The Volcanological and Seismological Observatory of Popayan (OVSPop), belonging to the Servicio Geologico Colombiano (SGC), operates a network of 24 seismological stations, consisting of short-period and broadband seismometers, in order to monitor the activity of the Nevado del Huila, Purace and Sotara volcanoes. Site effect determinations were performed on 20 stations in this network by using the coda wave method. An initial list of 348 events were selected, from which 207 tectonic and volcano-tectonic earthquakes were suitable for the study. The selection criteria involved those events with a range of local magnitudes between 2.0 and 5.2, as well as discarding those signals with high noise levels, oversaturated waveforms, events containing aftershock signals in the coda, as well as seismic events without seismic noise before and after the main event occurred. The method implementation required two main conditions: the average of the site effect values is equal to zero and each earthquake must be registered by 5 or more stations. The results show high amplification values at stations such as PAL, MAR and BUC, which are located on soft soils, and amplification factors close to and less than 1 for most of the remaining stations, which are located on hard rocks corresponding to lava flows from the monitored volcanoes. On average, the standard deviations of the obtained data are less than 0.3, although higher values were obtained for some stations and for some frequencies (up to 0.645 for the CGO station at 4 Hz). A comparison of the values found in this work with values of local magnitude corrections for the same stations shows a close correlation between the two values, which supports the idea that the coda waves are in fact dispersed S waves and that the estimation of local effects from coda waves may be a valid method for estimating the local effects of S waves.