Basically the only thing to consider is that you are giving displacement data for creating yout mathematical model. With the strain gauges you measure uSt, and you should convert it to absolute displacement. Best would be to acquire only the dynamic part of the strain signal in order to use the full ADC range and have the best resolution. Then yor FRFs will be microns/N instead m/s2/N... but the modal parameters can be calculated.
Be aware that the dynamic response of the Strain gauges is much more limited than an accelerometer. I mean, the reliability of the measurement is much more limited in frequency. But if you are considering Strain Gauges is probably beacuse your frequency range is for sure low.