The Implications of all Rules Not Being Created nor Administered Equally

There has been an assumption or even a paradigm focus in which it is assumed that rules are all created equally and are administered equally. On the surface this appears to be the case and what is the big deal. It is a big deal when it comes to how the rules get measured. If they are truly equal and are independent of each other, then they would be measured at a nominal level. However, if they are truly not equal and are dependent upon each other in that there is a ranking in which rules can be sorted on a specific metric, such as risk to the client. Well, that changes everything where we move from a nominal to an ordinal measurement strategy.

Let’s take a specific example, such as comparing a rule dealing with supervision of children and a rule that requires a signature of a parent or guardian in order to review children’s files so that the child can go on a field trip. It can be argued that the supervision of children places children at greater risk than if a signature is missing. To follow this thinking, then the rule dealing with supervision carries more weight than the signature rule. The rules are related in that one requires actual supervision while the other rule is a pre-requisite (parental approval) for going on a field trip where supervision will be required. However, missing a signature is much less of a concern than missing a child while on the field trip. These rules are dependent and related and there would definitely be a ranking with the supervision rule being weighted more heavily than the signature rule. With this ranking we have moved the nominal based rules to an ordinal based rule schema.

If we take this analogy to its logical conclusion then all the rules are inter-related and need to be rank- ordered accordingly. In other words, they need to be assigned weights based upon the relative risk to a child when non-compliance with the rule occurs. Many jurisdictions have done this type of weighting consensus either mathematically via a likert approach or by a more qualitative approach based upon group consensus. In either case, all the rules are rank ordered and weighted on the basis of risk assessment for morbidity or mortality if non-compliance is determined. The most comprehensive example of this approach in the publication Stepping Stones to Caring for Our Children.

This movement from nominal to ordinal measurement drastically changes the potential statistical analyses when utilizing these data to compare programs on various quality dimensions. For example, in studies involving the theory of regulatory compliance it became readily evident that utilizing the nominal measurement scale of rule violations was not as effective as utilizing an ordinal measurement scale. A Regulatory Compliance Scale based upon full, substantial, median, and low levels of regulatory compliance has been found to be much more advantageous in doing these types of analyses in early care and education program quality studies.

The measurement of rules needed to match the importance of the rule and how it was administered. It was the theory meeting up with the metrics of how to assess the rule. Weighting is critical because the theory is that all rules are not created equally as has been the predominant thinking and promulgation of rules.