The present study aims at investigating some English speaking problems found by the students of English language education department at State Islamic Institute of Kudus and exploring some implications for speaking curriculum development of English language education department. This study used a descriptive qualitative approach and the data were taken from semi-structured interviews, observation and focus group discussion (FGD). The results of the analysis reveal that the problems encountered by the students in English language education department in speaking English cover the lack of appropriate vocabulary, the lack of grammar mastery, the lack of correct pronunciation, the lack of input of English outside the class, the lack of confidence and the lack of English speaking curriculum development. There have been some implications for developing English speaking curriculum. First, the curriculum of speaking should be well designed. Second, lecturers should design English speaking curriculum integrated by technology and social media that makes students to enrich some culture, knowledge and experience around the world. Third, lecturers should facilitate the English speaking curriculum with the English speaking community so that the students will have more chance to speak English inside and outside the class.

Successfully solving reality-based tasks requires both mathematical and text comprehension skills. Previous research has shown that mathematical tasks requiring language proficiency have lower solution rates than those that do not, indicating increased difficulty through textual input. Therefore, it is plausible to assume that a lack of text comprehension skills leads to performance problems. Given that different sociodemographic characteristics and cognitive factors can influence task performance, this study aims to determine whether text comprehension mediates the relationship between these factors and competence in solving reality-based tasks. Additionally, it examines the impact of systematic linguistic variation in texts. Using an experimental design, 428 students completed three reality-based tasks (word count: M = 212.4, SD = 19.7) with different linguistic complexities as part of a paper-pencil test. First, students answered questions about the situation-related text comprehension of each text, followed by a mathematical question to measure their competence in solving reality-based tasks. The results indicate that: a) Tasks with texts of lower linguistic complexity have a significantly higher solution rate for both text comprehension (d = 0.189) and mathematical tasks (d = 0.119). b) Cognitive factors are significant predictors of mathematical solutions. c) Text comprehension mediates the relationship between the impact of students’ cultural resources and cognitive factors and their competence to solve reality-based tasks. These findings highlight the importance of linguistic complexity for mathematical outcomes and underscore the need to reinforce text comprehension practice in mathematical education owing to its mediating role.