In this study, the detection characteristics of volatile organic compounds (VOCs) were investigated in relation to two different experimental approaches (direct injection (DI) vs. solid phase micro extraction (SPME)) based on effective carbon number (ECN) theory. To this end, the calibration data obtained by both methods were evaluated in association with split ratio (such as 1:1 (splitless), 1:5, 1:10, and 1:20) The calibration results of eight different target VOCs were examined in terms of relative sensitivity at a given split ratio. If slope values are derived between ECN and Response Factor (RF) values of each compound, the DI-based data obtained at 1:5 split ratio yielded slightly positive bias of +3.6% relative to theoretic values of 20%. Likewise, the results at 1:10 split condition showed +1.2% bias. In contrast, the SPME-based results showed significantly large biases. The result of this comparative study suggests the need for more deliberate QA procedure to obtain quantitative data for the SPME method.