How an E-Learning Content Review Caught a Contradictory Assessment

One assessment question looked clear, and its answer key looked decisive. The problem was that the marked answer contradicted the process taught earlier in the module. The lesson told learners to verify the required information before moving a record to approval. The assessment rewarded the reverse order.

The error had survived because both versions appeared in the supplied material. One source showed the current sequence. Another retained an older sequence. If the assessment had been reviewed only for grammar, formatting, or difficulty, the contradiction could have reached learners.

This technical e-learning project changed how we checked the module. We stopped treating screens, interactions, and questions as separate pieces of copy. Instead, we treated the complete module as one connected learning system. That shift helped us resolve the source conflict, align every assessment item with an approved learning point, and catch the incorrect answer before delivery.

The source material did not contain one stable process

The module began with a familiar kind of input: presentation slides, process notes, reference material, and comments collected during earlier reviews. Each item contained useful information. Together, however, they did not form a clean script.

Some points appeared several times with small wording changes. Definitions varied between documents. One diagram presented the process as a simple sequence, while another inserted a verification step before approval. Reviewer comments clarified individual sentences but did not always state which source governed the complete workflow.

This is where technical e-learning differs from shortening a presentation. A learning module must decide what the learner should understand, practise, and remember. Repetition in the source does not automatically indicate importance. A neat diagram is not automatically current. A sentence approved in one context may become misleading when placed next to a different process stage.

Before writing, the content team therefore needs to answer three questions:

• Which source controls the process?
• Which statements describe the same idea, and which statements genuinely conflict?
• What must the learner be able to do after completing the module?

Without those answers, a writer can produce fluent screens while preserving contradictions that the source material never resolved.

Why the first screen-by-screen approach failed

The first approach followed the main presentation deck from beginning to end. That appeared sensible because the deck already had headings, diagrams, and a visible sequence. Each slide became a possible screen or interaction.

This method produced readable individual screens, but it had a structural weakness. It assumed that the deck was a single reliable source of truth. When supporting notes introduced a different term or process order, those differences were handled locally. The writer improved the sentence in front of them without testing its effect on the rest of the module.

The assessment was also drafted as a final task. Questions were written after the teaching screens, then checked for clarity and plausible distractors. That process tested whether each question worked on its own. It did not prove that the marked answer matched the exact instruction learners had received.

The result was local quality without complete continuity. The screens sounded correct. The questions sounded correct. The learning journey did not consistently teach and reward the same process.

We changed the method before completing the final draft. Source reconciliation had to come before screen writing, and assessment mapping had to begin before the module was considered stable.

We built a source-of-truth matrix

The source-of-truth matrix converted scattered reference material into an auditable content plan. It did not replace the approved sources. It showed how each source would be used and where a decision was still required.

For every important concept or process step, the matrix recorded:

• the proposed approved wording;
• the source location;
• alternative wording found elsewhere;
• any conflict that needed resolution;
• the learning objective supported by the point;
• the screen or interaction where it would be taught;
• the assessment item, if any, that would test it.

This made inconsistencies visible before they became polished copy. When two sources placed the same actions in a different order, the difference could no longer be treated as a minor edit. It became an explicit decision for the appropriate reviewer.

The matrix also prevented unnecessary rewriting. Once a technical term or process label had been approved, it remained stable across the introduction, explanation, interaction, feedback, and assessment. The surrounding explanation could be simplified, but controlled language did not drift from screen to screen.

Most importantly, every assessment question had to point back to a specific learning statement. If the team could not identify where the answer had been taught, the question was not ready. If the source did not support one unambiguous answer, the question needed revision or removal.

The module was organised around learning objectives

After the source conflicts were marked, the module was reorganised around learner actions rather than slide order. Each section had one clear purpose: introduce a concept, explain a process, let the learner apply it, or check understanding.

The workflow followed a connected sequence.

1. Define the learner action

Each learning objective described what the learner should be able to identify, distinguish, arrange, or apply. Broad objectives such as “understand the process” were replaced with observable actions. This gave the writer a clear test for relevance: if a detail did not support an objective, it needed a reason to remain.

