Chapter 6.2: Data Integrity

Validation & Verification Methods

9618 AS Computer Science

📚 Learning Objectives

Describe how data validation and data verification help protect integrity of data
Describe and use methods of data validation including range check, format check, length check, presence check, existence check, limit check and check digit
Describe and use methods of data verification during data entry and data transfer
Understand verification methods including visual check, double entry, parity check, checksum, and ARQ

🌟 What is Data Integrity?

Data Integrity refers to the accuracy, completeness, and consistency of data. It ensures that:

Data received is the same as data sent
Data copied is the same as the original
Data is correct and not corrupted

Data integrity can be compromised during data entry, data transmission, malicious attacks (malware, hacking), or accidental data loss (hardware issues).

1. Validation

📖 Definition

Validation is an automatic computer check to ensure that the data entered is sensible and reasonable. It does NOT check the accuracy of data.

Example: A secondary school student is likely to be aged between 11 and 16. The computer can be programmed only to accept numbers between 11 and 16. This is a range check. However, if a student's age is 14 but 11 is entered, it will be valid but incorrect.

1.1 Key Point

Validation ensures data is sensible and reasonable, but cannot guarantee it is correct. Validation is automatically carried out by computer software.

💡 Exam Tip

Remember: Validation checks if data is reasonable (within acceptable criteria), NOT if it is accurate (correct). For example, if you enter your age as 62 instead of 26, it is reasonable but not accurate.

1.2 Types of Validation

Validation Type	Description	Example
Range Check	Ensures a number falls within a set range	Percentage must be 0-100
Limit Check	Checks that a value does not exceed a maximum/minimum limit	Maximum 5 items per order
Length Check	Checks the length of a string	PIN must be exactly 4 digits
Type Check	Checks that input is correct data type	Age must be a whole number
Presence Check	Ensures field is not left blank	Registration form requires name
Existence Check	Checks that a referenced value exists in database	Student ID must exist in records
Format Check	Ensures data matches a pre-defined pattern	Email must contain @ and domain
Check Digit	Final digit calculated from other digits to detect errors	ISBN, barcodes

2. Validation Types in Detail

2.1 Range Check

📝 How It Works

Checks that only numbers within a specified range are accepted. For example, month in a date must not exceed 12, or a percentage must be between 0 and 100.

2.2 Limit Check

📝 How It Works

Checks only one limit - either upper limit or lower limit. For example, a maximum number of years for a person's age, or minimum order quantity.

2.3 Length Check

📝 How It Works

Checks that data contains an exact number or range of characters. For example, a telephone number must be 10-11 digits, or a PIN must be exactly 4 characters.

2.4 Format Check

📝 How It Works

Checks that characters entered conform to a pre-defined pattern. For example:

Date format: dd/mm/yyyy
Email: must contain @ and domain like .com
Postcode: specific format like "SW1A 1AA"

2.5 Presence Check

📝 How It Works

Ensures that an entry field is not left blank. For example, making sure a registration form is not submitted with blank required fields like name or email.

2.6 Existence Check

📝 How It Works

Checks that a referenced value exists in a database or list. For example, confirming a student ID entered exists in the school records, or a filename referred to in data entry actually exists.

3. Check Digits

📖 Definition

A check digit is the final digit included in a code; it is calculated from all other digits in the code. When a number is entered into a computer, the check digit is recalculated and if the same value is not generated, an error will occur.

3.1 Applications of Check Digits

ISBN (International Standard Book Numbers)
Barcodes (EAN, UPC codes)
VIN (Vehicle Identification Numbers)
Bank card numbers

3.2 Types of Errors Detected

Error Type	Example
Incorrect digit	5327 entered instead of 5307
Phonetic errors	13 (thirteen) instead of 30 (thirty)
Transposition error	4087 instead of 4807 (two digits swapped)
Missing/Extra digit	247 instead of 2407, or 42107 instead of 4207

3.3 ISBN Check Digit Calculation Example

📝 10-Digit ISBN Calculation: 1 84146 201 ?

Step 1: Each digit is given a weighting from 10 down to 1 (left to right)

Step 2: Multiply each digit by its weighting and sum:

(1×10) + (8×9) + (4×8) + (1×7) + (4×6) + (6×5) + (2×4) + (0×3) + (1×2)

= 10 + 72 + 32 + 7 + 24 + 30 + 8 + 0 + 2 = 185

Step 3: Divide total by 11: 185 ÷ 11 = 16 remainder 9

Step 4: Subtract remainder from 11: 11 - 9 = 2

Check digit = 2

Note: If remainder is 10, letter X is used as check digit.

