When someone conceives an idea then recommends its use within a specific area for example a community, he or she must have gone through its relative impacts to this society. This process will involve identifying overall benefits against the anticipated risks for the idea to be deemed ideal for the community. According to Emil J Posavac, a leading professor of psychology, the practice of weighing options before acting or confirming the real impacts of what is implemented is called program evaluation.
Program evaluation therefore is a systematic way of data collection, analysis and relaying the final information for use by the interested groups or key stakeholders. In some cases program evaluation is known as project evaluation. Posavac further intimates that program evaluation is a daily activity which is being performed by almost everybody including the lay groups not trained in this field. He qualifies his claim by acknowledging that at least everyone makes decision(s) every single moment and these decisions are derived from prior analysis of the options.
Among the attributes of program evaluation is reliability. This attribute can be defined as the ability to come to similar results when different related tests are done repeatedly, but with the same apparatus. Proper and quality evaluation must have the said attribute for the claim to be tenable. Those who have not studied social service research methods (hypothetical stakeholders) mainly rely on the observed events for their evaluation. When some of the phenomenal events happen repeatedly, but bringing the same results then these stakeholders can hypothesize the outcome.
When the outcome of final analysis of information collected brings a completely false result, this can be referred to the type II error. This kind of error is likely to appear in program evaluation due to several factors. Use of wrong source for information gathering with defective instrument for the results can be seen as a double tragedy. This kind of evaluation will definitely lead to the first and second degrees of errors which eventually give the results of total miss.