Scientific research and experiments

Scientific research is the process by which multiple, independent people continuously test scientific theories by making new observations. One purpose of research is to find new phenomena that existing theories fail to correctly explain. Unexplained phenomena are often found in new areas of science, such as cosmology or quantum mechanics. Unexplained phenomena are also found when new technologies, such as particle accelerators, make different kinds of observations possible that could not be made before. Science grows as new observations lead to new hypotheses that augment or change existing scientific theories. Read the text aloud Show The evolution of science
Scientific method
Show More about the scientific method
One outcome of research is to produce scientific evidence, such as drawings, measurements, data tables, graphs, or observations. The two most important characteristics of scientific evidence are that it be objective and repeatable. Objectivity means the evidence should describe only what actually happens as exactly as possible, without opinion, interpretation, exaggeration, embellishment, or bias. Reproducibility means that others who repeat the same experiment or make the same observation in the same way always observe the same results. Read the text aloud
An experiment is a controlled situation designed to collect scientific evidence on what happens under a controlled set of circumstances. A well-designed experiment changes one variable at a time so that any observed effects may be clearly associated with the variable that was changed. A poorly designed experiment is at best inconclusive, and at worst it supports erroneous conclusions. Experimental evidence should be critiqued by asking the following:
  1. Is the experiment objective? Does it generate unbiased observations?
  2. Do any observed effects result from changing the variable claimed, or could other variables have caused the effect?
  3. Do other researchers repeating the experiment observe the same result?
  4. Have the data been analyzed to understand the uncertainties in measurement?
  5. Are the observed effects greater than the uncertainties in measurement for this experiment? If not, the experiment may be inconclusive.
Read the text aloud
Well-designed experiments typically include a control, which is a standard set of conditions that are used for comparison to check or verify a result. For example, when testing the effects of a medicine on a sample of patients, the results might be compared to a control sample of patients who did not receive the medicine. Read the text aloud Show Galileo's evidence from his telescope

Clark steps outside on a winter North Dakota morning and throws a rock into the lake outside his tent. He sees that the rock does not sink. Clark forms a hypothesis about the behavior of rocks and water:

The colder water is, the slower a rock will sink.

Design an experiment to test this hypothesis. Predict whether this experiment will prove or disprove Clark’s hypothesis. If it would be disproved, what would be the revised hypothesis?


12Previous Page Next Page