Learner corpus research involves studying large collections of data to achieve a certain degree of representativeness, which means that it is often not doable to examine a full set of data qualitatively. An important issue, then, is how to select a subset for further qualitative analysis. This study illustrates a selection method, taking quantitative results as a starting-point, for a qualitative study of a lexicogrammatical pattern. Three configurations are examined, involving not only statistically significant differences (overuse and under-use), but also similarities (equal use). What is studied is the anticipatory it pattern ("It is however important to interpret these findings with caution") in apprentice writing in linguistics by learners and native speakers of English. The method yielded 463 tokens in 62 learner and 82 native-speaker essays. The research questions were (i) What are the connections between the selected subpatterns of anticipatory it and specific rhetorical moves? and (ii) Are there indications of learner behaviour in the connections between subpatterns and rhetorical moves? Most subpatterns were found to be specialised for a few moves. The two groups mostly used the subpatterns for the same rhetorical work, but the learners used important and clear subpatterns for a greater range of moves.