Abstract: Just the place for a Snark! I have said it thrice;/What I tell you three times is true (Carroll). Background: There is increasing interest in the role and practice of systematic reviews and meta-analyses of qualitative research. Experience of meta-ethnography (Noblit & Hare, 1988) as a technique of synthesis is growing, especially in the UK. In common with other techniques of synthesis for qualitative research, however, it relies partly on cumulative validity - findings are given more weight if they are reported by more than one study, or can be explained in terms of findings from another study. But is such an interpretation correct? Objectives: To explore the impact of different methodological and theoretical approaches to the findings of qualitative studies. Methods: A systematic review of 20 qualitative studies about women’s experiences of hysterectomy for benign conditions is used as an example. I describe the impact of expanding metaethnographic techniques of meta-data-analysis through the inclusion of the additional steps of meta-method and meta-theory described as part of meta-study (Paterson et al., 2001). Results: Understanding the methods that produced different findings (for example, interview reports as opposed to direct observation) allows alternative explanations for some findings to be developed. Expanded theoretical frameworks can challenge some simple themes, such as the finding that women require more information to make good decisions about hysterectomy. Understanding that such decision making takes place in the context of socio-cultural constructions of gender, provides a context for fears women have that go beyond factual understandings about what hysterectomy involves. Conclusions: I argue that it is imperative that syntheses of qualitative research take account of the use of and creation of theory. Similar findings may be the result of similar approaches or contexts while unique findings may be the result of innovative, but pertinent and important, insights, methods or theories. Studies that are better developed conceptually should carry more weight in the synthesis. Where similar findings are found across studies, this should not automatically be taken as an indication of validity. Similarly, ‘outlying’ findings should not be dismissed.