Skip to content
medlit.
FICTIONAL EXAMPLE · NO KEY NEEDED
Hand-authored teaching material, not real research or an AI-generated result. Uses MedLit’s real quotation checks. Nothing is added to your library.

Fictional example: a reading app and self-reported concentration

MedLit teaching material · not a published study
Randomized Controlled Trial
RoB 2-aligned structured questions
Fictional excerpt
Appraisal confidence

A score you can look inside.

9
/ 10
Source coverage, matched quotations, and assessable questions. Not a measure of a paper’s truth, accuracy, or clinical usefulness.
SOURCE AVAILABLE
Teaching excerpt
Review source limitations
QUOTATIONS MATCHED
5 of 6
Text match ≠ scientific validation
DOMAINS UNCLEAR
1 of 5
Missing information stays visible
This hand-authored example uses the real quotation validator. Neither its narrative nor a matched quote establishes scientific accuracy.
APPRAISAL OVERVIEW
Appraisal support (heuristic): higher
Applicability: not assessed
4 of 5 domains have determinate judgments supported by matched passages; 1 remain unclear. Text matching does not establish that a passage supports the model's interpretation.
Review the domain judgments and their source passages before relying on the generated narrative.
Key caveats before using or citing this appraisal:
Fictional, hand-authored teaching example. No provider was called and no scientific finding is established.
This is an abbreviated excerpt without effect estimates, protocol, reference list, or funding disclosures.
One or more generated quotations could not be matched to the supplied article text and were excluded from determinate judgments.
Full detail
A fictional reading-app study shows why a randomized design alone does not settle the risk of bias.
What they did
•
Assigned 120 university adults to a reading app or usual study practice for four weeks.
What they found
•
The excerpt reports 24 missing outcomes and does not provide an effect estimate.
Why it matters
Missing outcomes and unblinded self-report can complicate interpretation, even with random allocation.
Bottom line
Inspect the bias domains, compare their passages, and notice the unsupported reporting claim being marked unclear.
For: Anyone practicing critical appraisal. This example does not support a real health or learning decision.
Readability Metrics
Moderate
Reading Level
Grade 11
F-K Grade
~1 min
Reading Time
73
Words
Flesch Reading Ease: 42.5/100 (moderate difficulty)
Structured from the supplied article text. “Not reported” means MedLit did not verify enough source evidence; it is not proof the paper omitted the item.
Population
identified
120 adults from one university
1 verified passage
Intervention / exposure
identified
A reading app
1 verified passage
Comparator
identified
Usual study practice
1 verified passage
Outcomes
identified
Self-reported concentration score
1 verified passage
Timeframe
identified
Four weeks
1 verified passage
This heuristic combines source coverage, matched quotations, and assessable domains. It is not a probability that the findings are correct or a validated quality rating.
Strengths
✓
Randomization process: The example reports a generated random sequence and central concealment. These passages support this limited judgment; they do not certify implementation.
Weaknesses
⚠
Deviations from intended interventions: Participants knew their assignment and the analysis excluded non-completers. This raises concerns about departures from assigned intervention.
⚠
Missing outcome data: Twenty percent of outcomes are missing. Without reasons or group counts, the impact of missingness cannot be established.
⚠
Outcome measurement: The outcome is self-reported by participants who knew their assignment. Expectations could influence reporting.
⚠
Selection of the reported result: Teaching example: a deliberately unsupported preregistration quotation is excluded by the validator. Selective reporting therefore remains unclear.