📄 Difficulties in using correct English prepositions among EFL students

Summary

This study looks at why Saudi university learners of English as a foreign language often choose the wrong preposition, and what teachers might do about it. It focuses on two things: prepositions that a verb requires (for example, a verb that only takes at or in) and fixed prepositional phrases such as travel expressions with by or idioms beginning with on or out of. Ten students in levels seven and eight at a Faculty of Sciences and Arts took a written test with 50 verbs, where they had to supply the missing preposition. The author analysed the answers with simple descriptive statistics, reporting the share of correct and incorrect answers for each preposition. For most prepositions, accuracy was low. The paper explains this through Arabic–English mismatches, preposition polysemy and the lack of rules governing their use, and through curricula that neglect the topic. It recommends more targeted practice and more textbook coverage. The reported numbers have internal problems, and the paper's conclusion contradicts its own results section, so the findings should be treated as indicative only.

Key points

• The sample is 10 Saudi EFL students. The instrument is a written fill-in test of 50 verbs, analysed with the Descriptive Statistical Method.
• The prepositions tested fall into three groups: in/on/at/of, for/from/about and into/with/to. Reported accuracy includes at 4%, into 8%, for and from 14% each, to 16% and about 28%.
• The numbers do not agree with each other:
  • in and on are described both as mostly correct (70%/88%) and as mostly wrong (30%/12%).
  • with appears as both 8% and 20%.
  • The phrase-test figures exactly repeat the verb-test figures.
• The results section says the hypothesis about verb–preposition use was rejected, yet the closing paragraph claims both hypotheses were confirmed.
• Table 1 lists verbs grouped by the preposition they govern (e.g. concentrate/insist → on, stare/peer → at). Table 2 lists fixed prepositional phrases (on behalf of, out of order, by mistake, from time to time).

Interesting ideas

• I1. Organising verbs by the single preposition they select turns preposition choice into a lexical property of the verb rather than a free semantic decision.
• I2. The paper treats fixed prepositional phrases (by bus, on behalf of, out of order) as one unit with one function, not as a preposition plus a free noun.
• I3. The paper separates PPs that answer when/where/how (adverb-like) from PPs that answer which one/what kind (adjective-like). This is a practical test for deciding what a PP attaches to.
• I4. The test blanks the preposition after a verb and scores accuracy separately for each preposition rather than as one total.
• I5. The paper argues that L1 and L2 prepositions do not map one-to-one and that their meaning depends on context.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

• I1 (frames): a verb→governed-preposition table could flag such PPs as core arguments rather than adjuncts when assigning VerbNet roles, and the flag would be easy to explain.
• I2 (tokens): a multiword-expression lexicon built from such lists could mark these spans early, so later stages see them as one fixed unit instead of compositional PPs.
• I3 (tree): the question-type test could serve as an explainable heuristic or a silver-data check when choosing between obl and nmod attachment.
• I4 (learn): a masked-preposition probe with per-preposition breakdowns would show which governed prepositions the system has really learned, with random-fill baselines as a negative control.
• I5 (back): when translating through the meta-world, the target preposition should be chosen from the relation in the meaning graph and the target verb's requirements, not copied from the source word.