Female biased sex ratios are a common yet unexplained phenomenon in bryophyte populations, a pattern that, for some species, appears to be correlated with increased environmental stress. Natural populations of the dioicous moss Syntrichia caninervis, an important component of the Mojave Desert biological soil crust, are highly female-biased, based on expression of gametangia. This may be because males experience greater mortality at some point in the life cycle, and are therefore rare (the rare male hypothesis), or males may simply produce sexual structures less frequently (the shy male hypothesis). To distinguish between these two alternatives, we used double digest restriction-site associated DNA (RAD) sequencing to survey the clonal diversity within two Mojave Desert populations of S. caninervis and determine whether sex ratios inferred from genetic data are consistent with ratios based on sex expression. We first identified 200 candidate sex-associated loci in a sample of 11 females and 10 males by selecting RAD sequences that were only found in one sex. Next we searched for these markers within RAD sequences of 131 sterile branches of unknown sex from two sites that differed in water availability, and potentially level of stress. Samples that were found to only have potential sex-associated loci from a single sex were identified as that sex. About two thirds of the 200 candidate loci tested showed signature of sex linkage in the full dataset. The observed phenotypic female-male sex ratio was 18-1 for the higher elevation, less stressful site (SCH) and no sex expression was observed at a dryer lower elevation site (SCL). However, using the putative sex-linked markers, we found a 2-1 genetic female bias in SCH, suggesting that males in this population are “shy”, while SCL was entirely genetically female, suggesting that males are absent. Clonal diversity was higher in SCH than SCL (Simpson’s, p-value < 0.05) and females were more clonally diverse than males in SCH (Shannon index, p-value = 0.001). Higher clonal diversity in SCH suggests fewer genotypes may be capable of growth in the lower elevation, higher stress site. Together, these results suggest that (1) both the rare male and shy male hypotheses may contribute to observed phenotypic sex ratios in the field, and (2) sex-specific differences in life history and survival interact with environmental stress to determine the proportion of males in S. caninervis populations.