iPhone Step-Tracking Inaccurate vs Pedometers: UBC Study


According to a recent study by the University of British Columbia, the iPhone’s built-in pedometer missed about 1,340 steps during a user’s typical day when compared to a dedicated wrist-worn accelerometer. “In order to make accurate conclusions, we as researchers need to know that the data is actually representative of real behaviour,” said Mark Duncan, the study’s lead author.


33 participants were involved in the UBC study, which was divided into two parts i.e. a laboratory test and a test in regular living conditions. In the lab test, participants carried two iPhones, a personal iPhone and a shared one provided by the lab, which would allow researchers to see if different phone models produced different results. Participants walked on a treadmill for 60 seconds at various speeds and their steps were counted manually. 

Personal iPhones underestimated steps by 9.4 per cent at the slowest speed of 2.5 km/h. The shared iPhone fared slightly better at 7.6 per cent. At faster walking speeds, the phones were off by less than five per cent, which is generally considered acceptable for a pedometer.

For the other part of the study, participants fixed accelerometers to their waists for a full day, and recorded step readings from the iPhone at the beginning and end of the day. Over three days, the iPhone underestimated the accelerometer data by an average of 21.5 per cent, or 1,340 steps per day.

“The accelerometer in the iPhone actually does a pretty good job when tested under lab conditions,” said senior author Guy Faulkner. “You just have to have it on you at all times.”

Duncan however stressed that an average person shouldn’t be discouraged from using health apps for motivation. He added that people who are already tracking their steps, they can rest assured that if their phone says they’re getting the recommended 10,000 steps in a day, they are probably getting at least that many.

The study was published last month in the Journal of Sports Sciences.


  • DrDca

    I notice that the “research grade” pedometer is strapped to the subject’s waist, while the iPhone is held in her (widely swinging) arm. This discrepancy alone could account for the differences. I wear my iPhone strapped to my waist, and by actual count of the steps (yes, I can count that high!) it was within 2% over 1200 steps on several tests!

  • fredf1

    DrDca: given that it is an ‘accelerometer’ one might think that carrying the phone in your ‘widely swinging arm’ would be even more likely to pick up walking data compared to a waist.
    So I think you may have it backwards.

  • Aleks Oniszczak

    I’d have more confidence if they had themselves videotaped and then had a few people count their actual number of steps. If the people counting the steps agreed – then that would be an excellent baseline to compare to. Having two gadgets give you different readings does not tell us which is wrong. Maybe they are both wrong, who knows.

  • Joe

    I wish they had tested this alongside the Apple Watch. It’s fine if the iPhone isn’t totally accurate, that’s not its primary job. But if the Apple Watch produces this kind of inaccuracy, then I want my money back!

  • Janker

    I’d have more confidence if your reading comprehension was better: “Participants walked on a treadmill for 60 seconds at various speeds and their steps were counted manually.”

  • Aleks Oniszczak

    Ah yes I did miss that. Well good on them – looks like a solid study. Now excuse me while I skip off and buy some reading glasses.