THIRD EXAMINER · MANUSCRIPT AUDIT
Shear bond strength of a bulk-fill resin composite to dentin using four adhesive protocols: an in-vitro study
C59%
47 / 80 points
Major revision
Not ready to submit. Three findings block publication; each is fixable without new experiments.
Three things that block publication
1
C is not above the control: A 18.97 vs C 18.33 MPa, Tukey p = 0.925.
M1 · T04
2
Group B mean 12.58 MPa can't be computed from the raw data; the true mean is 15.28.
M2 · D02, H1
3
Methods say n = 10 per group; the data hold 9 (36 in total), with no exclusions reported.
M3 · INV-1
Scorecard
eight areas, each out of 10Ethics
8
Approval and tooth source stated; donor consent wording missing.
References
7
[14] cited for a claim it doesn't make; [22] details don't match its DOI.
Language
7
Clear English; results overstate, two passages read machine-written.
Methods reporting
6
Specimen count misstated (10 vs 9); exclusions not reported.
Novelty
5
Well-studied comparison; the bulk-fill pairing adds a modest increment.
Statistics
5
Right tests; checks unreported, one SE given as an SD.
Data plausibility
5
Three of four group means recompute; group B does not.
Internal consistency
4
Headline claim contradicts its own table; n contradicts F(3, 32).
Total 47 / 80 = 59%, grade C.A = ready with light polish · F = not defensible as written
Reported vs recomputed: 13 numbers checked
7 MATCH
4 MISMATCH
2 ADD
Full table on page 2: every reported value beside the value recomputed from the raw sheets.
Reported vs recomputed
| QUANTITY | REPORTED | RECOMPUTED | STATUS |
|---|---|---|---|
| Specimens per group | 10 | 9 | MISMATCH |
| Mean, group A (MPa) | 18.97 | 18.97 | MATCH |
| Mean, group B (MPa) | 12.58 | 15.28 | MISMATCH |
| Mean, group C (MPa) | 18.33 | 18.33 | MATCH |
| Mean, group D (MPa) | 11.08 | 11.08 | MATCH |
| SD, group A (MPa) | 1.45 | 1.45 | MATCH |
| SD, group B (MPa) | 2.43 | 2.43 | MATCH |
| SD, group C (MPa) | 2.28 | 2.28 | MATCH |
| SD, group D (MPa) | 0.82 | 2.46 | MISMATCH |
| One-way ANOVA | F(3, 32) = 24.21 | F = 24.21, p < 0.001 | MATCH |
| A vs C (Tukey) | C significantly above A | p = 0.925, not significant | MISMATCH |
| Normality (Shapiro–Wilk) | Not reported | lowest p = 0.395 | ADD |
| Equal variances (Levene) | Not reported | p = 0.647 | ADD |
Scorecard: the reason behind each score
| AREA | SCORE | REASON |
|---|---|---|
| Ethics | 8 / 10 | Approval and the source of the teeth are stated. Donor consent wording is missing. |
| References | 7 / 10 | Most references support their sentences. [14] is cited for a claim it does not make; [22] has details that do not match its DOI record. |
| Language | 7 / 10 | Clear and mostly correct English. The results chapter states conclusions more strongly than the data allow, two passages read as machine-written, and two Discussion sentences follow [7] without quotation marks. |
| Methods reporting | 6 / 10 | Protocols and thermocycling are clear. The specimen count is misstated (10 vs 9) and exclusions are not reported, as CRIS requires. |
| Novelty | 5 / 10 | Comparing universal-adhesive modes with an etch-and-rinse control is well studied; pairing them with a bulk-fill composite adds a modest increment. |
| Statistics | 5 / 10 | The right test and post-hoc. Assumption checks are not reported, one SE is given as an SD, and the conclusion contradicts the Tukey result. |
| Data plausibility | 5 / 10 | Three of the four group means recompute exactly from the raw sheets; group B does not (12.58 typed, 15.28 true). |
| Internal consistency | 4 / 10 | The headline claim contradicts its own table, and the stated n contradicts both the sheets and the degrees of freedom, F(3, 32). |
Fix roadmap, in order
- 1Rewrite the conclusion and the abstract: C is comparable to the control, not superior.Author Tukey A vs C, p = 0.925; A has the higher mean.
- 2Correct the group B mean to 15.28 in Table 4.2, the text and the abstract.Author It is the mean of the raw data; the typed value cannot be reproduced.
- 3State n = 9 per group; list each excluded specimen and why.Author + supervisor CRIS asks for the sample size and every exclusion.
- 4Report the group D SD as 2.46.Author 0.82 is the standard error.
- 5Add the normality and equal-variance results to methods and results.Author Reviewers ask for them; both pass, so this only strengthens the paper.
- 6Reword the sentence cited to [14]; correct the details of [22].Author Each reference must support its sentence and match its DOI record.
- 7Quote or rewrite the two sentences that follow [7], and rewrite the two passages that read as machine-written.Author Journals run a similarity check on every submission.
- 8Read the results chapter against every table once more.Supervisor Two of the three blockers are mismatches between text and tables.