3.4 EAN8 Barcode Check Digit Example

📝 EAN8 Barcode: 2142345?

Step 1: Multiply 1st, 3rd, 5th, 7th digits by 3:

(3×2) + (3×4) + (3×3) + (3×5) = 6 + 12 + 9 + 15 = 42

Step 2: Add remaining digits: 42 + 1 + 2 + 4 = 49

Step 3: Divide by 10: 49 ÷ 10 = 4 remainder 9

Step 4: Subtract from 10: 10 - 9 = 1

Check digit = 1 → Full barcode: 21423451

4. Verification

📖 Definition

Verification is checking that data has been accurately copied onto a computer or transferred from one computer system to another. Verification ensures data matches the original source.

⚠️ Validation vs Verification

Validation: Checks if data is sensible and reasonable (automatic)
Verification: Checks if data is accurate and matches source

4.1 Verification During Data Entry

Method	Description	Limitations
Double Entry	Data is entered twice; computer compares both entries and outputs error if different	Time-consuming; same mistake entered twice won't be detected
Visual Check	User manually compares input data against original source document on screen	Prone to human error; relies on user attention
Check Digits	Final digit calculated from other digits; computer recalculates to verify	Only works for numeric codes; requires algorithm

4.2 Double Entry Example

Password Creation: Users are asked to enter their chosen password twice. This verifies that the user typed it correctly. This check is important because password entry fields are normally asterisked out (***********), so users cannot see what they typed.

5. Verification During Data Transfer

When data is transferred electronically from one device to another, there is possibility of data corruption or even data loss. Several methods exist to minimize this risk.

5.1 Checksums

📖 Definition

A checksum is a value transmitted at the end of a block of data, calculated using the other elements in the data stream, used to check for transmission errors.

📝 How Checksums Work

Sending computer uses data and a predefined mathematical algorithm to calculate a checksum value
Sending computer sends the data plus the checksum value
Receiving computer uses the data received to calculate what it believes the checksum should be
The two checksum values are compared
If values match → data transmitted without errors
If values differ → request is sent for data to be retransmitted

💡 Checksum Calculation Rule

If sum of other bytes is 255 or less, checksum = that exact value
If sum of other bytes is more than 255, checksum = remainder of total ÷ 256

Example: Bytes total 1,151
1,151 ÷ 256 = 4 remainder 127
Checksum = 127

6. Parity Checks

📖 Definition

A parity check is a method to check whether data has been changed or corrupted following transmission. An extra bit (parity bit) is added to a string of binary code to ensure the number of 1-bits are either even or odd.

6.1 Even vs Odd Parity

Parity Type	Rule	Parity Bit Value
Even Parity	Total number of 1-bits (including parity bit) must be EVEN	Set to make total even
Odd Parity	Total number of 1-bits (including parity bit) must be ODD	Set to make total odd

📝 How Parity Checks Work

Sending and receiving computers agree the protocol (even or odd parity)
Sending computer adds the correct parity bit to the binary data (either 1 or 0)
Sending computer sends the binary data including parity bit
Receiving computer checks that the overall parity matches the agreed type
If parity is incorrect → receiving computer requests data be transmitted again

6.2 Parity Examples

Even Parity Example:
Data bits: 1 0 1 1 0 1 0 (four 1s = even)
Parity bit must be 0 to keep total even
Byte transmitted: 0 1 0 1 1 0 1 0

Odd Parity Example:
Data bits: 1 0 1 1 0 1 0 (four 1s = even)
Parity bit must be 1 to make total odd
Byte transmitted: 1 1 0 1 1 0 1 0

❌ Limitation of Parity Checks

If there are multiple errors in the same byte that still produce the same parity, the error will NOT be detected!

Example: "00001111" changes to "00000011" - both have even parity (four 1s and two 1s are both even... wait, 00001111 has 4 ones, 00000011 has 2 ones - both even!)

Actually: Original 00001111 has 4 ones (even), corrupted 00000011 has 2 ones (even) - parity unchanged, error undetected!

7. Parity Block Checks

📖 Definition

Parity blocks and parity bytes can be used to check an error has occurred AND where the error is located. Parity checks on their own only detect that an error occurred, not where.

