<?xml version="1.0" encoding="Windows-1252" ?>

<EUROFIRSCHEME>
<HEADER>
<EXPORTDATE>2007-11-04</EXPORTDATE>
<TITLE>EuroFIR Value Type Thesaurus version 1.0</TITLE>
<DEFINITION>EuroFIR Value Type Thesaurus version 1.0 Thesaurus Export File</DEFINITION>
<URI>http://www.eurofir.org/xml/scheme/EuroFIR_Value_Type_Thesaurus1.0.xml</URI>
<NOTE>This is the official XML export file for EuroFIR Value Type Thesaurus version 1.0</NOTE>
</HEADER>
<DESCRIPTORS>
<DESCRIPTOR>
<FTC>AR</FTC>
<TERM>as reported</TERM>
<BT>ROOT [00000]</BT>
<SN>Use when the Selected Value was assigned as reported, e.g. in a reference book, but not evaluated by the compiler. This may be appropriate for non-compositional values, for example conversion factors or density.  Use *As reported [AR]* together with Method Type *Method type not known [X]*.&#60;br&#62;</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-10-31</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>AV</FTC>
<TERM>average</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when the Selected Value was chosen as the average of values coming from different sources or different statistical distributions* (not necessarily normal distributions). &#60;br&#62;&#60;br&#62;</SN>
<AI>For Value Type AV, the only statistical fields that can be filled are N (number of contributing values), Minimum and Maximum.&#60;br&#62;&#60;br&#62;*A statistical distribution may consist of results from replicate analysis or results from separate samples (e.g. from differents projects) for which the separate samples&#38;#39;  value distributions do not differ significantly (statistically, usually tested with F- or similar test) and in which case all results can be combined.&#60;br&#62;</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>BL</FTC>
<TERM>below detection limit</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when it is known that the component is present in the food, but the component was not detectable with the applied method, and thus was below the limit of detection (LOD).&#60;br&#62;LOD must be provided within the corresponding method description. If LOD is not know, use *trace [TR]*&#60;br&#62;BL can be used with a blank Selected Value.  &#60;br&#62;</SN>
<AI>BL will often be used with a Selected Value &#62; 0. It is important when data are use for exposure assessment. A normal Selected Value can be 0.5 * LOD. For the statistical evidence of this, see http://www.who.int/foodsafety/publications/chem/en/lowlevel_may1995.pdf</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-12</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>BE</FTC>
<TERM>best estimate</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when the Selected Value was chosen as the “best” available. BE is weaker than *mean [MN]*, which applies to normal distributions, and *average [AV], which applies to any averaging of values.&#60;br&#62;*Best Estimate* can be used even if further statistical information is reported.&#60;br&#62;A value obtained from other component values by a single calculation is a normally assigned as *Best Estimate*.&#60;br&#62;</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>LT</FTC>
<TERM>less than</TERM>
<BT>ROOT [00000]</BT>
<SN>Use only if the Selected Value is the highest value, if there is no further statistical/technical information available, and if no other value type applies. Before applying this descriptor, refer to *below detection limit [BL]* and *trace [TR]*.&#60;br&#62;LT can be useful in case of calculated or imputed values. &#60;br&#62;The figure given in Selected Value should be interpreted as an upper limit.&#60;br&#62;</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-11</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>LZ</FTC>
<TERM>logical zero</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when the Selected Value iss assigned as &#38;#39;0&#38;#39; as the component is not assumed to occur in the food, e.g. alcohol in meat, or fat in mineral water (use LZ together with Method Type U). &#60;br&#62;In the case of the Selected Value being assigned as &#38;#39;0&#38;#39; due to regulatory requirements, use LZ together with method type *estimated according to regulatory requirements [L]*&#60;br&#62;&#60;br&#62;</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>MX</FTC>
<TERM>maximum</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when he Selected Value was chosen as the maximum value of the set of values, whether or not these values belong to the same statistical distribution*.  &#60;br&#62;</SN>
<AI>The Selected Value corresponds to the value reported in the Maximum field when details of the statistical results are given.&#60;br&#62;&#60;br&#62;*A statistical distribution may consist of results from replicate analysis or results from separate samples (e.g. from differents projects) for which the separate samples&#38;#39;  value distributions do not differ significantly (statistically, usually tested with F- or similar test) and in which case all results can be combined.</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>MN</FTC>
<TERM>mean</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when the Selected Value was chosen as the arithmetic mean of values deriving from the same statistical distribution*.  This descriptor may only apply to normal distributions.&#60;br&#62;For other value distributions and combinations, use *average [AV]*.  &#60;br&#62;</SN>
<AI>The Selected Value corresponds to the value reported in the Mean field when details of the statistical results are given.&#60;br&#62;&#60;br&#62;*A statistical distribution may consist of results from replicate analysis or results from separate samples (e.g. from differents projects) for which the separate samples&#38;#39;  value distributions do not differ significantly (statistically, usually tested with F- or similar test) and in which case all results can be combined.&#60;br&#62;</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>MD</FTC>
