In science, the sample is the group that is selected to be studied as part of both the experimental and control groups. Sampling bias is also referred to as selection biased. In order to avoid sampling bias, samples in scientific investigations should

• be large in order to produce accurate and valid results,
• reflect the characteristics of the larger targeted population, and
• be selected at random so that every individual in the larger target population has the same likelihood of being chosen.

Example: An advertisement states that a particular over-the-counter medicine is doctor recommended. First, you must ask yourself, “How many doctors were surveyed?” If the sample size is not large enough, then the statement might not be reflective of the entire population of doctors.

Inaccurate measurement of variables can also lead to bias in scientific investigations. This can happen when equipment is read incorrectly or was not calibrated to begin with. If the research study involves survey questions, the wording of the questions can lead to bias as well. The conditions in which the investigation is conducted can also lead to bias. Data is the number-one thing that can lead to bias. If the interval of values on the axes of the graph is too large or too small, it can distort the interpretation of the data. In order to be bias free, data collected from the experiment must be repeatable by other researchers.

Example: A window tinting company claims that its product has been proven to block 90% of UV rays. You must ask yourself, "This investigation was conducted under what conditions? Is the data from the investigation repeatable?"