01 | ONE WRONG MARK, TWO DIFFERENT QUESTIONS

The paper result says a student scored 34 out of 50. A teacher types 43 into the school system.

The value 43 is a whole number, is present and lies between zero and fifty. It meets those input rules. However, it does not match the paper result.

This example separates two questions: “Does the input meet the acceptability rules?” and “Was it copied or entered as intended?”

Validation addresses the first question. Verification addresses the second. Both help reduce data problems, but they examine different properties.

This lesson brings the previous checks together so you can choose and explain them in a scenario rather than relying on the similarity of the two words.

02 | VALIDATION: COMPARE WITH A RULE

Validation checks data against specified acceptability rules. For a whole-number test mark from zero to fifty, type and range rules define what is acceptable.

The check does not need to consult the student’s paper result to reject 500. The range rule alone establishes that 500 cannot be a mark for this test.

Other rules can require a supplied identifier, an exact character count, a permitted character pattern or consistency with a check-digit calculation.

Validation can be carried out automatically, but automatic processing is not its defining feature. Its defining comparison is data against a rule.

A teacher manually checking that a mark is within zero to fifty is still checking a validation rule. A computer comparing two entries can be performing verification. Do not classify methods solely by whether a person or computer performs them.

03 | VERIFICATION: COMPARE WITH A SOURCE OR ANOTHER ENTRY

Verification checks that data has been copied or entered as intended. A visual check compares entered data with its source. Double entry compares two entries of the same intended information.

In the mark example, a careful visual comparison notices that 43 differs from source 34. The range rule does not reveal that particular copying error.

For double entry, the teacher might enter 43 first and 34 second. The computer detects that the entries differ, so they need checking or re-entry.

A mismatch does not establish which entry is correct. Equal entries do not necessarily establish that they match a source. The same wrong mark entered twice would still agree.

Verification therefore provides evidence about copying or entry, with limits. It does not establish every underlying fact about the data.

04 | PUT THE DISTINCTION SIDE BY SIDE

Use the comparison below to connect purpose, evidence and methods.

“Valid” means the value satisfies the checks performed. “Verified” should describe the actual comparison made, such as matching the paper source or agreeing between two entries.

Neither word should be used to mean “guaranteed true in every respect”. A system can check several properties and still receive a wrong but plausible value.

The important reasoning question is: what is being compared with what? A permitted range is a rule. The original paper mark is a source. Two independently entered values are a pair of entries.

05 | INVESTIGATE THREE ENTRIES FROM THE SAME SOURCE

Keep source mark 34 and the accepted whole-number range zero to fifty. Compare these possible entered values.

Entry 43 demonstrates why validation alone is insufficient for copying. Entry 500 demonstrates why a range rule can reject input without first knowing the correct mark.

Entry 34 satisfies the range and matches the source. If the source itself recorded an incorrect mark, both checks could still succeed. Copying accurately and describing the true situation are different concerns.

When explaining a result, state the specific rule and comparison rather than writing simply “it is correct”.

06 | RECOGNISE THE VALIDATION METHODS

A range check compares a value with limits. A length check checks the number of characters. A type check checks the required kind of data. A presence check checks that required information was supplied.

A format check checks a specified pattern, such as two uppercase letters followed by four digits. A check digit tests a calculated relationship between digits using an agreed method.

These checks establish different rules. A six-character input can still have the wrong format; a whole number can still fall outside the permitted range; a code can pass a check-digit rule despite some undetected errors.

Do not treat a check digit as double entry. It is calculated redundancy checked by a rule, rather than a second independent entry of the full intended data.

07 | RECOGNISE THE VERIFICATION METHODS

A visual check involves a person comparing entered information with the original or intended source. Reading the screen and deciding a number looks reasonable is not the same as comparing it with that source.

Double entry receives the same intended data twice and compares the entries. It does not automatically compare either entry with an independent paper or database record.

A system displaying a value for the user to confirm can support a visual check when the user compares it with the source. Merely selecting “Yes” without making that comparison gives little evidence.

Look at the method, not the interface. A second field labelled “Confirm identifier” usually supports double-entry comparison; a screen showing the source alongside the entered record supports visual comparison.

08 | A VALID PATTERN CAN IDENTIFY THE WRONG PERSON

A student code must contain two uppercase letters followed by four digits. Intended code AB1042 is mistyped as AB1402.

AB1402 has the required length and pattern. It passes those validation rules. A source comparison with AB1042 reveals the copying error.

If the first entry is AB1402 and the second is AB1042, double entry flags a mismatch. If both are AB1402, it reports agreement even though the intended code differs.

