One of the most important and also one of the most basic concepts beginning science students need to learn is often overlooked. It is that the results of any scientific experiment always have a degree of uncertainty. Nature never lies, but the scientists' ability to conduct the experiment, measure, and interpret the results is predictably imperfect. How many times have you conducted an experiment and felt lucky to have gotten 60% yield? Almost anything can cause error in experimental results: a concentration that is inexact, a scale that only displays to the 1/100 of a gram, a shaky hand or the limits of hand-eye coordination in a timing experiment. Students understand the idea of error in experimental data on a basic level even from their first lab, but they tend to see it as the fault of their inexperience-and don't think it through enough to generalize the concept to include all scientific processes. It is generalization and bringing the concept to the conscious mind that needs to be taught.
Students should be cautioned that even the published results that are discussed in the news should be taken with a healthy dose of scientific skepticism. How were these experiments done? What was the level of uncertainty? The truth is that most of the students taking chemistry in high school will not continue a career in that area, but as American taxpayers and citizens nevertheless will be expected to make judgments about science. Commercials and websites claim brilliant scientific advances that will solve the customer's every problem; the government uses our tax dollars to fund research; arguments about genetic testing and new drugs are discussed in the news. Students need to be able to decipher and make sense of the science around them in order to be an informed citizen and to make wise decisions about their own life. Knowing to ask questions regarding scientific or pseudoscientific claims is a skill they can use to that end throughout their lives.
This website can be used to convey the concept of the extent of variation in even advanced scientific research. You will notice that many of the molecules in the database have multiple entries. Each of these entries has separate data for the same molecule, from research conducted by different groups using different methods or different crystals. Comparison of structures for the same species will push students to discover that scientists can come up with quite different results legitimately even when they are studying the same molecule using the same method. Much of the variability of the X-ray data comes from "lattice effects" interacting with the molecule or ion in different ways from different directions. An additional source of variation peculiar to X-ray studies relates to hydrogen atoms. One should not take the hydrogen atom positions in structures derived from X-ray analysis too seriously. This is because there aren't enough electrons around hydrogen atoms to give much scattering of the X-rays. In fact, many times the hydrogen positions in reported X-ray structures really weren't determined and are just "faked" by the analysis software. The best hydrogen position measurements are done using neutron diffraction, microwave spectroscopy, and electron diffraction.