Data analysis 40 pts
1. Develop a table of the statistical value for the true mean with 95% confidence
interval for each plateau both upward and downward. (2 Tables)
2. Develop a ±3σ range of expected measurements for each plateau up and down. (2
Tables)
3. Develop a statistical value for the true mean atmospheric pressure with 95%
confidence interval. Compare this value to the barometer reading in the thermo
lab.
4. Develop a statistical value for the true mean of the pressure induced by the weight
of the piston.
5. Develop a linear regression analysis of the entire data population for voltage (y-
axis) vs. applied pressure (x-axis) (linear curve fit to data)
6. Develop upper and lower limit lines based on a 95% confidence interval for both
slope and offset.
7. Using our calibration device and linear regression curve, compare the values of
pressure obtained with theoretically applying loads of .5-5 kg in .5 kg increments
to the calibration curve obtained from Omega.
8. Calculate the %error of the derived curve to the Omega curve as follows:
Graphs 40 pts
1. Plot all the raw data both upward and downward on a zero error plot for measured
voltage vs. applied pressure (data points only).
2. Show the upper and lower range of expected value curves based on the linear
regression analysis. (plot upper and lower offset curves)
3. Plot the expected voltage from both the linear regression curve develop in the lab
with the theoretical curve from Omega using the data from #7 above.
4. Plot the % error obtained from the comparison of experimental to Omega data.
Concluding Questions 20pts
1. Did you observe differences in the error depending on the direction the weights
were applied? If so why?
2. How well did your calibration curve compare to the one supplied by Omega?
3. Characterize the errors in comparing the two curves such as bias, accuracy,
precision etc.
4. What was the difference in measured pressure due to the weight of the piston?
5. What type of error does this introduce into the pressure measurement.