📝 How Parity Blocks Work

Data is arranged in a block of rows and columns
Each row has a parity bit (horizontal parity)
Each column has a parity bit in a parity byte (vertical parity)
By cross-referencing horizontal and vertical parity values, the exact position of an error can be pinpointed
The error can then be automatically corrected or a retransmission requested

7.1 Parity Block Example (Odd Parity)

8. Automatic Repeat Request (ARQ)

📖 Definition

ARQ (Automatic Repeat Request) is a form of error detection that uses a system of acknowledgements and timeouts. ARQ is often used to ensure reliable transmissions over an unreliable service.

📝 How ARQ Works

Sending computer transmits a block of data
Sending computer waits a period of time to see if receiving computer acknowledges receipt
If acknowledgement received → transmission successful
If no acknowledgement after set period (timeout) → data is automatically resent
This continues until the receiving computer acknowledges the data has been received

8.1 Key Components

Component	Description
Acknowledgement (ACK)	Signal sent by receiver to confirm data was received correctly
Timeout	Set period of time after which data is resent if no acknowledgement received
Retransmission	Automatic resending of data when timeout occurs without acknowledgement

💡 Exam Tip

ARQ combines well with other error detection methods. Data is first checked using parity or checksum, then if an error is found, no ACK is sent → timeout occurs → data is resent automatically.

9. Data Security vs Data Integrity

Data Security	Data Integrity
Deals with protection of data	Deals with validity of data
Protects data from illegal access/loss	Ensures data is not corrupted after input or transmission
Methods: Encryption, firewalls, passwords, access control	Methods: Validation, verification, checksums, parity checks
Prevents unauthorized users from accessing data	Ensures data accuracy and consistency

10. Key Takeaways

📌 Summary Points

Validation

Definition: Automatic check for sensible/reasonable data
Types: Range, limit, length, type, presence, existence, format, check digit
Key point: Does NOT check accuracy, only reasonableness

Verification During Data Entry

Double Entry: Enter twice, compare for match
Visual Check: Manual comparison with original
Check Digits: Calculated digit for error detection

Verification During Data Transfer

Checksum: Value calculated from data block to detect corruption
Parity Check: Extra bit to ensure even/odd number of 1s
Parity Block: Horizontal + vertical parity for error location
ARQ: Acknowledgements + timeouts for reliable transmission

11. Glossary

Term	Definition
Data Integrity	The accuracy, completeness, and consistency of data; ensures data is correct and not corrupted
Validation	Automatic computer check to ensure data entered is sensible and reasonable
Verification	Checking that data has been accurately copied or transferred, matching the original source
Range Check	Validation that ensures a number falls within a specified minimum and maximum range
Limit Check	Validation that checks a value against one limit (upper or lower) only
Length Check	Validation that checks the number of characters in data entry
Format Check	Validation that ensures data conforms to a pre-defined pattern
Presence Check	Validation that ensures a field is not left blank/empty
Existence Check	Validation that confirms a referenced value exists in a database or file system
Type Check	Validation that ensures the correct data type is entered
Check Digit	Final digit in a code calculated from other digits to detect entry errors
Double Entry	Verification method where data is entered twice and compared for match
Visual Check	Verification method where user manually compares input with original source
Checksum	Value transmitted with data block, calculated from data, used to detect transmission errors
Parity Bit	Extra bit added to binary data to make total number of 1s even or odd
Even Parity	Parity system where total number of 1-bits (including parity bit) must be even
Odd Parity	Parity system where total number of 1-bits (including parity bit) must be odd
Parity Block	Grid of data with horizontal and vertical parity bits to locate errors
Parity Byte	Additional byte containing vertical parity bits for each column in a parity block
ARQ	Automatic Repeat Request - error control using acknowledgements and timeouts
Acknowledgement	Signal sent by receiver to confirm successful data receipt
Timeout	Set period after which data is resent if no acknowledgement received
Transposition Error	Error where two adjacent digits are swapped (e.g., 4087 instead of 4807)
Phonetic Error	Error caused by similar-sounding numbers (e.g., 13 instead of 30)

12. Exam-Style Questions (Part 1)

1. Describe the difference between validation and verification. Give an example of each. [4 marks]

Answer:

Validation is an automatic check to ensure data is sensible and reasonable, but does NOT check accuracy
Verification checks that data accurately matches the original source
Validation example: Range check ensuring age is between 11-16 for a secondary school student
Verification example: Double entry of password to ensure it was typed correctly
Additional point: Validation is performed by computer software automatically; verification can involve user interaction

