Table 1.

Epistemic and linguistic uncertainties (Regan et al. 2002) relevant to community science in context of research addressing alien species

Epistemic uncertainties
Natural variation Relates to the changes (usually related to space and time) which are naturally and inherently present but often difficult to predict. Community science projects may be able to help reduce such biases, particularly related to natural spatial variation, by upscaling data collection. Example: Populations of alien species vary in different demographic attributes due to natural changes in fecundity and mortality.
Measurement error Occurs due to imperfect measuring equipment and observation techniques, including when the individual and/or equipment causes the error. Example: A participant under- or over-estimates the number of individuals of a species on site. The Global Positioning Satellite (GPS) of a smartphone is slow and records an observation in a different kilometre square or with high associated coordinate uncertainty.
Systematic error Arises due to biases in sampling procedure or measuring equipment. Example: For example, two species A and B look relatively similar. A volunteer recorder consistently records the presence of species A as the presence of species B as they do not realise that there are two different species; or, a volunteer recorder incorrectly sets up a GPS device and now all locations that the participant records are systematically wrong.
Model uncertainty Results from the necessity to represent the ‘true’ situation through the use of simplified models. Model uncertainty is generated from the fact there are a multitude of drivers that affect a process, and we will never capture the true scenario. Model uncertainty may be reduced through model validation methods (Zurell et al. 2010) and ensuring that findings are interpreted within the limits of the model.
Subjective judgement Arises through the interpretation of information. This is relevant from the perspective of both the project co-ordinators and the audience (e.g., volunteer recorders) receiving the information. Linguistic uncertainties may exacerbate subjective judgement. Example: A scientist believes that changes caused by an alien species in an ecosystem (e.g., soil pH) are generally deleterious for native species and therefore describes them in the report as detrimental for the ecosystem.
Linguistic uncertainties
Vagueness Language that permits borderline cases; common when using linguistic categories that underpin continuous measurements. Example: Asking volunteer recorders to provide linguistic size class categories, such as small, medium and large, for a specific species observation, may lead to inconsistencies.
Context dependence When the context under which something is required to be completely understood is absent. Example: Species may be thought of as either native or alien, depending on their geographical range. For instance, species translocated within a country may be perceived as native by some yet alien by others. Understanding the native range is necessary for context to determine if it is alien.
Ambiguity Where more than one meaning for a word or phrase may be interpreted and it is not clear which meaning is correct. Example: The term ‘invasive’ can be interpreted differently as current definitions use it to refer to alien species that are established and widespread across a landscape with no reference to impact, or alternatively, alien species that are perceived to have deleterious impacts.
Under specificity When there is an unwanted generalism and information is not clear due to the lack of detail. Example: Failure to clearly explain to project participants the level of details necessary for each species observation may lead to incomplete records and data gaps.
Indeterminacy of theoretical terms Occurs because language is imprecise, and words can change meaning with time. Example: Species names (both their scientific nomenclature and common names) can change over time, causing confusion to those who were aware of their previous names.