2. Resolve the process before explaining it

Conflicting sequences were sent for clarification before the final narrative was written. The team did not choose the version that was easiest to explain. The correct sequence had to come from the approved source or reviewer.

3. Build instruction and practice together

Teaching screens, examples, interactions, and feedback were drafted as connected components. Practice used the same terminology and logic as the explanation. Feedback did more than say whether a response was correct. It restated the rule that governed the answer.

4. Map every assessment item

Each question was linked to a learning objective and the exact point in the module where the answer was taught. Distractors had to be plausible without introducing a second interpretation. The marked answer had to agree with both the approved process and the learner-facing explanation.

5. Review the complete learner path

The final content review followed the learner’s experience from instruction to practice, feedback, and assessment. This horizontal review was added to the usual screen-by-screen edit. It tested continuity rather than treating every content block as an independent item.

What the continuity review caught

The continuity review found that the module taught learners to verify the required information before moving a record to approval. An assessment scenario then asked what should happen next when a record was ready to proceed. The answer key marked approval before verification as correct.

The question was grammatically clear. Its options were distinct. The problem sat in the underlying logic. The answer key had inherited the sequence from an older source item, while the teaching section followed the approved process used elsewhere in the module.

We traced the question through the matrix, compared it with the learning objective and teaching screen, and corrected the marked answer. The feedback was also rewritten so it explained why verification came first. That mattered because changing the answer key alone would have fixed the score without reinforcing the intended behaviour.

The same review checked every other question against its source and teaching point. It also confirmed that process terms remained consistent and that no interaction introduced a shortcut the module had not approved.

This was the specific value of review on the project. It did not merely improve style. It prevented the module from teaching one action and rewarding another.

What this changes about e-learning content review

E-learning quality cannot be judged by reading screens in isolation. A module is correct only when its parts agree. The objective, explanation, example, interaction, feedback, and assessment must all point to the same knowledge or behaviour.

For similar technical learning work, check the following before delivery:

• Is there one approved source for every process step and technical term?
• Have contradictions between source files been resolved rather than edited around?
• Does every screen support a defined learning objective?
• Is every assessed answer taught clearly in the module?
• Do feedback messages explain the same rule used in the lesson?
• Are controlled terms consistent across narration, on-screen copy, interactions, and questions?
• Has someone reviewed the learner path by concept and process, not only by screen?
• Can every correction be traced to an approved source or reviewer decision?

The main lesson is simple. Good e-learning writing is not the production of polished screens. It is the construction of a consistent learning path. Source reconciliation establishes what is correct. Learning design decides what must be taught. Continuity review proves that the module teaches, practises, and assesses the same thing.

That is how a small answer-key error becomes useful evidence about the whole production method. The catch showed where the original workflow was weak, and the revised workflow made the content easier to verify before delivery.

FAQs

What was actually wrong with the assessment?

One question looked clear and its answer key looked decisive, but the marked answer contradicted the process taught earlier in the module. The lesson told learners to verify required information before moving a record to approval; the assessment rewarded the reverse order.

How did an error like that survive review?

Because both versions appeared in the supplied material. One source showed the current sequence and another retained an older one. Reviewing the assessment only for grammar, formatting or difficulty would not have caught it, because each item was correct in isolation.

What is a source-of-truth matrix?

A record that converts scattered reference material into an auditable content plan. For every important concept it captures the proposed approved wording, the source location, alternative wording found elsewhere, any conflict needing resolution, the learning objective it supports, the screen that teaches it, and the assessment item that tests it.

Why did the screen-by-screen approach fail?

It assumed the main presentation deck was a single reliable source of truth. When supporting notes introduced a different term or process order, the writer improved the sentence in front of them without testing its effect on the rest of the module. The result was local quality without continuity.

When should assessment items be written?

Before the module is considered stable, not as a final task. Drafting questions last tests whether each one works on its own; it does not prove the marked answer matches the instruction learners actually received.