Mindset (beliefs about the malleability of intelligence) has been studied in a variety of contexts for decades. Recent research highlights the importance of contextual factors in moderating mindset's impact on student outcomes. The commonly-used original mindset measure is context-general. Recently, a mindset measure that is specific to science and math undergraduates was developed: the Undergraduate Lay Theories of Abilities (ULTrA) Survey. I hypothesized that a context-specific measure of mindset would associate more strongly with undergraduates' outcomes than a context-general measure. I surveyed 1537 undergraduates with Dweck's 3-item original mindset measure, ULTrA, and measures of outcomes (sense of belonging, goal orientation, self-handicapping, evaluative concern and intent to persist in science) and collected course grades. Structural equation modeling indicated that the fixed factor of the ULTrA exhibited stronger and more consistent relationships with outcomes than the 3-item original mindset measure and predicted unique variance in outcomes above and beyond what the original mindset measure accounted for. The academic outcomes (intent to persist and course grade) were significantly related to ULTrA, but not the original mindset measure. Our results provide evidence that the context-specific ULTrA survey can detect relationships with undergraduate outcomes that the context-general original mindset measure can fail to detect.