In some contexts (e.g., the control condition), the standard full formulation of LV-GD (Table 2 ID = 5) did not reach statistical significance for the association between gaming and learning, although its associations were still better than those of KE-GD.
← all excerpts
Using latent variable models to make gaming-the-system detection robust to context variations.
2
—
—
The sentences
Meanwhile, we also noticed that the reliability of both KE-GD and LV-GD has a decreasing trend with the decrease of format context similarity, suggesting some degree of influence from the contexts on the gaming estimates.