Documenting Excellence
A consulting practice focusing on working with colleges', organizations', and individuals' utilization of quantitative and qualitative assessment tools to analyze and document their quality outcomes through providing staff development, research design and analysis, and psychometric evaluations.

Further Information

Primer of Statistical Tools

The relationship of cause and effect, probability and determinism - is one of the most difficult concepts for many people to understand.  When social scientists talk about something "causing" some result, they generally mean that the probability, or chance, of that results in increased by the cause.  The most frequent patterns are not a certain cause in which 100% of the cases exposed to the cause show the result.

For example, it is frequently stated that "growing up in single-parent family causes delinquency."  Such a statement will tend to appear in general discussions and public sources.  It is not the way a researcher would state the relationship.  A researcher might say that "the probability of delinquent behavior is doubled for young males raised in a single-parent household." 

However, even that statement makes it sound like many boys growing with such a family relationship will turn out delinquent.  This assumption makes the mistake of assuming determinism in the cause and effect statement.  Then, many people argue that they or friends were in such a household and they were no delinquent, so the cause and effect statement must be wrong.  This is a logical fallacy of overgeneralization or rejection based on the exception.

In our example, the observed pattern in the US is that in general about 7% of males in households with two patents  will have some level of serious contact, other than minor misdemeanors such as speeding, with the criminal justice system by their 18th birthday.  The rate of serious contact for males in households with one parent is about 14%, or twice as high. 

Notice that there is a substantial change in the probability for serious criminal justice contact.  So, the cause and effect statement is valid.  But, most boys are not delinquent.  So, single-parenting does not determine the outcome delinquency by causing 100% of boys from single-parent homes to become delinquent..

In a similarly way,  patterns of information from surveys and public opinion are focused on modes and medians, not on a patterns that is valid for all of the populations.  As the cartoon illustrates, variation among  humans may lead to many exceptions to a pattern that describes an average or typical case.  It is this relationship that makes research both important and yet not always predictive in understanding our shared world.