We examine the measurement noise of electrical resistivity tomography and assess its effect on the inverted results. The observed and numerically simulated resistivity datasets are analyzed regarding noise distributions. We evaluate and present the contact resistance, reciprocal and repeating errors, potential noise, artificial effect on 2D resistivity measurement, inversion misfit, and model accuracy. The result shows considerable measurement noise variation for dry and wet conditions. This study uses a 3% repeatability error cut-off, and about 3.2% of the dry season and 0.83% of the wet season datasets are above cut-off values.  The result also exhibits an inverse relationship between the precipitation and reciprocal error. The resistivity measurement in dry conditions generally indicates high contact resistance, repeatability error, and reciprocal errors, resulting in significant data discarding. We also reveal the misfit between observed and model-predicted resistivity data; a high discrepancy is exhibited for noisy data, leading to substantial model error. The depth of investigation (DOI) threshold depth decrease with increasing measurement noise. This study will give insight into measurement noise evaluation, allow cut-off value, assess data noise propagation and its effects on the data misfits and inverted models, and reduce model misinterpretation.