Bias in Statistics > Accidental Bias

## What is Accidental Bias?

Accidental bias is the introduction of one or more nuisance variables into your experiment. These nuisance variables may be known or unknown to the researcher. But in either case they are**unwanted and systematically affect your experimental units**— which means you must take them into account when analyzing your research. The name

*accidental*bias is a bit of a misnomer, because these aren’t variable that are “accidentally” introduced into an experiment by, say, dropping a beaker or missing a page in a document; Nuisance variables are usually intrinsic to an experiment and can’t be avoided by being careful. It could more aptly be called

*chance-variable*bias; Soares (1985) suggests a suitable name would be “lurking-variable bias”.

## How to Avoid Accidental Bias

Randomization is one way to control for accidental bias (Suresh), as long as you are careful to balance treatment assignments by identifying all factors which may affect outcomes. Most randomization techniques will minimize the effects of nuisance variables, except for the truncated binomial design (Chow & Chang). Along the same vein, a random allocation design can minimize the effects of this type of bias in sequential experiments (Soares). When combined with complete randomization, random allocation eliminates accidental bias completely in large sample sizes (over 100) (Chow & liu).

**References**:

Chow, S. & Chang, M. (). Adaptive Design Methods in Clinical Trials.

Chow, S. & Liu, J. (2004), Design and Analysis of Clinical Trials: Concepts and Methodologies.

Soares, J. (1985) Optimality of random allocation design for the control of acc. bias in sequential experiments. Journal of Statistical Planning and Inference 11 81-87.

Suresh, K. (2011) An overview of randomization techniques: An unbiased assessment of outcome in clinical research. J Hum Reprod Sci. Jan-Apr; 4(1): 8–11.

doi: 10.4103/0974-1208.82352. Retrieved January 8, 2017 from: NIH.

If you prefer an online interactive environment to learn R and statistics, this *free R Tutorial by Datacamp* is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try *this Statistics with R track*.

*Facebook page*and I'll do my best to help!