In this paper, we explored the scientific literacy of a general sample of the Slovak adult population and examined factors that might help or inhibit scientific reasoning, namely the content of the problems. In doing so, we also verified the assumption that when faced with real-life scientific problems, people do not necessarily apply decontextualized knowledge of methodological principles, but reason from the bottom up, i.e. by predominantly relying on heuristics based on what they already know or believe about the topic. One thousand and twelve adults completed three measures of scientific literacy (science knowledge, scientific reasoning, attitudes to science) and several other related constructs (numeracy, need for cognition, PISA tasks). In general, Slovak participants' performance on scientific reasoning tasks was fairly low and dependent on the context in which the problems were presented--there was a 63% success rate for a version with concrete problems and a 56% success rate for the decontextualized version. The main contribution of this study is a modification and validation of the scientific reasoning scale using a large sample size, which allows for more thorough testing of all components of scientific literacy.