The property of the set of measurements of being very reproducible or of an estimate of having small random error of estimation. Precision is to be contrasted with accuracy, which is the property of being close to some target or true value. Precision is a quality associated with a class of measurements and refers to the way in which repeated observations conform to themselves; and in a somewhat narrower sense refers to the dispersion of the observations, or some measure of it, whether or not the mean value around which the dispersion is measured approximates the "true" value. OECD Source: The International Statistical Institute, "The Oxford Dictionary of Statistical Terms", edited by Yadolah Dodge, Oxford University Press, 2003.
A probability sample is a sample selected by a method based on the theory of probability (random process), that is, by a method involving knowledge of the likelihood of any unit being selected. OECD Source: United Nations Statistics Division, "Handbook of Vital Statistics Systems and Methods, Volume 1: Legal, Organisational and Technical Aspects", Studies in Methods, Series F, No. 35, United Nations, New York, 1991.
Any method of selection of a sample based on the theory of probability; at any stage of the operation of selection the probability of any set of units being selected must be known. It is the only general method known which can provide a measure of precision of the estimate. Sometimes the term random sampling is used in the sense of probability sampling. OECD Source: A Dictionary of Statistical Terms, 5th edition, prepared for the International Statistical Institute by F.H.C. Marriott. Published for the International Statistical Institute by Longman Scientific and Technical.
A sample in which the individual units are selected by some purposive method. It is therefore subject to biases of personal selection and for this reason is now rarely advocated in its crude form (ISI).
A sample which has been selected by a method of random selection as contracted with one chosen by some method of purposive selection (ISI). Random sampling allows each element of the target population (as represented by the frame) a positive chance of being selected in the sample. This likelihood of being selected is the inclusion probability (or inclusion density for continuous populations); its inverse is the sample weight.
In this "Salmon Monitoring Advisor" web site, we use the term "reference site" to be synonymous with "control". That is, a reference site is a spatial/temporal location that is similar (ideally identical) to another site that only differs from the reference site by being affected to a greater (or lesser) extent by some mechanism that affects salmon. Of course, no two sites can be identical, but the careful choice of one or more reference sites will permit reasonably rigorous conclusions about differences in responses at those sites to the causal mechanism. More information on reference and control sites and the different uses of these terms can be found in Downes et al. (2002, page 122) and Roni et al. (2005, page 22).
Resilience is the term used to describe the extent to which an ecological system responds to disturbances without substantially changing its structure and function (Holling 1972). For instance, imagine two lakes, one that includes only a small number of species and/or functional groups of phytoplankton, zooplankton, and fish, and another thathas multiple species and/or functional groups in each of those categories. The simple lake ecosystem will probably be less able to maintain its previous structure and function after major disturbances such as massive nutrient loading, warming, or changes in seasonal timing of events than the second lake. An often-forgotten second component to the resilience concept that Holling (1972) described is that frequent disturbances help select for characteristics among component populations that lead to greater resilience.
Response designs are the descriptions of how the measurements will be made in the field and how the measurements will be summarized into intermediate results (i.e., what we call metrics in our STRIDE nomenclature).
A sample survey is a statistical technique that allows investigators to describe or characterize an entirety, like a stream network, or the lakes or wetlands in a region, without having to sample everywhere in that entirety. Sample surveys rely on selecting part of the resource of interest, characterizing that part, and then making inferences to the entirety. Sample surveys are especially useful if a census of the resource cannot be conducted (i.e., too expensive; too time consuming; technically not feasible...) A sample survey is a survey which is carried out using a sampling method, i.e. in which a portion only, and not the whole population is surveyed. OECD Source: The International Statistical Institute, "The Oxford Dictionary of Statistical Terms", edited by Yadolah Dodge, Oxford University Press, 2003.
That part of the difference between a population value and an estimate thereof, derived from a random sample, which is due to the fact that only a sample of values is observed; as distinct from errors due to imperfect selection, bias in response or estimation, errors of observation and recording, etc. The totality of sampling errors in all possible samples of the same size generates the sampling distribution of the statistic which is being used to estimate the parent value. Sampling errors arise from the fact that not all units of the targeted population are enumerated, but only a sample of them. Therefore, the information collected on the units in the sample may not perfectly reflect the information which could have been collected on the whole population. The difference is the sampling error (Eurostat, Quality Glossary). OECD Source: The International Statistical Institute, "The Oxford Dictionary of Statistical Terms", edited by Yadolah Dodge, Oxford University Press, 2003.
The frame is the representation of the target resource used in the selection of the sample. For discrete populations, the frame is the list containing each population element, the list of lakes or streams in the region of interest, sometimes referred to as a "list frame". For continuous resources, such as stream networks, a digital map of the stream network is the usual form of the frame. Accurate representations of stream networks therefore become critical as they become the functional target population. A list, map or other specification of the units which define a population to be sampled. The frame may or may not contain information about the size or other supplementary information of the units, but should have enough details so that a unit, if included in the sample, may be located and taken up for enquiry (ISI). Also, under sampling frame: A list of all members of a population used as a basis for sampling. Without such a frame, or its equivalent, methods of sampling with assured properties such as unbiasedness are not available. The frame in effect defines the study population (ISI; Thompson, 1992)
Sampling in which every member of the population has an equal chance of being chosen and successive drawings are independent as, for example, in sampling with replacement. OECD Source: A Dictionary of Statistical Terms, 5th edition, prepared for the International Statistical Institute by F.H.C. Marriott. Published for the International Statistical Institute by Longman Scientific and Technical.
