The correlation coefficient (Pearson's r) between the oral thermometer and the Spree Band was calculated to be 0.192798 (p-value of 1.0967E-21). On the other hand, the correlation coefficient between the pulse ox and the Spree Band was 0.690806 (p-value of 0.42712).
Inferential Stats
An unpaired T-test calculated the statistical significance of the recorded temperature and heart rate of the Spree Band compared to a golden standard. This T-test produces a P-value that indicates the likeliness that the null hypothesis is true. The null hypothesis states that the compared values of two populations are not statistically significant-meaning that the difference in value between two populations isn’t definite. Usually, we can reject the null hypothesis if the P-value lies below .05 as it shows that there is a 5% chat the differences are due to chance, rather than an outside factor.
Since the experiment draws its data from two population that are not related in the sense that it records the difference of something after a process, but rather the accuracy of which two devices record temperature, the T-test must be unpaired.
When the computer ran the numbers between the recorded temperatures of the oral thermometer and the spree band into a t-test, the program outputted the P-value was 3.3•10^(-20), which validates the claim that the Spree Band and the oral thermometer have recorded distinctly different values that are due to an external factor rather than pure chance.
Design Flaws and Recommendations
The purpose of this lab was to compare a new device which monitors body temperature and heart rate. The new device is from Spree sports, whose claim is the device accurately measures body temperature and heart rate. The new method was tested against the gold standard, an oral thermometer for temperature and a pulse ox for heart rate. The average temperature recorded using the gold standard was 96.64716º, while the average temperature for the Spree was 95.53086º. The standard deviation notes the average distance each data point is from the mean. The standard deviation calculated for the temperature of the gold standard is 1.922672º, while the standard deviation calculated for the temperature of the Spree 0.870378º. A smaller standard deviation signifies that the data points are close to the mean on average. Because the Spree has a smaller standard deviation value, the data collected is fairly close to the mean (95.53086º). The average heart rate using the gold standard was 98.0898 beats per minute (bpm), while the average heart rate for the spree was 95.9538 bpm. The standard deviation calculated using the pulse ox is 23.0305 bpm, while the standard deviation value for the spree band is 24.8775 bpm. Because the standard deviation values calculated were high in number, it can be concluded that the data points collected are spread from the average.
Pearson’s r value reveals the correlation between the two groups. The r value calculated for temperature is 0.192798, while the r value calculated for heart rate is 0.690806. An r value greater than zero signifies that there is a positive correlation between the two groups. A negative r value signifies that there is a negative correlation between the variables, and an r value equaling zero means there is no correlation. The r values calculated for this experiment were positive, resulting in a positive correlation between the Spree band and the gold standard.
The Spree band and pulse ox, which is the gold standard for measuring heart rate, are correlated. The Pearson’s r value calculated for the correlation between the two methods and heart rate is 0.690806, proving that there is a positive correlation. P-values are calculated probabilities. A p-value less than 0.05 rejects the null hypothesis, while a p-value greater than 0.05 accepts the null hypothesis. The p-value calculated for heart rate is 0.42712, meaning that the null hypothesis was rejected. There is not a significant difference between the Spree band and the the gold standard. The Spree method accurately measures heart rate. However, there is not a relationship between the Spree band and oral thermometer when measuring temperature. The Pearson’s r-value calculated is 0.192798. This value is significantly smaller than the value calculated for heart rate monitoring. The p-value calculated for temperature is 1.0967E-21. Because this value is very small (less than 0.05), the null hypothesis is accepted, meaning there is no correlation between the Spree band and gold standard.
Experimental Design of Own Device
Question:
Will the transdermal insulin patch, including a glucometer, more accurately deliver insulin to diabetic patients than the standard needle prick along with self-hormone delivery?
Hypothesis:
The transdermal insulin patch will constantly monitor the glucose concentration in the blood, and deliver the hormone more efficiently than the gold standard.
Groups:
Gold standard: needle prick along with self delivery
Transdermal insulin patch users
Experiment:
A large population size is required (100 patients per group, 50% being type 1 diabetic, 50% being type 2 diabetic, totaling 200 patients). To decrease any influencing factors, patients will be allotted one day’s worth of meals, for two consecutive days, equalling six meals (three a day). The first day is to control the diet of the patient since that is a major factor contributing to diabetes. The second day is to control the measurements taken, since day two is trial day. Glucose concentration will be recorded 16 times a day (hourly) from the time the patient wakes up, to the time the patient goes to bed, allowing for 8 hours of sleep. If the glucose concentration is at a concerning level, group 1 will self deliver insulin, while group 2 will automatically have insulin delivered through the bloodstream by the transdermal patch. Every 2 hours, the patient will record their energy level on a scale of 1-10 which will be provided:
This leads to 8 energy level recordings, which will then be averaged and compared between the two groups. The purpose of waiting two hours for energy recordings is to allow the insulin to be absorbed by the blood and perform its duty, take glucose from the blood. With a balanced diet, the glucose concentration of the subject will be controlled and easily managed.
Inferential Statistics:
Inferential statistics are helpful in generalizing the information collected from an experiment. The statistics that will be calculated from this experiment are:
The average (mean) of the two groups will be calculated. Calculating the average between the two groups will allow analyzers to clearly see the difference between the two groups.
A correlation coefficient will also be useful to calculate because it shows the direct relationship between the two groups.
In this experiment, glucose concentration and energy levels are compared between two groups. The control group considered is the gold standard (needle prick with self delivery), which is then compared with the new method (transdermal insulin patch).
Sources:Welbourne, T. M. (2015, July 23). How the Spirit of Giving Can Help Companies Reap Higher Performance. Retrieved September 20, 2017, from https://www.tlnt.com/how-your-workers-spirit-of-giving-helps-to-reap-higher-performance/