2. A school is creating a database of student information. Describe three different validation checks that could be used on the data entry form. [6 marks]

Answer:

Presence check on name field to ensure it is not left blank
Range check on age field (e.g., 11-18) to ensure reasonable age for student
Format check on email field to ensure it contains @ and a valid domain
Length check on phone number to ensure correct number of digits
Type check on grade field to ensure only letters (A-F) are entered
Limit check on maximum number of subjects a student can select
Additional point: Existence check to verify student ID exists when referencing another record

3. Explain how a checksum is used to verify data during transmission. [5 marks]

Answer:

Sending computer calculates a checksum value using a mathematical algorithm on the data block
The checksum value is transmitted along with the data
Receiving computer recalculates the checksum from received data using same algorithm
The two checksum values are compared
If values match, data was transmitted without error; if different, retransmission is requested
Additional point: Checksum is calculated by summing bytes (if >255, use remainder when divided by 256)

4. A barcode has the first seven digits: 5 0 1 2 3 4 6. Calculate the check digit using the EAN8 method. Show your working. [4 marks]

Answer:

Step 1: Multiply 1st, 3rd, 5th, 7th digits by 3: (3×5) + (3×1) + (3×3) + (3×6) = 15 + 3 + 9 + 18 = 45
Step 2: Add remaining digits: 45 + 0 + 2 + 4 = 51
Step 3: Divide by 10: 51 ÷ 10 = 5 remainder 1
Step 4: Subtract from 10: 10 - 1 = 9
Check digit = 9
Additional: Full barcode would be: 50123469

5. Explain the purpose of a parity bit and describe the difference between even parity and odd parity. [4 marks]

Answer:

A parity bit is an extra bit added to binary data to help detect transmission errors
Even parity: The parity bit is set so that the total number of 1-bits (including parity) is an even number
Odd parity: The parity bit is set so that the total number of 1-bits (including parity) is an odd number
Sender and receiver must agree on parity type before transmission begins
Additional point: If received data has wrong parity, an error is detected and retransmission is requested

12. Exam-Style Questions (Part 2)

6. A byte using even parity is transmitted as: 1 0 1 1 0 1 0 0. Explain whether an error has been detected if this byte is received. [3 marks]

Answer:

Count the number of 1-bits in the received byte: 1, 0, 1, 1, 0, 1, 0, 0 = four 1s
Four is an even number, which matches the even parity requirement
No error detected - the parity check passes
Additional point: However, if two bits were corrupted (swapped), the parity would still be even and the error would NOT be detected - this is a limitation of parity checks

7. Describe how a parity block check can be used to both detect and correct an error in transmitted data. [5 marks]

Answer:

Data is arranged in a block of rows and columns
Each row has a horizontal parity bit calculated and transmitted
Each column has a vertical parity bit in a parity byte
If an error occurs, the row with wrong parity and column with wrong parity are identified
The intersection of the row and column pinpoints the exact location of the error
The bit at that location can be automatically corrected (flipped from 0 to 1 or 1 to 0)
Additional point: This method can only correct single-bit errors; multiple errors may not be locatable

8. Explain how Automatic Repeat Request (ARQ) ensures reliable data transmission. [4 marks]

Answer:

Sending computer transmits a block of data
Sender waits for an acknowledgement (ACK) from the receiver
If ACK is received within the timeout period, transmission is confirmed successful
If no ACK is received before timeout expires, data is automatically resent
Process continues until receiver acknowledges successful receipt
Additional point: ARQ works with error detection methods - if parity/checksum fails, no ACK is sent, triggering retransmission

9. Describe three types of errors that a check digit can detect. Give an example of each. [6 marks]

Answer:

Incorrect digit: A single wrong digit is entered (e.g., 5327 instead of 5307)
Transposition error: Two adjacent digits are swapped (e.g., 4087 instead of 4807)
Phonetic error: Numbers that sound similar are confused (e.g., 13 "thirteen" instead of 30 "thirty")
Missing digit: A digit is omitted (e.g., 247 instead of 2407)
Extra digit: An additional digit is added (e.g., 42107 instead of 4207)
Additional point: Check digits are used in ISBNs, barcodes, bank cards, and VIN numbers

10. Compare double entry and visual check as methods of verification during data entry. Include advantages and disadvantages of each. [6 marks]

Answer:

