Experiments on plasma turbulence and related transport processes generally require diagnostics with high spatio-temporal resolution. Standard tools for this purpose are statistical methods like cross-correlation analysis or conditional averaging. High spatial resolution is achieved at the expense of an asynchronous recording and a subsequent averaging process. Recently, we had shown that diagnostics arrays in combination with a super-resolution algorithm can achieve a comparably high spatial resolution [1]. This paper compares the results of both techniques applied to identical experimental conditions. It addresses the question of whether the higher efforts for spatio-temporal diagnostics are needed to study the dynamics of large-scale structures in a turbulent system and to what extent the results of statistical approaches are reliable.