As much as we try to eliminate bias from our work reports, our testing methods and even our daily life, some degree of bias will inevitably be present in any situation involving human beings. Bias is an important factor for administrators to take into account when they create a test. Overwhelming bias in the test creation process can dramatically skew results. A test is usually at fault for being biased when results are significantly different from one group of test-takers to another - in other words, it is completely unreliable. Item bias can be a result of many unresolved issues during the test creation process. These may include poor grammar, selective cultural phrases and a test creation staff which does not represent the diversity of the test takers. Here are some methods administrators can implement to help avoid the pitfalls of biased tests.