The phrase "deviations from a linear" is correct and usable in written English.
You can use it when you want to describe a situation or outcome that has not followed a linear path or pattern. For example, "The study found evidence of significant deviations from a linear relationship between the two phenomena."
The test for possible non-linear relationships between the SDQ scores and the dependent variable (time to receiving health-related benefits) showed that there were no significant deviations from a linear relationship for any genders (p for non-linearity ≥0.157).
Deviations from a linear temperature dependence of the resistivity increase with increasing solute concentration.
Clear experimental evidence is provided confirming the predictions of Tang et al. 1999 [19] regarding deviations from a linear relation between the thermodynamic equilibrium temperature and Ni concentration.
Deviations from a linear model are quantified statistically by generating surrogate data sets that, by construction, possess only linear (second-order) correlations.
For certain components where the deviations from a linear approximation are non-negligible (gas spring in the displacer cylinder, check valves and piston valve, and heat exchangers), a non-linear description is employed.
Deviations from a linear path were observed to repeat to within a few nanometers about nominal variations of less than 10 nm for a traverse distance of 10 mm.
(The small deviation from a linear function at the bond coat/copper interface is due to the sinusoidal shape of the interface).
Positive deviation from a linear dependence of the glass transition temperature with composition was observed and a molecular interaction parameter determined.
Data at 4 Tesla showed at short stimulus duration a nonlinearity, i.e., deviation from a linear model, with an index up to 400%, whereas data at 7 Tesla exhibited a nonlinearity index up to 40%.
There was a strong positive correlation between the projected shoot area obtained by image analysis and shoot fresh weight in the two rice cultivars, Fatmawati (R2 = 0.97) and IR64 (R2 = 0.98) (Figure 2) and there was no indication of any deviation from a linear relationship even at the highest biomasses measured in this experiment.
A Estimated effect (standard deviation) from a linear model.
