Scribd
Upload
35 views

02 Making Decision Data D2L

The document discusses making decisions with data including fitting lines to data, living with uncertainty in measurements, accuracy and precision of measurements, communicating experimental results, describing variation in data, relative standard deviation as a measure of precision, and preparing calibration curves using linear regression and correlation coefficients.

Uploaded by

Paige Burbadge
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
35 views

02 Making Decision Data D2L

The document discusses making decisions with data including fitting lines to data, living with uncertainty in measurements, accuracy and precision of measurements, communicating experimental results, describing variation in data, relative standard deviation as a measure of precision, and preparing calibration curves using linear regression and correlation coefficients.

Uploaded by

Paige Burbadge
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

MAKING DECISIONS WITH DATA

2011 Pearson Education, Inc.

Goal:
Fitting a Line to Data Or Line of Best Fit

2011 Pearson Education, Inc.

Living with Uncertainty


EVERY Measurement we make will have some uncertainty (error). Classify the following pipette errors:

Suppose the tip of a TD pipet has a small chip that changes the size of the opening. Repeated measurements with the pipet will always be off, but by the same amount and in the same direction. Systematic or Random?
A perfectly good TC pipet is used to deliver 20.0 mL. Systematic or Random? Which problem could be eliminated?
2011 Pearson Education, Inc.

Measurements can be accurate or precise


In the lab, the true value may be an accepted value (rather than a known value), m. ABSOLUTE ERROR: e = lx ml RELATIVE ERROR: er = (lx ml)/m This can be expressed as a fraction, percent, or ppm.

2011 Pearson Education, Inc.

Communicating Experimental Results


Suppose you did three trials to determine the mass of a drain cleaner. You could report the average value (but with a small number of samples we have the experimental mean X ).

2011 Pearson Education, Inc.

Describing variation within a set of data that, in theory, should not vary
Rather than just examine the simple range of values, we can compare each value to the experimental mean. This comparison of variation is called the standard deviation
Note, this is not the true standard deviation, s (requires more values to cancel all random errors).

2011 Pearson Education, Inc.

The Helpful Relative Standard Deviation


RSD = s/mean This can be expressed as a percent. RSD can also be known as the coefficient of variation. A smaller value indicates better precision. Practice: A student found the following values from three trials for the molarity of his solution: 0.01005 M; 0.01018 M; 0.00998 M What is the experimental mean? What is value of s? What is the value of RSD?
Mean = 0.01007 M s = 0.000101 M RSD = 0.0100 M

2011 Pearson Education, Inc.

Preparing a Calibration curve or comparing results to predicted response

Linear Regression: Fitting x and y values into a linear equation.

2011 Pearson Education, Inc.

Best fit (correlation)


Correlation Coefficient: (r) How well does the best fit line describe the data? When |r| = 1, the fit is perfect. Increases in r = greater confidence.

2011 Pearson Education, Inc.

You might also like