BME100 f2015:Group13 1030amL3: Difference between revisions
(→Graph) |
|||
(3 intermediate revisions by the same user not shown) | |||
Line 142: | Line 142: | ||
==Inferential Statistics== | ==Inferential Statistics== | ||
In order to validate the design of our product we needed to test it on many subjects. We tested our device on 50 subjects, picking from both genders and people of different physical size and stature. Our sample was random, but only included people between the ages of 20 and 50 which is our target population. In order to test the accuracy of our device, we had subject perform physical activity, light or heavy, over the course of an hour and recorded the number calories burned in 3 min intervals and found the total calories burned. The subjects simultaneously tested both our Uwatch and a brand from our competitors, the Nike Fuel Band to see how accurate our device’s measurements are. | In order to validate the design of our product we needed to test it on many subjects. We tested our device on 50 subjects, picking from both genders and people of different physical size and stature. Our sample was random, but only included people between the ages of 20 and 50 which is our target population. In order to test the accuracy of our device, we had subject perform physical activity, light or heavy, over the course of an hour and recorded the number calories burned in 3 min intervals and found the total calories burned. The subjects simultaneously tested both our Uwatch and a brand from our competitors, the Nike Fuel Band to see how accurate our device’s measurements are. | ||
[[Image:Data_Chart.jpg|125px|Calories Burned]] | |||
As for the data we ran a paired t-test to evaluate the statistical significance between the two sets of data, which were the U-watch and the Nike Fuel Band. The test is paired because only one group was tested using the two devices. The resulting '''p-value was 0.503552128'''. Because this value is over .05, there proves to be no statistical significance between the Uwatch and the Nike Fuel Band. | As for the data we ran a paired t-test to evaluate the statistical significance between the two sets of data, which were the U-watch and the Nike Fuel Band. The test is paired because only one group was tested using the two devices. The resulting '''p-value was 0.503552128'''. Because this value is over .05, there proves to be no statistical significance between the Uwatch and the Nike Fuel Band. | ||
The Pearson’s '''r-value of 0.999360395''' that was found for the data set shows a strong correlation between the U-watch measurement and that of competitors. | The Pearson’s '''r-value of 0.999360395''' that was found for the data set shows a strong correlation between the U-watch measurement and that of competitors. These two values mean that the measurement readings for calories burned that the Uwatch produces are accurate and can be trusted by consumers. | ||
Revision as of 17:45, 6 October 2015
BME 100 Fall 2015 | Home People Lab Write-Up 1 | Lab Write-Up 2 | Lab Write-Up 3 Lab Write-Up 4 | Lab Write-Up 5 | Lab Write-Up 6 Course Logistics For Instructors Photos Wiki Editing Help | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
OUR TEAMLAB 3A WRITE-UPDescriptive Statistics
ResultsHeart Rate Temperature
AnalysisWe decided to use a T-test because there were only two groups being compared in each situation: the spree headband and pulse ox for heart rate, and the spree headband and thermometer for body temperature. As the devices were all used on the same person during our data gathering, we used a paired T-test comparing the spree app results of heart rate and the pulse ox. Then we compared the spree app temperature results to the thermometer. We made two separate graphs, one for pulse and the other for body temperature. Our p values were 0.390 for our heart rate data and 4.696 E-77 for our temperature data. We also calculated Pearson's R values for the measurements, obtaining the following numbers: 0.823329451 for heart rate and 0.167425906 for body temperature. The purpose of the Pearson's R values are to show the correlation between the actual (pulse ox and thermometer) and theoretical (spree headband). In terms of heart rate, the measurements the spree headband and pulse ox made correlated fairly well (since R values close to 1 indicate high correlation). This is consistent with the fact that we obtained a p value above 0.05 (0.39), indicating that there is very likely no statistically significant difference between the measurements made by the two devices. Since the measurements made by the headband are so similar to those made the pulse ox, our gold standard device, they are likely accurate and reliable measurements. The Pearson's R value for body temperature showed that the measurements made by the spree headband and thermometer did not correlate much at all (since R values close to 0 indicate little to no correlation). This is consistent with the fact that we obtained a p value far below 0.05 (4.696 E-77), indicating that we are more than 95% confident that the differences in the measurements made by the two devices are statistically significant.
Summary/DiscussionThere were several design flaws with the product (the spree headband). For starters, the apparatus is not aesthetically pleasing, being a heavy metal headband that must be placed around the forehead. This also made it uncomfortable to wear for long periods of time. Making a device for a different part of the body, such as the wrist, would improve user comfort. Another flaw is the fact that the device did not measure temperature well at all. Instead of showing an actual numerical value for temperature, it would represent the temperature using a three color-based temperature meter. To make the measurements more accurate, the device would need to measure the temperature with an actual number. Also, the device didn’t measure the distance traveled or speed at a certain instant accurately at all. Both these could be helped by putting better sensors within the device. The low Bluetooth range was another problem because the user would go out of range very easily, decreasing the convenience and usability of the device. Lengthening the Bluetooth range would allow the user to set the device down and do their workout without losing the signal.
LAB 3B WRITE-UPTarget Population and NeedOur target population for our device is male and female adults, 20-50 years old, concerned or very intuitive with their health. We are basing this design off the needs of the consumer. These needs include going to the doctor less, and being able to treat small problems with their body by themselves at home.
Device Design
Inferential StatisticsIn order to validate the design of our product we needed to test it on many subjects. We tested our device on 50 subjects, picking from both genders and people of different physical size and stature. Our sample was random, but only included people between the ages of 20 and 50 which is our target population. In order to test the accuracy of our device, we had subject perform physical activity, light or heavy, over the course of an hour and recorded the number calories burned in 3 min intervals and found the total calories burned. The subjects simultaneously tested both our Uwatch and a brand from our competitors, the Nike Fuel Band to see how accurate our device’s measurements are. As for the data we ran a paired t-test to evaluate the statistical significance between the two sets of data, which were the U-watch and the Nike Fuel Band. The test is paired because only one group was tested using the two devices. The resulting p-value was 0.503552128. Because this value is over .05, there proves to be no statistical significance between the Uwatch and the Nike Fuel Band. The Pearson’s r-value of 0.999360395 that was found for the data set shows a strong correlation between the U-watch measurement and that of competitors. These two values mean that the measurement readings for calories burned that the Uwatch produces are accurate and can be trusted by consumers.
Graph
|