Skip to main content
. Author manuscript; available in PMC: 2018 Dec 1.
Published in final edited form as: J Mem Lang. 2017 Aug 25;97:135–153. doi: 10.1016/j.jml.2017.07.008

Table 4.

Change in misfit for the full model (Model 1) as measured by χ2 during backward-building.

Path Removed df χ2 Δ df Δ χ2 p of Δ χ2
None 323 528.629 - - -
WMC ↔ Inhibition 324 528.654 1 0.025 0.8744
Fluency ↔ Inhibition 325 529.826 1 1.172 0.2790
Decoding ↔ Reasoning 326 531.733 1 1.907 0.1673
Reasoning ↔ Inhibition 327 533.066 1 1.333 0.2483
Decoding ↔ WMC 328 535.291 1 2.225 0.1358
Inhibition → Comprehension 329 535.298 1 0.007 0.9333
Perceptual Speed → Comprehension 330 535.446 1 0.148 0.7004
Fluency → Comprehension 331 538.120 1 2.674 0.1020
Decoding → Comprehension 332 540.466 1 2.346 0.1256
WMC → Comprehension 333 542.882 1 2.416 0.1201

Notes: ↔ indicates a covariance path

→ indicates a regression path.