0

Overview:

  • For a study, two independent groups evaluated a system using two different usability measures (SUS and PSSUQ).
  • The PSSUQ collects slightly better scores when translated into the range 0-100 and directly compared to the SUS.
  • Also, statistical difference for the two groups based on SUS is found. However, the same is not the case for PSSUQ

Question:

As they are proven to correlate, I expected them to behave similarly. How can we interpret the difference in results? Is the sample size (about 30) just to small? Why is it that results are significantly different looking at SUS but for PSSUQ not? What could be possible reasons?

Improve this question
3
  • 1
    Just to confirm, how controlled was the test? Time of the day, background and demographics of participants? Are you suggesting that results from PSSUQ were correct and SUS not?
    – Ren
    Commented Aug 21 at 6:45
  • SUS and PSSUQ were filled out by the same participant in that order right after each other. I am not suggesting anything just wondering how to interpret these results. E.g. are both invalid now because they differ? Or could the difference still be significant and maybe sample size was too small for PSSUQ to detect?
    – anjelomerte
    Commented Aug 21 at 14:04
  • One of the ways will be to identify the outliers in the data sample and have a conversation with the corresponding participants about their replies to get more context as well as verify which method proved more reliable.
    – Ren
    Commented Aug 22 at 14:25

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.