Analysis of Activity Monitors

Note: this analysis was scored using the criteria described here in late 2011/early 2012. Since then some device capabilities have changed, some devices are no longer sold.

Measuring activity has become relatively simple as the cost of 3D accelerometers has fallen and they have become incorporated into more devices. I evaluated a range of dedicated activity monitors as well as smartphone apps that claim to measure activity. The results for activity life-logging devices and apps are shown in Table 3. Since our research questions are considering multi-day life-logging, it is clear that the only appropriate devices in the list are the FitBit and the BodyMedia armband: the other devices and apps last less than a day since they require constant wireless communication with a smartphone. A number of apps do score higher on the charge/sync criterion but because they only calculate data while the phone screen in on their accuracy is so low as to be useless. Although the BodyMedia armband showed a higher score for accuracy over the FitBit due to its measurement of galvanic skin response (sweat), temperature and heat flux in addition to the accelerometer data (which is the only thing the FitBit records), this difference in calorie burn calculation was small. In order to determine the ‘ground truth’ a heart rate monitor and treadmill were used for calibration. The BodyMedia armband was very close to the values given by the treadmill/heart monitor, but the FitBit was less than 5% lower on step count and less than 10% lower on calorie count. Since the research questions are looking at non-specialist users (i.e. neither athletes nor those with a diagnosed medical problem), this small difference is not relevant. As the FitBit was the least invasive, had a better charge/sync rating and had the most open data format it was considered a good choice for activity life-logging. Smartphone based activity monitors proved inappropriate because they either relied on an inaccurate GPS reading (which does not work indoors) or measured activity only when the phone screen switched on and unlocked.

 

Name/URL Charge/ Sync Invasive -ness Accuracy Open Data
FitBit
www.fitbit.com

(original model evaluated, new models released in 2012)

5

-once/week 1 hour charge on stand, wireless sync

4

-clip to belt, bra, or pocket

4

-uses accelerometer to guess calories based on height and weight

 

4

– API allows access to most data

BodyMedia

 

£150 plus monthly susbscription

www.kiperformance.co.uk

 

2

-2-3 day capacity, cable charge, cable sync with custom software

1

-strapped tightly to upper arm, uncomfortable

5

-uses GSR, heat flux, temperature and accelerometer at 1 min intervals to calculate calories

1

-custom software to access data, subscription required to access data, no export available

Polar Heart Rate ANT+ Chest Strap

£60

2

1

5

4

Various iOS and Android activity monitor apps

£0-£5

5

1

1

4

Nike+ (shoe sensor)

£15

 

2

-shoe sensor lasts months but smartphone app must be running, lasts only a few hours

4

-strapped to shoe laces or inside show but must run smartphone app

4

-measures accurate step count, guesses calories based on height and weight

1

-proprietary software, no data export

Table 3: Activity Life-logging Device Analysis

Experiments with Personal Informatics Devices (Lifelogging) for self-hacking, persuasion, influence, nudge, and coercion (PINC)