Since the introduction of raw Global Navigation Satellite Systems (GNSS) measurements for Android devices in 2016, the applicability of GNSS chips on Android has increased significantly. Prior to the raw measurements availability, the device users were only able to access GNSS National Marine Electronics Association (NMEA) data which included limited rudimentary GNSS related information such as position, carrier-to-noise-density ratio (C/No), speed over ground, and heading. The introduction of raw measurements has allowed the users to directly compute additional parameters such as three-dimensional velocity, which would allow computation of higher accuracy GNSS navigation solutions including altitude rate, heading velocity, and acceleration, compared to NMEA-based solutions. One significant application for higher detail GNSS information is GNSS spoofing detection. GNSS spoofing refers to the inaccurate computation of GNSS navigation solutions due to intentional or unintentional artificial false satellite signals. There are already several methods to detect spoofing, such as the analysis of C/No and automatic gain control (AGC), or navigation solution anomaly detection using inertial sensors. The method that will be explored in this paper will be a navigation domain analysis, but instead of using NMEA messages, raw GNSS measurements will be used to derive the navigation solutions. Using the raw carrier phase measurements, three-dimensional velocity, acceleration, and heading will be derived, and compared to onboard inertial navigation sensors such as magnetometer, accelerometer, and barometer, which are immune to GNSS spoofing due to their measurements being unaffected by GNSS measurements. If the comparison metrics exceed a predesignated threshold, potential spoofing alert is triggered. The novelty of this study lies in looking into the feasibility of using a smartphone to detect potential GNSS spoofing using raw GNSS measurements-derived navigation solutions, and onboard inertial sensors. In order to trigger potential spoofing, the noise of the inertial measurements from both the GNSS chipset, and the inertial sensors will be assessed under various dynamic environments, in order to derive dynamics-dependent anomaly detection threshold. Afterwards, the statistical validity of the thresholds will be assessed using two drive tests: one to determine the proposed thresholds, and one to test the effectiveness of those proposed thresholds.