Data Set (column in a Data Table): These are results of an experimental determination of a property from a particular source (i.e., article or reference). It is possible to have more than one data set from a single source for a given compound and property, if, for example, different methods or different samples were used.

Default Representations (Navigation Tree item): Default Representations are critically evaluated property representations generated dynamically by the ThermoData Engine (TDE) software. As part of the evaluation process, the program chooses amongst models of different complexity based upon data-quality and range considerations. The Default Representation is the model chosen by the software for a particular data scenario. The program user can select an Alternative Equation for secondary fitting, if desired.

Evaluated Data (Navigation Tree item): These are results of the dynamic data evaluation (a critical evaluation) generated by the ThermoData Engine (TDE) program. NOTE: These should not be confused with Predicted Data, which are generated with a particular estimation method, such as group-contribution or corresponding states.

Experimental Data (Navigation Tree item): Experimental data are defined as those obtained as the result of a particular experiment on a particular sample by a particular investigator. The feature that distinguishes experimental data from predicted and critically evaluated data is use of a chemical sample.

Invalid Data (column entry in a Data Table): This term is assigned by the program to data that are invalid by definition. For example, values involving negative temperatures or pressures. Origins of these problems include typographical errors in the bundled database of experimental values, typographical errors in the original data report, misidentified compounds in either the database or original report, errors in units, and others.

Large Deviation (column entry in a Data Table): This term is assigned by the program to data that show a relatively large deviation from the fitting equation or weighted average value. Rejection involves multiple considerations including deviations from predicted data, deviations from fitted equations or weighted averages, and relative local data quality. Local data quality is the data quality near the data point under consideration. Complete descriptions of the methods used for data rejection will be published separately.

Navigation Tree: The leftmost program frame that includes expandable nodes entitled Compound, Experimental and Predicted Data, Default Representations, etc.

Navigation Tree End Node: The final (end) expansion item for an entry in the Navigation Tree.

Out of Range (column entry in a Data Table): This term is assigned by the program to data that are invalid by definition. The meaning is similar to that for Invalid Data, except that for Out-of-Range data the lack of validity is with the variable(s) rather than with the property. (For example, vapor pressures for T > T_c, or densities for the liquid phase for p < p_sat are 'out of range.')

Predicted Data: Property values generated with a particular estimation method. (In version 1.0 of the program, only group-contribution and corresponding states methods are used to derive these values.) These values are used to check the validity of experimental data and to aid in the extrapolation or interpolation of experimental values. NOTE: These values should not be confused with Evaluated Data, which are the critically evaluated results generated by the ThermoData Engine software.

Rejected (column in a Data Table): This column is used to indicate data that were rejected by the ThermoData Engine software during the critical evaluation process. Entries in this column are Invalid Data, Out of Range, and Large Deviation.

Smoothed Data (column in a Data Table): At times, data are reported in original sources in the form of equations only or as tables generated from fitted equations. Such values are not primary experimental data and are given the 'smoothed' designation.

Source (column in a Data Table): A 'source' is a data source, such as a journal article, thesis, report, prediction, or evaluation. Data sources are displayed in various formats.

• Experimental Data:
The source codes (e.g., 2003 chi ste 0) represent the year, the first 3 letters of the first author's name, the first 3 letters of the second author's name, and an integer extension used to resolve otherwise duplicate entries. Right clicking the data point number to the left of the source code is used to access full Bibliographic Details.
• Predicted Data:
The method of prediction is indicated for each data point. A typical entry might be: Predicted by Sastri-Rao Method (for viscosity of the saturated liquid). References to descriptions for each prediction method are available. (Prediction Methods & References).
• Evaluated:
These are results of the dynamic data evaluation (a critical evaluation) generated by the ThermoData Engine (TDE) program. NOTE: This should not be confused with Predicted Data, which are generated with a particular estimation method, such as group-contribution or corresponding states. These data are listed only within the Evaluated Data node of the Navigation Tree. These data are always labeled with the term: Evaluated.
• User Data:
Data supplied by the user of the program are designated with the phrase User Data (default) or some other phrase supplied by the user. Procedures for Addition of User Data are provided.