goodness of fit

n. 1. This refers to how well a statistical model fits a set of observations. The chi-square test statistic is used to see whether the observed proportions in two or more categories differ significantly from a priori, or theoretically expected, proportions. In other words, it looks for a discrepancy between observed values and the expected values. This test can be used in statistical hypothesis testing, as when one tries to see whether two samples are drawn from the same population, or when one tests for the normality of residuals. It is also used to see whether outcome frequencies follow a specified distribution. 2. The goodness of fit model takes into account the interactionist belief that an individual is a product of physical and environmental factors. The belief is that children of certain temperaments (a biologically based factor) may be at risk for future pathology. However, having a difficult temperament does not predict a negative outcome. Parental attitudes and practices, dynamics of the family system, early life experience, and stressors in the community combine with biological factors to influence the child. If the goodness of fit exists between the child’s biology and his/ her environment, then there will be a positive result. This result cannot be identified by any one characteristic, as it is not a sum of factors, but the way they interact with one another.

- EF