The significance of the integrated Sachs-Wolfe effect revisited

We revisit the state of the integrated Sachs–Wolfe (ISW) effect measurements in light of newly available data and address criticisms about the measurements which have recently been raised. We update the data set previously assembled by Giannantonio et al. to include new data releases for both the cosmic microwave background and the large-scale structure of the Universe. We find that our updated results are consistent with previous measurements. By fitting a single template amplitude, we now obtain a combined significance of the ISW detection at the 4.4σ level, which fluctuates by ∼0.4σ when alternative data cuts and analysis assumptions are considered. We also make new tests for systematic contaminations of the data, focusing in particular on the issues raised by Sawangwit et al. Amongst them, we address the rotation test, which aims at checking for possible systematics by correlating pairs of randomly rotated maps. We find results consistent with the expected data covariance, no evidence for enhanced correlation on any preferred axis of rotation, and therefore no indication of any additional systematic contamination. We publicly release the results, the covariance matrix and the sky maps used to obtain them.