Tests and Measurement Discussion
Factors That Lead to Biases in Testing Process
In the United States, the media constantly reports statistical information to the general public as “interesting findings,” usually presented in a factual manner. In some arguments, there are statistics that prove both opposing sides. But how could statistical data be true OR false, or both, or proven then unproven? One of the factors leading to biases in the testing process is a problem called selective reporting. Selective reporting occurs when investigators are more likely to publish findings when they match predictions than when they do not. For example, when one is testing for a difference in scores between Blacks and Whites, and there is none, or if the results shows a figure different than what was expected, the results are not reported. It is important to understand that when test differences are reported, not all data are published, and the published data may be a less-than-perfect representation of the population.
When a test itself is biased, it systematically underestimates a group’s performance. If fact, some people believe that any test showing a difference between two groups must be biased against the lower-scoring group. In fact, this may not be true and the word bias must indicate that the test scores either exaggerate a difference between groups ore report a difference that doesn’t really exist at all. To determine if a test is actually biased or not, one has to find out whether the certain groups actually perform better than the test predict they will.
Another factor that leads to test biases is historical period in the United States. During the late 1910s and early 1920s, hostility towards immigrants was widespread. IQ tests during that time when immigrants were looked down upon, biased tests proved Immigrants to be intellectual below the American average. After evolution of the accuracy of IQ tests, researchers have found that immigrants or people who do...