Property Management Sampling Approaches and Techniques
DAU GLOSSARY DEFINITION
“Population” means an aggregation of elements, (e.g., documents, records, articles, or actions) selected for review due to common characteristics (also referred to as universe or lot). The entire set of data from which a sample is selected and about which the auditor wishes to draw conclusions.
“Sample” means a number of items, (e.g., documents, records, articles, or actions), selected from a population for a review in order to draw inferences regarding and generalizable to the status of the population.
“Sample item” means the individual items constituting a sample.
“Sampling” means the selection and evaluation of less than 100 percent of the population of audit relevance such that the auditor expects the items selected (the sample) to be representative of the population and, thus, likely to provide a reasonable basis for conclusions about the population.
“Random sampling” is a method of selecting a sample (random sample) from a statistical population in such a way that every possible sample item that could be selected has a predetermined probability of being selected.
“Statistical Sampling” is a technique by which evaluation of attributes or transactions, or fact, or conditions obtained from randomly selected samples forms the basis for determining the degree to which an entire population conforms to standards.
“Judgment Sampling” is a process used to evaluate areas, items, or actions, based upon the reviewer’s professional judgment.
“Purposive Sampling” is a process used to evaluate areas, items, or actions involving credible, known, suspected deficiencies or reported conditions of a critical/substantial nature.
The purpose of this article is to describe the two general approaches to sampling and the sampling techniques used by property administrators and others responsible for performing contractor property management system analyses (akin to performance audits described in the Generally Accepted Government Auditing Standards). The two sampling approaches are probability (statistical) and nonprobability (non-statistical) sampling. The most commonly used methods of sampling by property administrators are statistical sampling, judgment sampling, and purposive sampling. These sampling methods are referenced in the "Guidebook for Contract Property Administration (December 2014)." Prior to proceeding with sampling one must have a sampling plan. Double sampling plans are used by Department of Defense property administrators. It is important that relevant and homogenous populations be defined and properly framed before proceeding with sampling, (e.g., all records of Government property accountable to contracts administered by the Defense Contract Management Agency).
The basis of the statistical sampling is the use of a random sample selection technique. The statistical approach includes simple random sampling and complex random sampling (systematic, cluster, stratified, and double). Simple random sampling is the most widely used when performing a property management system analysis (PMSA). The "Guidebook for Contract Property Administration" states that property administrators should use statistical sampling whenever possible. Proper use of random sampling assures that each item within the population has an equal probability of being selected as a part of the sample. Use of random sampling also allows findings from the sample to be generalized to the population that it was selected from. In other words, a conclusion can be formed about the population based on the results derived from the analysis of the sample. Random sampling also protects against bias. The outcomes and processes identified in FAR 52.245-1(f)(1) that are typically audited using statistical sampling during a property management system analysis are:
- Acquisition of Property;
- Receipt of Property;
- Records of Government Property;
- Physical Inventory;
- Utilizing Government Property;
- Identification of Excess;
- Maintenance; and
Some of the processes above may have process segments for which judgment sampling is used. Ultimately, each property administrator should apply his/her own judgment regarding which sampling approach and technique are most suitable for each process and process segment population.
The statistical approach usually employs the use of computer-based technology for random sample selection. There are many tools on the Internet that can be used to assist in generating a random sample. Examples of such tools are provided at the end of this article along with a Defense Acquisition University Tool that consists of a Microsoft Excel template, directions for generating random samples, and double sampling plans.
Non-statistical sampling does not include the use of a random selection technique. It relies heavily on the experience and professional judgment of property administrators. The non-statistical approach includes both judgment and purposive sampling methods. “Using Statistical Sampling” is one of a series of papers issued by the Government Accounting Office’s prior Program Evaluation and Methodology Division. The paper refers to judgment sampling as "discretionary" and further states:
In this type of sampling, the evaluator bases the selection of a sample on knowledge or judgment about the characteristics of the population. Haphazard or “catch as catch can” samples-for example, grabbing a few items “at random”-are usually included in the category of judgment sampling.
Judgment sampling is useful for auditing processes and process segments that do not lend themselves to statistical methods of sampling (e.g., storage and protection of property). Purposive sampling is closely related to judgment sampling. As a matter of fact in the area of research, judgment sampling actually falls under the category of purposive sampling. Property administrators may choose to use purposive sampling when it is determined, based on known, respected, or reported conditions, that there is potential for a negative systemic impact. For example, another functional specialist, such as a quality assurance representative, may inform a property administrator of contractor actions that he or she witnessed multiple times that could have a substantial impact on other processes within the system.
In conclusion, statistical and non-statistical approaches require property administrators to use professional judgment in planning, selecting, and analyzing a sample. Sampling in any case is a tool. Property administrators should always use sampling with his or her knowledge, experience, and again, professional judgement.