Unsystematic Variance

Unsystematic variance is variability within individuals and/or groups of individuals. This variability is essentially random; some individuals change in one direction, others in an opposite direction, and some do not change at all. For example, some individuals may feel better than they did yesterday, others feel worse than they did yesterday, and some feel the same as they did yesterday. Another example would be in taking tests. Some individuals see test improvement the next day, others see their performance decline, and still others see essentially the same test performance. When taken together, the overall result of this variance is little overall change. Most researchers would consider unsystematic variance a fact of life that cannot be totally eliminated or accounted for at this time. However, it may be possible to account for some unsystematic variance at a later time when our understanding of human behavior and our measurement systems evolve to allow for such accountability (Sidman, 1960). There are two major sources of unsystematic variance--measurement error and individual differences (see Figure 6.1).

Measurement Error

Measurement error exists because a measurement device (e.g., a test, a person doing an observation) does not measure an attribute the same way every time. Measurement error occurs frequently. In fact, we could assume that all measurements have some measurement error. For example, if you weighed 150 pounds but the scale weighed you at 148 pounds, there would be measurement error on that scale of 2 pounds. In other words, the scale has error of measurement built into it. In order for this measurement error to be considered unsystematic, the scale would weigh you at 148 pounds at one time, 153 pounds at another, and 150 pounds a third time. The sum of all your weight differences from the average weight (the sum of all weighings divided by the number of weighings) would be zero if we did an infinite number of weighings. In the end, the measurement error would be a wash, and that average would be the best estimate of your true weight.

We can never weigh ourselves an infinite number of times. Therefore, measurement error must be estimated. Measurement error is directly related to the reliability of the measurement instrument (i.e., how consistently the device measures an attribute)--the less reliable the measurement device, the larger the measurement error. Researchers are concerned with the reliability of their measurement devices since they desire to reduce their measurement error.

Individual Differences

Another source of unsystematic variance is individual differences. Most researchers assume that intrinsic factors, or factors unaccounted for, cause people to be different from one another. For example, one reason why you do well in reading and a peer does not may be due to an inborn ability to achieve at reading or to factors of which the researcher is unaware, such as past learning histories. Therefore, some people may respond favorably to reading instruction whereas others may not.