1. They report a positive correlation (r=0.71) between digital burnout and poor psychological health. But the GHQ-28 is scored so that higher scores = worse health. So a positive correlation makes sense. But here’s the catch: if they didn’t reverse-score items properly (or mislabeled directions), that correlation could be totally misleading. Did they actually verify the scoring direction? Because if they messed that up, their whole conclusion flips…..
2. They claim digital burnout explains 50.9% of the variance in psychological health, but none of the demographic variables (age, course load, department) were significant. That’s… suspicious. Either digital burnout is incredibly powerful (which is possible but rare), or there’s multicollinearity or overfitting going on. Did they check for suppressor variables? Because something’s off when a single predictor dominates like that…..
3. Look at the plots: one shows a tight cluster, another shows a scattered mess. The correlation for digital aging (r=0.34) is weak, but they lump it into the same “strong correlation” narrative. And the plot for digital deprivation looks curvilinear—did they test for nonlinearity? If not, their Pearson r might be totally inappropriate….