Dentistry Thesis Statistics
Dentistry theses need slightly different statistics from other health fields: many studies compare the two halves of the mouth (split-mouth) or several teeth/implants within the same patient, so the measurements aren't independent; clinical indices are measured repeatedly over time; and how consistently two clinicians rate the same case (examiner agreement) has to be reported separately. GetBayes takes your data and research questions, runs the tests that correctly model this dependency structure (paired comparisons, repeated measures, survival analysis, kappa/ICC agreement coefficients), and reports them in the format dental journals expect, with effect sizes and confidence intervals — the analysis itself usually takes 15 minutes, and your report is in your hands the same day, often within hours or even minutes.
For implant and restorative studies we use Kaplan-Meier and log-rank for success/survival rates; repeated-measures analyses for periodontal indices (plaque index, gingival index, probing depth, clinical attachment level); and often nonparametric or count-data models for caries counts like DMFT/dmft. When the sample is small (a single clinic, a specific patient group) we carefully check the normality assumption and switch to nonparametric alternatives when needed.
Who is this service for?
Research assistants working on a dentistry specialty (residency) thesis
PhD and master's students in dentistry
Clinicians running studies in implantology, periodontology, orthodontics, restorative or endodontics
Researchers doing in-vitro work (bond strength, microleakage, material comparison)
Researchers whose manuscript to a dental journal came back with a statistics revision
The analyses we run most in dentistry research
| Study type / question | Typical analyses |
|---|---|
| Implant/restoration success and survival | Kaplan-Meier survival curve, log-rank test, Cox regression (for risk factors) |
| Split-mouth design (two sides in one patient) | Paired-samples t-test/Wilcoxon, repeated-measures or mixed model |
| Agreement between two clinicians/examiners | Cohen's kappa (categorical), intraclass correlation (ICC) and Bland-Altman (continuous) |
| Change in periodontal indices over time | Repeated-measures ANOVA or Friedman test + post-hoc comparisons |
| Caries experience (DMFT/dmft) comparison | Mann-Whitney/Kruskal-Wallis or count-data models (Poisson/negative binomial) |
| Material/technique comparison (in-vitro) | One-/two-way ANOVA + Tukey, or a nonparametric alternative when needed |
How the process works
- 01
Send your data and design
Share your measurement data and study design (split-mouth or single-group, how many time points) — Excel, a Google Forms export or an SPSS file all work.
- 02
Free assessment
Within 24 hours we tell you which analyses are needed, flag anything that needs cleaning in the data, and give a clear written price.
- 03
Analysis and reporting
Test selection matched to the dependency structure, normality checks, examiner agreement, and APA-formatted tables with effect sizes.
- 04
Delivery and support
Your publication-ready report arrives the same day; we keep answering questions ahead of your advisor meeting or defense.
Common mistakes in dentistry theses
These are what committees and reviewers flag most — we account for them from the start:
Analyzing split-mouth data or multiple teeth/implants from one patient as independent observations — measurements from the same patient are dependent and need a paired or mixed model.
Never reporting examiner agreement (kappa/ICC) — dental journals want to see reproducibility of clinical measurements.
Reporting implant/restoration success as a simple percentage without accounting for follow-up time — survival analysis (Kaplan-Meier) handles differing follow-up durations correctly.
Applying a test that assumes a normal distribution to right-skewed count data like DMFT — these data usually need a nonparametric or count-data model.
Frequently asked questions
How do you analyze my split-mouth study?
Measurements from the two sides of the same patient are dependent, so instead of an independent-samples test we use a paired-samples t-test, or a Wilcoxon signed-rank test if normality doesn't hold. With multiple time points or multiple sites, we fit a repeated-measures or mixed-effects model.
Two clinicians rated the same cases — how do you report agreement?
If the rating is categorical (e.g. present/absent, a classification) we report Cohen's kappa; if it's a continuous measurement (e.g. probing depth, a cephalometric distance) we report the intraclass correlation coefficient (ICC) and a Bland-Altman agreement plot. We can also report intra-examiner reliability from the same clinician's repeated measurements.
How many cases do I need for an implant survival analysis?
The exact number depends on the expected failure rate and follow-up duration; usually the number of events (failures) matters more than the total number of implants. We can size the study using the logic on our sample size calculation page and prepare the ethics-committee justification text with you.
Can I get the report in SPSS format?
Yes — your tables and findings are prepared in the layout SPSS users and committees are accustomed to, in APA format; results are identical to SPSS.
I'm using patient and imaging data — how is confidentiality handled?
We work with de-identified data; your data is not shared with third parties, is used solely for your study, and is automatically deleted no later than 90 days after delivery — you can request earlier deletion at any time.
How much does the analysis cost?
Price depends on the complexity of the study design and the number of variables. See our statistical analysis pricing page for current market ranges, or send your data for a free assessment within 24 hours.
Start your dentistry thesis analysis
Send your data and study design — we'll reply within 24 hours with a free assessment.
Last updated: July 12, 2026