TLI (Tucker-Lewis Index)
A good TLI is .95 or higher, with .90 sometimes treated as acceptable. TLI is an incremental fit index, also called the non-normed fit index, that compares your model with a baseline model while penalizing complexity.
What it means
Like CFI, TLI compares your model to a null model of uncorrelated variables, but it rewards parsimony by accounting for degrees of freedom. Because it is non-normed, TLI can occasionally fall below 0 or rise slightly above 1.
Why it matters
TLI and CFI usually agree, but TLI is stricter with complex models, so reporting both gives a fuller picture of fit. It is part of the standard fit-index set for CFA and SEM.
How to report it
Report TLI to two decimals. A common guideline is TLI ≥ .95 for good fit (Hu & Bentler, 1999; https://doi.org/10.1080/10705519909540118), with .90 sometimes treated as acceptable. Check it with the fit index checker.
Reference table
| Cutoff | Interpretation |
|---|---|
| ≥ .95 | Good fit |
| ≥ .90 | Sometimes acceptable |
Frequently asked questions
What is a good TLI value?
TLI of .95 or higher is read as good fit (Hu & Bentler, 1999), and .90 or above is sometimes acceptable. Because TLI is non-normed it can exceed 1 slightly.
How is TLI different from CFI?
Both are incremental indices, but TLI penalizes model complexity through degrees of freedom and is non-normed, so it tends to be slightly lower and stricter than CFI for complex models.
How to cite this page
If this page helped your research, you can cite it (APA 7):
Szymkowiak, A. (2026). TLI (Tucker-Lewis Index). Explonia. https://explonia.com/glossary/tli/
Last updated: July 2026
Run your whole study in one place
Build it, collect it, and get publication-grade analysis with a draft APA paragraph you can edit. Free to start. No credit card.
Start free