Recently, the Self-Compassion Scale (SCS) has been criticized for problems with psychometric validity. Further, the use of an overall self-compassion score that includes items representing the lack of self-compassion has been called into question. I argue that the SCS is consistent with my definition of self-compassion, which I see as a dynamic balance between the compassionate versus uncompassionate ways that individuals emotionally respond to pain and failure (with kindness or judgment), cognitively understand their predicament (as part of the human experience or as isolating), and pay attention to suffering (in a mindful or over-identified manner). A summary of new empirical evidence is provided using a bi-factor analysis, which indicates that at least 90 % of the reliable variance in SCS scores can be explained by an overall self-compassion factor in five different populations, justifying the use of a total scale score. Support for a six-factor structure to the SCS was also found; however, suggesting the scale can be used in a flexible manner depending on the interests of researchers. I also discuss the issue of whether a two-factor model of the SCS—which collapses self-kindness, common humanity, and mindfulness items into a “self-compassion” factor and self-judgment, isolation, and over-identification items into a “self-criticism” factor—makes theoretical sense. Finally, I present new data showing that self-compassion training increases scores on the positive SCS subscales and decreases scores on the negative subscales, supporting the idea that self-compassion represents more compassionate and fewer uncompassionate responses to suffering.