<TERM>median</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when the Selected Value corresponds to the value reported in the Median field when details of the statistical results are given.</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-10</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>MI</FTC>
<TERM>minimum</TERM>
<BT>ROOT [00000]</BT>
<SN>Use when the Selected Value was chosen as the minimum value of the set of results, whether or not these belong to the the same statistical distribution*. &#60;br&#62;</SN>
<AI>The Selected Value corresponds to the value reported in the Minimum field when details of the statistical results are given.&#60;br&#62;&#60;br&#62;*A statistical distribution may consist of results from replicate analysis or results from separate samples (e.g. from differents projects) for which the separate samples&#38;#39;  value distributions do not differ significantly (statistically, usually tested with F- or similar test) and in which case all results can be combined.&#60;br&#62;</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>MT</FTC>
<TERM>more than</TERM>
<BT>ROOT [00000]</BT>
<SN>Use only if the Selected Value is the lowest value, if there is no further statistical information available, and if no other value type applies.&#60;br&#62;MT is also useful in case of calculated or imputed values, e.g. in recipe calculation.&#60;br&#62;The figure given in Selected Value should be interpreted as an lower limit.&#60;br&#62;</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-11</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>E</FTC>
<TERM>other value type</TERM>
<BT>ROOT [00000]</BT>
<SN>Use for other Value Type not defined in this thesaurus.  Use of this term should be avoided, if used only with great caution. See also Additional Information.&#60;br&#62;</SN>
<AI>E = else.&#60;br&#62;Except for its use specified in a EuroFIR thesaurus or guidelines, please inform the maintainer of this thesaurus of the circumstances requiring the use of this descriptor in case a new Value Type should be defined.&#60;br&#62;</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>TR</FTC>
<TERM>trace</TERM>
<BT>ROOT [00000]</BT>
<SN>Use Trace only when there is evidence that some amount of the component is present but no precise figure can be given and no information on e.g. LOQ/LOD.&#60;br&#62;The TR descriptor is weaker than *below detection limit [BL]*.&#60;br&#62;Further information about the exact definition of Trace should be provided ia a Remarks field in the appropriate in Value, Method Specification, Component or Reference record.</SN>
<AI>The trace decsriptor is often used in printed food composition tables to indicate real values lower that the &#38;quot;resolution&#38;quot;, i.e. the number of decimals, used for the component the tables. Representation of a value of 0.05 is not possible, if the component is shown with with 1 decimal as the resulting value would be 0 (zero).&#60;br&#62; In such cases, the value is indicated by e.g. a &#38;quot;tr&#38;quot; or &#38;quot;trace&#38;quot; to indicate that there is some amount of the component present.</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-11</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>UD</FTC>
<TERM>undecidable</TERM>
<BT>ROOT [00000]</BT>
<SN>Use this Value Type with a blank Selected Value, if values exist but no decision can be made, e.g. because the data differ too much. Method information or some statistical parameters (e.g. minimum and maximum) might however be available.&#60;br&#62;If no information about the Selected value is available, use &#38;quot;unknown [N]&#38;quot;.&#60;br&#62;</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-11</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>N</FTC>
<TERM>unknown</TERM>
<BT>ROOT [00000]</BT>
<SN>Use this value type together with a blank Selected Value and with Method Type X , if compilation work has shown the value to be unknown, i.e. there is no literature available and no estimation or calculation possible. &#60;br&#62;If a Selected Value is present, but nothing is known about it origin or derivation, etc., use &#38;quot;value type not known [X]&#38;quot;.</SN>
<AI>Value Type &#38;quot;N&#38;quot; is only useful in the table of aggregated values of a FCDB, in cases where there is no Selected Value, no statistics, etc.&#60;br&#62;In cases where the compiler is completely clueless as to what value to assign. It signals missing data for future analyses and enables tables to be printed correctly.</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-11</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>X</FTC>
<TERM>value type not known</TERM>
<BT>ROOT [00000]</BT>
<SN>Use when the Value Type for the given value is not known.&#60;br&#62;&#60;br&#62;&#60;br&#62;</SN>
<AI>Value type &#38;#39;unknown [X]* apply to data taken from a publication where nothing is known about how the value was calculated.</AI>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-10-31</DATEUPDATED>
<DATECREATED>2007-09-11</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
<DESCRIPTOR>
<FTC>W</FTC>
<TERM>weighted</TERM>
<BT>ROOT [00000]</BT>
<SN>Used when the Selected Value was chosen as the weighted average of values from different sources or different statistical distributions, for example weighting by brands by market share, by number of samples, etc.</SN>
<CLASSIFICATION>False</CLASSIFICATION>
<ACTIVE>True</ACTIVE>
<DATEUPDATED>2007-09-10</DATEUPDATED>
<DATECREATED>2007-09-10</DATECREATED>
<SINGLE>1</SINGLE>
</DESCRIPTOR>
</DESCRIPTORS>
</EUROFIRSCHEME>
