied it in situations where the available paths were singular, limited, and robust. Mar 29, 2016 at 15:40
-
I take the term linear to mean that there is at least a defined start and/or defined end point. It might vary depending on the individual or task where you meander within the system, but the end point is the same. I use the term non-linear to mean that you can start at more than one point and finish at more than one point.– Michael Lai ♦Mar 29, 2016 at 21:27
If you want to measure the cognitive workload the most famous scale is NASA-TLX scale. SWAT is other one which is widely used. However they are multidimensional which means mental workload is calculated based on more than one scale, unlike Cooper-Harper.
If you want to measure the usability you can use SUS scale, which is the widely approved and used for usability evaluation.
answered Mar 29, 2016 at 7:46
-
1+1 I found NASA-TLX and SWAT both very good scales, so thanks for contributing to the answer. Have you used it much in past projects? SWAT seems a little outdated and simplified versions have been proposed, maybe you would like to update your answer as well?– Michael Lai ♦Mar 29, 2016 at 21:49
-
1Yes I've Used NASA-TLX and it gives valid and comprehensive results. There is a calculation stage which is OK to be ommited. Actually, in cognitive psychology field this is the most used and referenced scale for measuring mental workload. So it is a safe bet as its validity and diagnosticity has been tested by numberious researchers. Mar 30, 2016 at 10:34