Quasi-equilibrium (QE) theory posits that, over large enough spatial and temporal scales, temperature profiles in the tropical troposphere are constrained by moist convection. The degree to which quasi-equilibrium holds true is investigated over different temporal and spatial scales, as well as over various regions of the globe, by regressing temperature perturbations at various pressure levels on column average temperature perturbations (and on near-surface temperature perturbations) and then comparing these linear regression slopes to similar slopes derived from a distribution of calculated moist adiabats. For NCEP reanalysis data at monthly and larger time scales and large space scales the shapes of these curves are fairly similar for the tropical free troposphere. At daily time scales and sufficiently large space scales, NCEP reanalysis data, AIRS Level 2 satellite data, and TOGA COARE IOP (Intensive Observing Period) rawinsonde data vertical perturbation profiles still show general agreement with the moist adiabat perturbations in most of the free troposphere, though the relationship to the atmospheric boundary layer exhibits some departures. Furthermore, over at least these shorter time scales there are large negative linear regression coefficients at or above 100 mb in all data sets analyzed, a feature not present for purely moist adiabatic perturbations. One hypothesis to explain this, that vertical velocity present by continuity above areas of convective heating can cause adiabatic expansion and thus negative temperature perturbations, is investigated. Implications are considered for models that assume a moist adiabat-like structure for tropospheric temperature profiles and perturbations.