The determination of the glass transition temperature, Tg, of polymers by differential scanning calorimetry (DSC) and thermomechanical analysis (TMA) is sometimes problematic and rather subjective. This was shown previously in the ICTA certificate (distributed by NBS as GM-754) for the certified reference material polystyrene (PS). The not very good reproducibility of the measured value of the onset is due to a variety of instrumental and experimental parameters. This is true also for the determination of the glass transition by TMA measurements. The main reasons are temperature gradients caused by the relatively high sample mass required for DSC and by the limited heat transfer in TMA, respectively. Our own experiments which were carried out with polystyrene and with [poly(ethyleneterephthalate)] (PET) proved that a combination of DSC with TOA (thermo-optical analysis or hot stage microscopy under polarized light) can solve some of these problems. TOA is a nonsubjective method since the changes in birefringence and light transmittance during the glass transition which are visible under the microscope are measured with a photocell. TOA allows Tgmeasurements of small samples (fraction of milligrams).