Quantitative applied linguistics research often takes place in restricted settings of an intact language classroom, workplace, phonetics laboratory or longitudinal sample. In such settings the samples tend to be small, which raises several methodological problems. The main aim of the current paper is to give a detailed explanation of methodological and practical implications inherent in a robust statistical method called bootstrapped quantile regression (BQR) analysis. Importantly for applied linguistics research, the BQR method could help to deal with methodological difficulties inherent in small sample studies. The current study employed a moderately small sample (N = 27) of students learning the Japanese language in a Malaysian public university. It examined the relationships between the students' language learning motivation (specifically, integrative orientation), the students' images or stereotypes about Japan and their global attitudes toward the target language country and its people. The findings indicated that there was a statistically significant relationship between the students' attitudes toward the target language country and their integrative orientation. In addition, these attitudes were found to be the most constant determinant of the integrative orientation. Besides the applied linguistics research, the BQR method can be used in a variety of the human sciences research where a sample size is small.
International attention on the environmental impacts of China's Belt and Road Initiative (BRI) is increasing, but little is known internationally about the large corpus of Chinese BRI environmental research. We present the first systematic review of the Chinese and English-language BRI environmental research, supported with text mining and sentiment analysis. We found that the research is dominated by Chinese authors writing about BRI routes within China in Chinese, even though concerns around BRI are largely about impacts and benefits within host countries, and the volume of publications in English is recently catching up. Different disciplines and methods are well-represented across languages, apart from specific types of Chinese social science papers. The sentiments of academic research are largely neutral and less polarised than media discourse. We recommend that scientists and practitioners should pay more attention to BRI environmental impacts in developing countries and proactively engage local voices.