Abstract：At same time and point on the earth, crustal deformation and vibration are identical, but the recording of different observation instruments differs. One important metric affecting earthquake forecast is that the precursory signal in one kind of record cannot be found in another at the same site. We calculate the amplitude-spectrum of the data from extension-meter and water-tube tiltmeter and analyze the relation between amplitude-spectrum and capability for anomaly signal detecting. The results show that when an anomaly signal is weak, it may appear only in one recording but not in others at the same site. So, if an anomaly signal in one recording cannot be found in others at the same site, it cannot be evidence for falsification. This argument is supported by field data in Yunlong and Yongsheng stations.