E

*EDI (electronic data interchange)

The electronic exchange of data usually in forms that are compatible so that software or a combination of individuals and software can put in a compatible form at the receiving end if necessary. An example might be pulling data of one type of data base management system into a sequential format and then transferring the data to a second location where the data are stored in a format different from the originating data base management system.

*Note: This definition may not be fully satisfactory.

Source: UNECE Data Editing Group

Edit Rule

See: Checking Rule

Edit Rule Specification

See: Checking Rule Specification

Editing Bounds

Bounds on the distribution of a measure used on survey data so that when raw, unedited data are outside the bounds the data is subject to review and possible correction (change).

Source: UNECE Data Editing Group

Editing Code

See: Editing Indicators

Editing Efficiency

An efficient edit is good at identifying suspicious data. Edits that incorrectly flag large amounts of valid data are not very efficient. If such edits require an analyst to re-contact the respondent to verify the data, the analyst will have less time to perform other tasks, such as converting non-respondents. The hit rate may be used as an indicator of editing efficiency.

Source: UNECE Data Editing Group

Editing Flag

See: Editing Indicators

*Editing Indicators (editing flag, editing code)

A flag or code that is added during the edit process. The flag might indicate that:

(1) a field in a record was targeted for change because it failed an edit;

(2) the field in the record was changed; or

(3) an override code was entered so that the edit system would no longer fail the record or the field in the record.

*Note: This definition may not be fully satisfactory.

Source: UNECE Data Editing Group

Editing Match (matching fields, statistical match)

For hot-deck imputation, the fields in an edit-failing record that do not fail edits are matched against edit-passing records. Often the edit-passing record that is closest to the edit-failing record interms of some metric is chosen as the donor record. If the edit-passing records are in random order, then the (possibly erroneous) assumption is that the donation is at random from a valid set of donors. This type of matching is sometimes referred to as statistical matching.

Source: UNECE Data Editing Group

Edit(ing) Matrix

A matrix used in editing. In hot-deck imputation, the matrix contains information from edit-passing records that is used to donate information to (impute values in) the edit-failing record. Typically, there will be a variety of matrices that correspond to the different sets of variables that are matched on. The matrices are updated continuously as a file of records is processed and additional edit-passing records become available for updating the matrices.

Source: UNECE Data Editing Group

*Editing Measure

A formal measure such as the Hidiroglou-Berthelot statistic that allows delineation of a targeted subset of records for manual follow-up. The statistic might be a measure such as size that allows the most important respondents to be manually reviewed.

*Note: This definition may not be fully satisfactory.

Source: UNECE Data Editing Group

Editing of Individual Data

The lowest logical level of checking and correction during which the relationships among data items are not considered.

Source: UNECE Data Editing Group

See also: Editing of Individual Records, Editing of Logical Units

Editing of Individual Records

Logical level of checking and correction during which the relationships among data items in one record are considered.

Source: UNECE Data Editing Group

See also: Editing of Individual Data, Editing of Logical Units

Editing of Logical Units

A logical level of checking and correction during which the relationships among data in two or more records are considered, namely in a group of records that are logically coupled together.

Source: UNECE Data Editing Group

See also: Editing of Individual Data, Editing of Individual Records

Editing Procedure

The process of detecting and handling errors in data. It usually includes three phases:

the definition of a consistent system of requirements,
their verification on given data, and
elimination or substitution of data which is in contradiction with the defined requirements.

Source: UNECE Data Editing Group

Editing Rationality

An editing process that focuses on improving the incoming data quality and hence the overall quality of the survey data. This may include moving most of the editing as close as possible to the collection of data, limiting manual follow-up to those flagged records with the heaviest potential impact on estimates, and applying a total quality management approach to data editing.

Source: UNECE Data Editing Group

Electronic Data Interchange

See: EDI (Electronic Data Interchange)

Electronic Questionnaire

Questions are in a software system that can be answered by an individual. Examples might be a software system that is on a laptop computer where respondents can answer questions directly into the laptop (possibly without knowledge on the part of interviewers of the details of the answers) or through queries on an Internet page.

Source: UNECE Data Editing Group

See also: Generated Versions of Questionnaires

Error Detecting Character

Character added to the basic characters of the code word. Its relationship to the basic charactersis specified previously. The relationship is specified so that typical errors in the code word transmitted break this relationship.

Source: UNECE Data Editing Group

Error Detecting Characters (check digit)

Code or given set of code words whose relationship to the set of valid codes is known, such that when a transcription error occurs, the relationship is violated and the error is detected with certainty or very high probability.

Source: UNECE Data Editing Group

Error Detection

An activity aimed at detecting erroneous data, using predefined correctness criteria. The correctness criteria can be defined through various checking rules.

Source: UNECE Data Editing Group

Error Diagnostics

See: Report on Errors

Error Localization

The (automatic) identification of the fields to impute in an edit-failing record. In most cases, an optimization algorithm is used to determine the minimal set of fields to impute so that the final (corrected) record will not fail edits.

Source: UNECE Data Editing Group

Error Statistics

A statistical report of errors found. It usually provides the error rate in individual data items and the occurrence of particular kinds of errors.

Source: UNECE Data Editing Group

Evaluation of Editing

There are several methods available to evaluate an edit system. One method is to compare the raw (unedited) data file with the edited file for each question. By ordering the changes by descending absolute magnitude, the cumulative impact of the editing changes to the total editing change or to the estimated item total can be displayed in graphs and tables. This technique can be used to identify questionnaire problems and respondent errors. When forms are reviewed before data capture or when general editing changes are noted on the questionnaires, evaluation studies can be carried out by selecting a sample of forms and analysing the effect of the editing procedures on individual data items. A widely used technique to evaluate new editing methods is to simulate the new process using a raw data file from the survey. By replacing values flagged according to the new method with the values from the tabulation file containing the data edited by the alternative editing process, it becomes possible to compare estimates from this newly edited file with estimates from the tabulation file.

Source: UNECE Data Editing Group

Expert System

An expert system is an intelligent computer program that uses knowledge and inference procedures to solve problems that are difficult enough to require significant human expertise for their solution. Every expert system consists of two principal parts: the knowledge base and the inference engine. The knowledge base contains both factual and heuristic knowledge. Factual knowledge consists of items commonly agreed upon by spokesmen in a particular field. Heuristic knowledge is the less rigorous, more experiential and more judgmental knowledge of performance or what commonly constitutes the rules of "good judgement" or the art of "good guessing" in a field. A wisely used representation for the knowledge base is the rule or if /then statement. The "if part" lists a set of conditions in some logical combination. Once the "if part" of the rule is satisfied, the "then part" can be concluded or problem solving action taken. Expert systems with knowledge represented in rule form are called rule-based systems. The inference engine makes inferences by determining which rules are satisfied by facts, ordering the satisfied rules, and executing the rule with the highest priority.

Expert data editing systems make so-called intelligent imputations based on a specified hierarchy of methods to be used in imputing an item. One item may use a deterministic approach followed by a hot-deck approach, while another item might require a model-based approach. Each item on the questionnaire would be resolved according to its own hierarchy of approaches, the next being automatically tried when the previous method has failed.

Source: UNECE Data Editing Group

Explicit Edit

An edit explicitly written by a subject matter specialist. (Contrast explicit edits with implied or implicit edits).

Source: UNECE Data Editing Group

Explicitly Defined Conflict Rule

A conflict rule which in defined by the people responsible for the correctness of the data.

Source: UNECE Data Editing Group