This work aims to understand dispersion phase behavior as a function of solvent quality and particle volume fraction for polystyrene (PS) particles stabilized with Pluronic triblock copolymer in 0.5 M NaCl. Measurements of kT interactions between a single colloidal particle and a wall with total internal reflection microscopy (TIRM) indicate a continuous evolution in particle attraction as a function of diminishing solvent quality. The directly measured potential for relatively thin Pluronic layers displays attraction reasonably predicted by Lifshitz theory using only the PS particle properties and neglecting the Pluronic layer properties. The TIRM potential is quantitatively compared with light scattering measurements of temperature-dependent aggregation in dilute, Brownian dispersions for a large range of PS particle radii, a, relative to adsorbed Pluronic layers with thickness, delta. The aggregation behavior is well described as an irreversible flocculation process using the directly measured potential and the Lifshitz prediction for dispersions with a/delta > 40. For dispersions with a/delta < 40, when polymeric interactions dominate, the core PS particle attraction alone is not sufficient to predict the observed aggregation. The directly measured potential and dilute, Brownian aggregation measurements are used to interpret rheological measurements of a fluid-gel transition for high volume fraction dispersions with a/delta = 9. After considering the possible roles of Pluronic desorption and applied shear, a continuum polymeric van der Waals contribution to the net composite interparticle attraction is considered as the source of the observed phase behavior. The polymeric van der Waals contribution for dispersions with small a/delta is addressed with approximate uniform and nonuniform film models. The nonuniform film model predicts sufficient attraction to account for the measured aggregation and fluid-gel data, but the effect of retardation appears to be important for constructing an accurate theoretical phase diagram.