Double Entry: Data is entered twice and computer compares both entries
Advantage: Automated checking, no human error in comparison
Disadvantage: Time-consuming; same error entered twice won't be detected
Visual Check: User manually compares entered data with original source on screen
Advantage: Quick to perform; good for small amounts of data
Disadvantage: Relies on human attention; prone to tiredness/errors
Additional point: Double entry is better for critical data like passwords; visual check is better for occasional verification

13. Exam Success Tips (Part 1)

🧠 Memory Trick: Validation vs Verification

VAlidation = VAlue - Checks if VALUE is reasonable/sensible
VErification = VErify - Checks if data matches source (VERIFIED)
Think: "Is it VALID?" vs "Has it been VERIFIED?"
Validation = Automatic (computer does it)
Verification = Can involve user interaction

💡 Validation Types - Remember "RFLPECT"

Range - Numbers within min/max
Format - Pattern matching (dd/mm/yyyy)
Length - Exact or range of characters
Presence - Not blank
Existence - Exists in database
Check digit - Calculated verification digit
Type - Correct data type

💡 Parity Rules - Easy Memory

EVEN parity: Total 1s must be EVEN (2, 4, 6, 8...)
ODD parity: Total 1s must be ODD (1, 3, 5, 7...)
Parity bit is set to MAKE the total correct
If data already has correct parity → parity bit = 0
If data needs adjustment → parity bit = 1

💡 Checksum Calculation Rule

If sum ≤ 255: Checksum = exact sum value
If sum > 255: Checksum = remainder when sum ÷ 256
Example: Sum = 1,151 → 1,151 ÷ 256 = 4 r 127 → Checksum = 127

❌ Common Mistakes to Avoid

Don't say validation checks "accuracy" - it only checks reasonableness!
Don't confuse range check (min AND max) with limit check (one limit only)
Don't forget: Parity checks can miss multiple errors that maintain parity
Don't mix up even parity and odd parity in your answer
Don't forget to show working in check digit calculations

13. Exam Success Tips (Part 2)

🧠 ARQ Memory Trick

A - Acknowledgement sent by receiver
R - Retransmit if no ACK received
Q - Quality assured through timeout mechanism
Think: "Always Require Quality" for reliable transmission

💡 Answer Structure Tips

For "describe" questions: Give step-by-step details of how something works
For "explain" questions: Give reasons WHY something happens
For "compare" questions: Use table format and mention BOTH items
For "calculate" questions: Always show your working step by step
Check mark allocation: [4 marks] = 4 distinct points needed

⚠️ Must-Know Calculations

EAN8 Check Digit:

Multiply positions 1, 3, 5, 7 by 3
Add remaining positions (2, 4, 6)
Divide by 10, get remainder
Subtract remainder from 10

ISBN-10 Check Digit:

Multiply each digit by weight (10 down to 1)
Sum all products
Divide by 11, get remainder
Subtract remainder from 11 (if 10, use X)

🌟 Quick Reference Table

Topic	Key Point
Validation	Checks reasonableness, NOT accuracy
Verification	Confirms data matches original source
Parity Check	Extra bit for even/odd count of 1s
Parity Block	Locates exact error position
Checksum	Value from algorithm to detect corruption
ARQ	ACK + Timeout = Reliable transmission
Check Digit	Final digit calculated from code

13. Exam Success Tips (Part 3)

📌 Final Exam Reminders

Read questions carefully - note the command word (describe, explain, compare, calculate)
Check mark allocations - they tell you how many points to make
Use diagrams where helpful - they can earn marks and clarify answers
Show all working in calculation questions - marks for method even if answer wrong
Use technical terminology: parity bit, checksum, acknowledgement, timeout
Time management - don't spend too long on one question
Review your answers if time permits

💡 When to Use Each Method

Situation	Best Method(s)
Ensuring password typed correctly	Double entry verification
Checking age is reasonable	Range check validation
Detecting transmission errors	Checksum, Parity check, or ARQ
Finding exact error location	Parity block check
Verifying barcode correctness	Check digit
Ensuring field not empty	Presence check validation
Reliable data transmission	ARQ + Checksum/Parity

❌ Last-Minute Warnings

Parity check limitation: Two errors in same byte can go undetected
Double entry limitation: Same error twice won't be caught
Validation limitation: Cannot detect reasonable but wrong data (age 62 vs 26)
Range check requires both min AND max; limit check uses only one
Always state whether using even or odd parity in answers