PCS- Sources of Bias

The lesson on Sampling Techniques presented many different types of samples. Be aware that poor sample designs can produce misleading conclusions. Some issues that compromise the results of conducting a study on a sample of the population include a voluntary response sample which consists of people who choose themselves by responding to a general appeal, like a telephone call-in poll. This is one of the common bad sample designs. Another is convenience sampling which chooses the individuals easiest to reach. Both voluntary response samples and convenience samples choose a sample that is almost guaranteed not to represent the entire population. These sampling methods display BIAS or systematic error by favoring some parts of the population over others. Restating, a sample is said to be biased when there is a systematic failure to represent its population, that is, it may overlook an important group in that population. Every effort should be made to eliminate or minimize the occurrence of bias when conducting a survey or performing an experiment.

The statistical remedy for these personal choice bias samples is to allow impersonal CHANCE to choose the sample. A sample chosen by chance prevents favoritism by the sampler or self-selection by respondents. Choosing a sample by chance attacks bias by giving all individuals an equal chance to be chosen. The simplest way to use chance to select a sample is to place names in a hat (from the entire population) and draw out a handful (the sample). This is the idea of SIMPLE RANDOM SAMPLING (SRS). The random digit table introduced in the simulations lesson can be used for sample selection.

Example of bias

A survey was conducted with students in grades 9 -12 to decide on a school mascot. Since the school was playing only a varsity schedule for the first year, only juniors and seniors participated in the survey. Asking only juniors and seniors to participate made this sample biased because 9^thand 10^thgraders were also students at the school and made up part of the population.

Another Example of bias

A sample of 17-23 year old college students would not be a representation of the entire 17-23 year old population in the country. Conclusions based on this sample would be limited to only those in that age group that attend college and would not apply to the 17-23 year old age group in general.

Other forms of bias may result from the method of sampling. Telephone surveys with "ring-no response" and mailed surveys with no reply are an example of non-response bias. When a specific subgroup of the population is under-represented, that is called under coverage bias. Interviewer or respondent behavior or a poorly worded question is called response bias. Here's a great example of improper and suggestive question wording: "Some cell phone users have developed brain cancer. Should all cell phones come with a warning label explaining the danger of using cell phones?" When respondents choose themselves, or self-select, as in a telephone call in survey usually people with strong opinions respond and this is called voluntary response bias.

IMAGES CREATED BY GAVS