BME100 f2014:Group18 L2
BME 100 Fall 2014 | 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 2 WRITE-UPDescriptive Statistics
Results
Analysis
Summary/DiscussionTo begin our experiment we performed a ANOVA on our human study. The data we were provided with described the levels of dosages in mg and compared them to our inflammation levels in pg/ml. So first we found the average values for each dosage level (ie 0mg-15mg). We then found the standard deviation as well as the standard error. We then graphed our averages for each dosage vs the dosage levels, and we added in our error bars. Next we ran the actual ANOVA test on the human study. With this information we obtained our P-value, which was less than the targeted 0.05. From here we had to run a Bonferroni Correction, because the p-value provided in the ANOVA doesn't tell us what it is comparing. The Bonferonni Correction required us to perform a T-Test for each comparrison (ie 0mg vs 5mg, 0mg vs 10mg ect). From here we had to create a corrected p-value. This was defined as our original p-value divided by how many new t-tests we performed. Once we determined that our new p-value vs our t-test comparison values, all of which were less than our new p-value. This shows that our data is statistically significant. In another words, our data was good. From here we performed a T-test on the Rat Study. We did a t-test because we were only comparing 2 columns of data. It was unpaired because we were testing different subjects(rats in this case). So first, we had to find the averages for 0 pills and 1 pill and our standard deviations for each as well. Essentially we had to find all the same values as in the ANOVA (ie avg, std dev, endpoint, std error). Once again we made a graph comparing the LPS Dosages vs our Inflammotin. From here we ran a t-test on our data and obtained a p-value which we compared to the alpha value of 0.05. Our p-value was larger than this, so we concluded that our data was NOT statistically significant. Again, another words our data was no good. |