Validation checks

Home
Errors
Types of errors
Transcription Errors
Transmisson errors
Processing Errors
Reducing errors
Verification
Validation
Validation checks
Batch processing
Accuracy and validity

Field Presence Check

 To ensure that all necessary fields are present.

This checks that an entry has been made for the field.  For example, the Surname field in an order form cannot be left blank. is a required field.

Field Length Check

To ensure that an item of data has the correct number of characters.

It determines the minimum and maximum length of the field. It can make sure the minimum has been entered - for example a date must have at least 6 numbers in it e.g. 16/10/04. If only 5 numbers have been entered, it will generate an error.

It can also check the maximum length, for example a phone number cannot have more than 12 numbers.

Range Check

 To ensure that data value is within a pre-determined range.

 This checks a value to be within a certain range of values.  For example, the month of a year must be between 1 and 12.  Numbers less than 1 or greater than 12 would be rejected.

Format check (also called a picture check)

 To ensure the individual characters that make up the data are valid - e.g. no letters in numerical data.

This checks that data is of the right format, that it is made up of the correct combination of alphabetic and numeric characters.  A National Insurance number must be in the form of XX 99 99 99 X.  The first two and the last characters must be letters.  The other six characters are numbers.  The total length is nine characters.  Any other format is rejected.

Batch header check

 This is concerned with Batch processing - see later.

The total number of records in the batch should be calculated by the computer and compared with the figure on the batch header.  The control totals and hash totals are also calculated and compared.

Check Digit
Allows a number to be self-checking.

This is used to check the validity of code numbers, for example product codes in a supermarket or bank account numbers.  These numbers are long and prone to data entry errors.  It is crucial that such numbers are entered correctly so that the right record in the file or database is identified.

A check digit is an extra digit added to the end of the original code.  The value of the check digit is determined by the value and positioning of the other digits: for any given code there is only one possible check digit.  When code has been entered, the check digit is calculated and compared to the entered value.  If the two digits do not match, an error is reported.
 

The example below of a check digit was copied from: http://www.advanced-ict.info/

The check digit is the digit in the units column (1). The actual data is the number from position 2 upwards.

To find the check digit to attach to a number perform the calculation as from position 2. The check digit is what must be added to the result to make it divisible by 11.

This shows validation of the number 83232. Note that the final 2 is the check digit. The number is valid because the final total is exactly divisible by 11.

Note that 10 is a possible value for the check digit. This is represented by an X. Use of a check digit is guaranteed to detect the two most common human errors in entering a long sequence of numbers - entering one digit incorrectly or transposing two digits. Used whenever a particular long number is to be used repeatedly - e.g. part number or account number. Also used to validate data captured by bar code reader. Validation is performed on input.

Note that check digits are used to validate numbers that are being used as identifiers rather than as numbers. There is no need to store the check digit within the computer once the input has been validated.

© www.teach-ict.com  All Rights Reserved