Several methods have been proposed to estimate the sensation level (SL) at which children receive amplified speech from their hearing aids. The present study compared the SL estimates obtained with two such methods: (1) a sound field aided audiogram approach, and (2) an electroacoustic approach that incorporated the use of a probe tube microphone system (Seewald, Ross, & Stelmachowicz, 1987). Sound field aided thresholds were obtained for 13 hearing-impaired subjects at eight audiometric frequencies. For the electroacoustic approach, in situ thresholds were obtained using a button-type hearing aid receiver attached to a custom earmold. Real ear aided responses were measured using a 70 dB RMS speech-weighted composite noise signal (Frye, 1986). A comparison of the frequency-specific SL estimates derived from the two different methods revealed that the sound field aided audiogram approach yielded higher SL estimates for 74% of the individual comparisons. A detailed analysis of the findings obtained from two subjects suggested that when the results of the two methods did not agree, the differences were due to an interaction between signal level and the unique input/output characteristics of the subjects' hearing aids. A precautionary measure is suggested for those who wish to use sound field aided threshold data to estimate the SLs at which children receive amplified conversational speech.