Generalizability theory (G-theory) extends classical reliability theory by partitioning measurement error into multiple, separable sources, thereby supporting the design of more dependable measurement procedures. Existing software for G-theory — most notably the GENOVA Suite and EduG — is widely used but predates contemporary R-based reproducible workflows and lacks interactive exploration of decision (D) study trade-offs. We introduce gtheory-app, a free and open-source Shiny application that brings G-theory analysis into the browser. The application accepts long-format data uploads with auto-detected delimiters or score-only files paired with a user-specified design, supports balanced and unbalanced data, and handles fully crossed, nested, and mixed (random + fixed) designs with an arbitrary number of facets. Variance components are estimated via lme4 (REML) with a classical ANOVA cross-check available for balanced data. The application then reports universe-score variance, relative and absolute error variances, the generalizability coefficient (Eρ²), the index of dependability (Φ), Φ(λ) for criterion-referenced decisions, a G-facets sensitivity table, and an interactive 3D surface graph over any pair of user-selected facets with the resulting D-study. Every numerical output is paired with a context-sensitive, plain-language interpretation that names the user's actual facets, places reliability values in conventional bands, and identifies the largest single error term as a target for redesign. We illustrate the application on three classes of designs: a fully crossed three-facet design, a nested-items design, and a multi-facet chain-nested design. We argue that gtheory-app fills a clear gap in the open-source psychometric software ecosystem and lowers the barrier to G-theory for applied researchers, educators, and assessment professionals. CHANGELOG: 0.8.1 (May 2026) Fixed - Mixed variance components now match EduG exactly. v0.8.0 computed Mixed VCs by solving `M_mixed × σ² = MS` where `M_mixed` was built from `M_random` by zeroing entries whose `β\α` set contained any fully-fixed facet. For many components (notably D in example 16) every off-diagonal collapsed and the equation returned `σ²_random` instead of the correct Mixed value. v0.8.1 replaces that matrix solve with the EduG-equivalent additive correction: ``` σ²_mixed(α) = σ²_random(α) + Σ_β σ²_random(β) × M_random[α,β] / M_random[α,α] ``` where β ranges over components strictly containing α, with `β\α` consisting entirely of **fully-fixed** facets (`N = n`). Hand-verified against EduG's example 16 on D, M, C:DS, P:C:DS, K:O, and I:K:O — all six recover EduG's reported Mixed values to displayed precision. σ²(X̄..) computed from the corrected Mixed values lands on 0.015, matching EduG. - G-Study contributions and final Coef_G now match EduG. With the corrected Mixed VCs flowing through `compute_g_coefficients`, the dependent quantities — `σ²(τ)`, `σ²(δ)`, `σ²(Δ)`, `Coef_G_rel`, `Coef_G_abs`, `σ²(X̄..)`, and the per-component `Differ./Rel/Abs` cells — all align with EduG to displayed precision. The visible discrepancies you saw in v0.8.0 (D's absolute err. var. 0.0111 vs 0.005; σ²(X̄..) 0.0213 vs 0.015; Coef_G_abs 0.5752 vs 0.65) are all eliminated. Cosmetic - Intro paragraph no longer lists per-version highlights.- Author and personal website (`carloscollares.com`) added to intro.- Footer credits GNU General Public License v3.0 alongside the documentation link. Inherited from v0.8.0 - `(N − n)/(N − 1)` correcting coefficient (per EduG).- Coefficient applied only to MAIN (effect) fixed facets in each component, never to transitive nesting parents.- All `D × I` components route to relative error.- Universe-size column accepts `INF` / `Inf` / `infinite`.- Restricted Φ(λ) fallback. Known limitation - The Universe radio still uses "random / fixed" rather than inferring sampling status from N vs n the way EduG does. A future version will drop the radio and let N alone determine fixed / finite random / infinite random. Earlier versions See v0.8.0, v0.7.0, v0.6.0, v0.5.0, v0.4.0, v0.3.0, v0.2.x, v0.1.xentries in those folders' CHANGELOGs.