|Moderator:Jouni (see all)|
<section begin=glossary />
- Variable is a description of a particular piece of reality. It can be a description of a physical phenomenon, or a description of value judgements. Also decisions included in an assessment are described as variables. Variables are continuously existing descriptions of reality, which develop in time as knowledge about the topic increases. Variables are therefore not tied into any single assessment, but instead can be included in other assessments. A variable is the basic building block of describing reality.<section end=glossary />
What should be the structure of a variable such that it
- is able to systematically handle all kinds of information about the particular piece of reality that the variable is describing, especially
- it is generic enough to be a standard building block in decision support work (including interpretation of scientific information and political discussions),
- is able to systematically describe causal relationships between phenomena and variables that describe them,
- enables both quantitative and qualitative descriptions,
- is suitable for any kinds of variables, especially physical phenomena, decisions, and value judgements,
- inherits its main structure from universal objects,
- complies with the PSSP ontology,
- can be operationalised in a computational model system,
- results in variables that are independent of the assessment(s) they belong to;
- results in variables that pass the clairvoyant test.
- can be implemented on a website, and
- is easy enough to be usable and understood by interested non-experts?
Variable is implemented as a web page in Opasnet wiki web-workspace. A variable page has the following structure.
|Attribute||Sub-attribute||Comments specific to the variable attributes|
|Name||An identifier for the variable. Each Opasnet page have two kinds of identifiers: the name of the page (e.g. Variable) and the page identifier (e.g. Op_en2022). The former is used e.g. in links, the latter in R code.|
|Question||Gives the question that is to be answered. It defines the scope of the variable. The question should be defined in a way that it has relevance in many different situations, i.e. makes the variable re-usable. (Compare to an assessment question, which is more specific to time, place and user need.)|
|Answer||An answer presents an understandable and useful answer to the question. Its essence is often a machine-readable and human-readable probability distribution (which can in a special case be a single number), but an answer can also be non-numerical such as "very valuable" or a descriptive table like on this page. The units of interconnected variables need to be coherent with each other given the functions describing causal relations. The units of variables can be used to check the coherence of the causal network description. This is a so called unit test. Typically the answer contains an R code that fetches the ovariable created under Rationale/Calculations and evaluates it.|
|Rationale||Rationale contains anything that is necessary to convince a critical reader that the answer is credible and usable. It presents the reader the information required to derive the answer and explains how it is formed. Typically it has the following sub-attributes, but also other are possible. Rationale may also contain lengthy discussions about relevant topics.|
|Data||Data tells about direct observations (or expert judgements) about the variable itself.|
|Dependencies||Dependencies R↻ tells what we know about how upstream variables (i.e. causal parents) affect the variable. In other words, we attempt to estimate the answer indirectly based on information of causal parents. Sometimes also reverse inference is possible based on causal children. Dependencies list the causal parents and expresses their functional relationships (the variable as a function of its parents) or probabilistic relationships (conditional probability of the variable given its parents).|
|Calculations||Calculations R↻ is an operationalisation of how to calculate or derive the answer. Formula uses algebra, computer code, or other explicit methods if possible. Typically it is R code that produces and stores the necessary ovariables to compute the current best answer to the question.|
|Data not used||Data not used are relevant for the research question, but for some reason they were not used in producing the current answer. I may be that the data was found after the synthesis, and an update has not yet been done; or it has been unclear how to merge these to the existing data. In any case, it is important to be differentiate and be explicit about whether data is irrelevant (and therefore removed from the page) or relevant but not used (and therefore waiting for further work).|
In addition, it is practical to have additional subtitles on a variable page. These are not attributes, though.
- See also
- Keywords (not always used)
- Related files
The structure is based on extensive discussions between Mikko Pohjola and Jouni Tuomisto in 2006-2008 and intensive application in Opasnet ever since.
For more detailed description about variables as information objects, see knowledge crystal.
- List of all variables in Opasnet
- Universal object
- Open assessment
- A previous version of this page contains much of the discussion from the Intarese deliverables D17 and D18, which has been edited with a hard hand.