BME100 f2016:Group1 W1030AM L3
|BME 100 Fall 2016|| Home |
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
Wiki Editing Help
LAB 3 WRITE-UP
Inferential Statistics and Analysis
T-Test (two tailed, paired): 0.47754541
The value of the Pearson's r correlation coefficient of 0.401 is closer to 0 than to 1, which leads to the assumption that there is about 40% of similarity between the data sets. This is not enough the prove the reliability of the measurements of the Spree headband compared to the ones of the Golden Standard. The result of the t-test is 0.477.
The very low value of cc. 0.037 compared to the ideal value of 1.00 for the Pearson's r coefficient shows that there is almost no linear relationship between the variables, thus between the two data sets. As there is no assumed association between the measurements of the Golden Standard and the Spree headbands, the conclusions driven from the t-test therefore may well be invalidated. The result of the t-rest is 1.097.
Design Flaws and Recommendations
The measured values of body temperate and heart rate of the two devices differed in many ways. First of all, the values of the Spree headband were rounded to integers, being much less accurate compared to the ones of the Golden Standard. It can be assumed, that the sensor of the Spree headband isn't at the best place in order to measure heart rate and temperature. The missing data might be the signs of the sensor having fallen out. The presence of sweat, hair, make-up might also affect the work of the headband.
To provide more accurate measurements, the Spree device should be stable and at the best possible place on the body, thus close to some blood vessel close to the skin surface in order to measure more reliably. The use of a more advanced technology doing the measurements is also a considerable advice, so that Spree can cope with its competitors on the market. Other recommendations include making the device waterproof, placing it onto a place on the body where less visible, more comfortable.
Experimental Design of Own Device
The first phase of the experiment would be geared toward testing how the machinery is working. There would be 200 random volunteers involved in this phase's testing. Each subject would be given a pill box with "sugar" pills in each pill dispenser box to take home with them for one week. Before they take it home they would be taught what to expect from the device and how it works as well as test the bio-fingerprint scan. Everyday, for the one week period the subjects would be contacted and asked specific yes or no questions in order to ensure the machinery is working well. This is an example of a possible survey: "Were any of the pills damaged?" "Is the alarm at a proper volume?" "Did the alarm remind you on time?" "Were the pills accessible on time?"
Phase two of the experiment would be conducted in a similar way. The main objective would be to gauge the overall customer satisfaction. Again there would be 200 random subjects, however 100 of these subjects would be random prescription pill users and 100 would be random opioid users. Each of these subjects in phase two will be of varying ages from young (18-30), middle aged (30-55), and elderly (50+). Again the bio-fingerprint scan would be tested and each subject would be taught how to use the device as well as what to expect from it. They would take it home for a week and everyday they would be contacted in order to answer another set of specific yes or no questions, asking about the convenience the device offers. An example of a possible survey is: "Is it easy to use?" "Were any of the pills damaged?" "Do you find it helpful?" "Do you feel there is a danger of somebody else gaining access to the pills?"
Answers will be recorded and the results of these surveys will provide the data necessary to know whether or not the device is convenient and works properly.