The goal of this work was to evaluate changes in dietary fiber measured by the traditional enzymatic-gravimetric method (AOAC 991.43) and the more recently accepted modified enzymatic-gravimetric method (AOAC 2011.25), mono- and disaccharides, and starch as a function of assessed ripeness in a controlled study of a single lot of bananas and in bananas at the same assessed stages of ripeness from bananas purchased in retail stores, from different suppliers. Sugars, starch, and dietary fiber were analyzed in bananas from a single lot, at different stages of ripeness, and in retail samples at the same assessed stages of ripeness. Mean fiber measured by the traditional enzymatic-gravimetric method (EG) was ~2 g/100g and not affected by ripeness. Mean fiber assessed with the recently modified method (mEG) was ~18 g/100g in unripe fruit and decreased to 4-5 g/100g in ripe and ~2 g/100g in overripe bananas. Slightly ripe and ripe bananas differed by ~1.1 g/100g in the controlled single-lot study but not among retail samples. There was a large increase in fructose, glucose and total sugar going from unripe to ripe with no differences between ripe and overripe. Aside from stage of ripeness, the carbohydrate composition in retail bananas is likely affected by differences in cultivar and post-harvest handling. Results from this study demonstrate the importance of measuring dietary fiber using the mEG approach, developing more comprehensive and sensitive carbohydrate analytical protocols and food composition data, and recognizing the impact of different stages of maturity and ripeness on carbohydrate intake estimated from food composition data.