Also check the association between the value and the record. A correctly copied mark entered beside the wrong student can evade a check that looks only at the mark’s digits.

Different checks address different questions. The more precise your description of the error, the easier it is to suggest a suitable check.

09 | USE BOTH FOR A SCHOOL RESULTS FORM

Consider a form requiring a student code and a whole-number mark from zero to fifty. A sensible design can apply validation rules and use verification when transferring the source data.

Require both fields, check the identifier’s specified format, handle the mark as a whole number and test its range. Then compare the student association and entered mark with the paper source.

This is one possible workflow, not a rule that all systems must perform checks in exactly this order. However, safe input handling must establish a suitable representation before numeric comparisons.

If a check fails, communicate the problem and obtain a correction. Do not silently choose a plausible replacement mark and assume it is the intended value.

10 | SEE THE DIFFERENT COMPARISONS IN PSEUDOCODE

These two fragments illustrate different comparisons. Assume Mark is already a whole number. SourceMark is the mark obtained from the reference source.

IF Mark >= 0 AND Mark <= 50
  THEN
    OUTPUT "Range accepted"
  ELSE
    OUTPUT "Range rejected"
ENDIF

The first checks Mark against fixed limits. It establishes only whether the range rule is satisfied.

IF Mark = SourceMark
  THEN
    OUTPUT "Entered mark matches the source mark"
  ELSE
    OUTPUT "Entered mark differs from the source mark"
ENDIF

The second compares entered Mark with SourceMark. It illustrates the source-agreement question. In the syllabus’s visual check, a person makes that comparison; the fragment here makes the comparison’s meaning explicit rather than redefining a visual check as automatic code.

A double-entry routine instead compares FirstEntry with SecondEntry. Which operands are compared determines the claim that can be made.

11 | MATCH THE CHECK TO THE ERROR

A numeric value of 500 for a fifty-mark test is addressed by a range rule. Text “thirty-four” where the input routine requires an integer is addressed by type handling. A missing required mark is addressed by presence.

A valid-looking 43 copied instead of 34 is addressed by comparison with the source or potentially detected through independent double entry. It cannot reliably be rejected just by narrowing the range unless the real requirements justify that range.

A short code is addressed by length, and a misplaced letter by the relevant format rule. A format check might incidentally catch a copying mistake, but that does not turn it into verification: it still compares the entry with a pattern.

When asked to suggest a check, explain what it detects in the given scenario and mention any important remaining limit.

12 | AGREEMENT AND ACCEPTABILITY STILL HAVE LIMITS

If the paper mark is wrong and the computer faithfully copies it, a source comparison passes. If the same wrong code is entered twice, double entry passes.

A valid mark may belong to the wrong person. A valid identifier may have been selected intentionally but incorrectly. A check-digit rule can miss particular digit changes.

Passing several checks increases evidence about the properties examined, but it does not create an absolute guarantee of factual correctness.

This is not a reason to omit checks. It is a reason to understand their purpose, choose complementary methods and state conclusions accurately.

For example, “The input meets the range rule and matches the recorded source” is a stronger, clearer statement than “The system has proved the mark is true”.

13 | STRUCTURE A COMPARE ANSWER

The command word Compare asks you to identify similarities and/or differences. State both methods in relation to each other rather than giving unrelated notes.

A clear response is: “Validation checks data against acceptability rules, such as a permitted range, whereas verification checks that copying or entry was as intended, such as comparing with the source. Both can detect some input errors, but neither guarantees factual accuracy.”

For Explain, connect a method to a reason: “A range check accepts forty-three because it lies between zero and fifty, so it cannot detect that the source mark was thirty-four. A visual comparison with the paper can reveal that difference.”

Use the scenario’s exact values and rules as evidence. Avoid the shortcut “validation is automatic and verification is manual”, which fails for computer-performed double entry.

14 | APPLY THE DISTINCTION TO NEW SCENARIOS

An order form asks for a quantity from one to twenty. Range and type rules establish an acceptable numeric quantity. Comparing it with the customer’s written order checks the transfer of the intended quantity.

A user chooses a new contact address and enters it twice. Pattern rules check its structure; double entry checks agreement. Neither proves that messages can be delivered to that mailbox.

A library copies a book code from a label. Length, format and check-digit rules can test representation and consistency. Visual comparison can check that the correct label was copied.

In every case, ask the same two questions: what rules must the entry satisfy, and what evidence shows it agrees with the intended source or entry? You are now ready to test those behaviours with normal and abnormal data.