Spatial sampling designs consist of three major components: response design, spatial design and temporal design. Developing a spatial sampling design, i.e. a monitoring program, to meet the objectives of a study requires decisions on the three components with an understanding that a decision for one component imposes restrictions on the others.
Spatial balance refers to the idea that sample points be distributed in some regular or nearly regular pattern across the resource of interest. Simple random sampling, for example, can leave "holes" in the distribution of candidate sample points across the resource or lead to unnecessary clumping.. Systematic sampling can align with the spatial structure of natural features such that sites of a particular type are over-emphasized or under-emphasized. The ridge and valley region of Nevada is an example of a situation in which a systematic sample could align with a natural resource.
The time of year when productivity of phytoplankton and zooplankton at the bottom of the salmon food chain in the ocean greatly increases due to mixing of nutrient-rich waters at depth with surface waters that are exposed to sunlight.
The target population refers to the resource to be described. For example, it might be all the registered voters in the United States; or all citizens over the age of 65. It might be the number of fish in a stream network in a particular watershed, the biological condition of streams and rivers in a state, or the habitat condition of streams in a national forest. Critical in developing the design is an explicit definition of the target population. The definition of the target population should contain specific information about the stream network: its spatial extent, its flow status (the perennial network? Includes the intermittent channels?); its size (all stream sizes? Just first order streams?). The definition should be specific enough that an individual could determine whether a location on a stream network is part of the target population; in some cases, membership in the target population might be determined after data have been collected at the site. In statistical usage the term population is applied to any finite or infinite collection of individuals. It has displaced the older term ‘universe’…(ISI).
Target populations can be divided into discrete subpopulations, or strata, on which to increase/decrease sample size. A stream network’s elevation could be used to divide the population into elevation strata, allocating an equal number of sites per stratum (likely yielding inclusion probabilities that vary by stratum because the amount of stream length in each stratum likely would vary). Stratification is the process of grouping members of the population into subgroups before sampling. The strata should be mutually exclusive and also collectively exhaustive so that no population element is excluded Stratification consists of dividing the population into subsets (called strata) within each of which an independent sample is selected. The division of a population into parts is known as strata, especially for the purpose of drawing a sample, an assigned proportion of the sample then being selected from each stratum. The process of stratification may be undertaken on a geographical basis, e.g. by dividing up the sampled area into sub-areas on a map; or by reference to some other quality of the population, e.g. by dividing the persons in a town into strata according to gender or into three strata according to whether they belong to upper, middle or lower income groups. The term stratum is sometimes used to denote any division of the population for which a separate estimate is desired, i.e. in the sense of a domain of study. It is also used sometimes to denote any division of the population for which neither separate estimates nor actual separate sample selection is made. (The International Statistical Institute, "The Oxford Dictionary of Statistical Terms", edited by Yadolah Dodge, Oxford University Press, 2003). OECD Source: Statistics Canada, "Statistics Canada Quality Guidelines", 4th edition, October 2003, page 21.
The entire length of time the study will be operated. The goal of the study will be to provide some kind of summary statement covering this entire study period.
Survey design covers the definition of all aspects of a survey from the establishment of a need for data to the production of final outputs (the microdata file, statistical series, and analysis). The survey design addresses the following issues: what statistics are produced, for which population, when, and with what accuracy; what data are to be collected for which units of the population of interest, and what are the methods by which those data are to be collected and processed to produce the required statistics. Operational, organisational and administrative issues are usually addressed (Lessler, J.T. and Kalsbeek, W.D. (1992), "Non Sampling Error in Survey", New York: John Wiley or US department of Commerce (1978), "Glossary of Non Sampling Error Terms: An Illustration of a Semantic Problem in Statistics", Statistical Policy Working Paper 4, Office of Federal Statistical Policy Standards). OECD Source: Statistics Canada, "Statistics Canada Quality Guidelines", 4th edition, October 2003, page 8. NOTE: See ISI definition of sample design for a long list of qualifiers as the term tends to be used in several ways. Also see Spatial sampling designs above for one way of using the terms.
An experimental design laid out without any randomization. The term is difficult to define exactly because in one sense every design is systematic; it usually refers to a situation where experimental observations are taken at regular intervals in time or space (ISI).
Component of monitoring design that describes the temporal periods for the study and the sequence of re-visits at each site selected by the spatial design.
A simple case of a multi-stage sample. In this case the population to be sampled is first classified into primary units, each of which consists of a collection of the basic sampling unit, the secondary unit. A sample of these primary units is taken, constituting the first stage, and these are then subsampled with respect to their secondary units: this constitutes the second stage. OECD Source: A Dictionary of Statistical Terms, 5th edition, prepared for the International Statistical Institute by F.H.C. Marriott. Published for the International Statistical Institute by Longman Scientific and Technical.
Variable probability sampling selects members of the population proportional to an auxiliary variable. If the auxiliary variable is based on membership in subgroups, then it is similar to stratified sampling, except that the number of samples allocated to a subgroup is not fixed but random.
The variance is the mean square deviation of the variable around the average value. It reflects the dispersion of the empirical values around its mean. OECD Source: Eurostat, "Assessment of Quality in Statistics: Glossary", Working Group, Luxembourg, October 2003.
The total variation displayed by a set of observations, as measured by the sums of squares of deviations from the mean, may in certain circumstances be separated into components associated with defined sources of variation used as criteria of classification for the observations. Such an analysis is called an analysis of variance, although in the strict sense it is an analysis of sums of squares. Many standard situations can be reduced to the variance analysis form. OECD Source: A Dictionary of Statistical Terms, 5th edition, prepared for the International Statistical Institute by F.H.C. Marriott. Published for the International Statistical Institute by Longman Scientific and Technical.
Precision
