1.6: Scientific experiments

Last updatedSave as PDFView Spots

The spots on this child’s tongue are an early sign of vitamin C deficiency, which is also called scurvy. This disorder, which can be fatal, is uncommon today because foods rich in vitamin C are readily available. They include tomatoes, peppers and citrus fruits such as oranges, lemons and limes. However, scurvy was a well-known problem on Navy ships in the eighteenth century. Scurvy was said to cause more deaths in the British fleet than French and Spanish guns. At the time, the cause of scurvy was unknown and vitamins had not yet been discovered. Anecdotal evidence suggests that eating citrus could cure scurvy. However, no one knew for sure until 1747, when a Scottish naval doctor named John Lind did an experiment to test the idea. Lind’s experiment was one of the first clinical experiments in the history of medicine. Figure\(\PageIndex1\): Scorbutic languageWhat is an experiment?

An experiment is a special type of scientific research that is conducted under controlled conditions. Like all research, an experiment generates evidence to test a hypothesis. But unlike some other types of research, an experiment involves manipulating a few factors in a system to see how it affects the outcome. Ideally, experiments also involve controlling for as many other factors as possible to isolate the cause of the experimental results.

An experiment usually tests how a particular variable is affected by some other specific variable. The affected variable is called the dependent variable or outcome variable. The variable that affects the dependent variable is called the independent variable. It is also called the manipulated variable because this is the variable that is manipulated by the researcher. Any other variable (control variable) that may also affect the dependent variable remains constant, so only the effects of the independent variable are measured. Lind’s scurvy experiment

Lind began his scurvy experiment aboard a British ship after he had been at sea for two months and sailors had begun to show signs of scurvy. He chose a group of 12 sailors with scurvy and divided the group into 6 pairs. The 12 sailors received the same diet, but each couple also received a different daily supplement to the diet (Table\(\PageIndex1\)). Table\(\PageIndex1\): Lind Scurvy ExperimentPair of SubjectsDaily Dietary Supplement Received by this Par11 quart of cider25 drops sulfuric acid36 tablespoons vinegar41 cup sea water52 oranges and 1 lemon6spicy paste and a shot of barley water

Lind’s experiment ended after just five days when fresh citrus ran out of par 5. However, the two sailors of this couple had already fully recovered or improved a lot. The sailors of pair 1 (receiving the cider room) also showed some improvement, but the sailors of the other pairs showed none.

Can independent and dependent variables be identified in Lind’s experiment? The independent variable is the daily supplement that pairs receive. The dependent variable is improvement/no improvement in scurvy symptoms. Lind’s results supported the citrus cure for scurvy, and it was soon adopted by the British navy with good results. However, the fact that scurvy is caused by a vitamin C deficiency was not discovered until nearly 200 years later. Sampling

Lind’s scurvy experiment included only 12 subjects. This is a very small sample by modern scientific standards. The sample in an experiment or other research consists of the individuals or events that are actually studied. It rarely includes the entire population because doing so would likely be impractical or even impossible.

There are two types of errors that can occur when studying a sample rather than the entire population: casual error and bias.A chance error occurs if the sample is too small. The smaller the sample, the greater the likelihood that it will not fairly represent the entire population. The probability error is mitigated by using a larger sample.Bias occurs if the sample is not randomly selected with respect to a variable in the study. This problem is mitigated by taking care to choose a random sample.

A reliable experiment must be designed to minimize both potential sources of error. You can see how sources of error were addressed in another landmark experiment: Jonas Salk’s famous 1953 trial of his newly developed polio vaccine. The massive Salk experiment has been called the “largest public health experiment in history.” Salk Polio Vaccine Experiment

Imagine a nationwide epidemic of a contagious flu-like disease that primarily attacks children and often causes paralysis. That is exactly what happened in the United States during the first half of the twentieth century. Beginning in the early twentieth century, there were repeated cycles of polio epidemics, and each seemed to be stronger than the last. Many children ended up on life support in so-called “iron lungs” (see photo below) because their respiratory muscles were paralyzed by the disease. Figure\(\PageIndex2\): This photo shows the iron lung ward at a California hospital in 1953, the same year Salk undertook his nationwide vaccine experiment.

Polio is caused by a virus, and there is still no cure for this potentially devastating disease. Fortunately, it can now be prevented with vaccines. The first polio vaccine was discovered by Jonas Salk in 1952. After testing the vaccine himself and his family members to evaluate its safety, Salk undertook a nationwide experiment to test the vaccine’s effectiveness using more than a million schoolchildren as subjects. It’s hard to imagine a nationwide trial of an experimental vaccine using children as “guinea pigs.” It would never happen today. However, in 1953, polio struck so much fear in the hearts of parents that they accepted Salk’s word that the vaccine was safe and gladly allowed their children to participate in the study.

Salk’s experiment was very well designed. First, it included two very large random science experiments samples of children — 600,000 in the treatment group, called the experimental group, and 600,000 in the untreated group, called the control group. Using very large, random samples reduced the potential for chance error and bias in the experiment. Children in the experimental group were injected with the experimental polio vaccine. Children in the control group were injected with harmless saline (salt water). Saline injection was a placebo. A placebo is a “fake” treatment that doesn’t actually have any effect on health. It is included in trials of vaccines and other medical treatments, so subjects will not know which group (control or experimental) they have been placed in. Using a placebo helps researchers control the placebo effect. It is a psychologically based reaction to a treatment that occurs only because the subject is treated, even though the treatment has no real effect.

Experiments in which a placebo is used are usually blinded experiments because the subjects are “blind” to their experimental group. This helps prevent bias in the experiment. Often, even researchers don’t know which subjects are in which group. This type of experiment is called a double-blind experiment because both subjects and researchers are “blind” to what subjects are in each group. Salk’s vaccine trial was a double-blind experiment, and double-blind experiments are now considered the gold standard of clinical trials of vaccines, therapeutic drugs, and other medical treatments.