Disadvantages in preparing and publishing scientific papers caused by the dominance of the English language in science: The case of Colombian researchers in biological sciences

Abstract

The success of a scientist depends on their production of scientific papers and the impact factor of the journal in which they publish. Because most major scientific journals are published in English, success is related to publishing in this language. Currently, 98% of publications in science are written in English, including researchers from English as a Foreign Language (EFL) countries. Colombia is among the countries with the lowest English proficiency in the world. Thus, understanding the disadvantages that Colombians face in publishing is crucial to reducing global inequality in science. This paper quantifies the disadvantages that result from the language hegemony in scientific publishing by examining the additional costs that communicating in English creates in the production of articles. It was identified that more than 90% of the scientific articles published by Colombian researchers are in English, and that publishing in a second language creates additional financial costs to Colombian doctoral students and results in problems with reading comprehension, writing ease and time, and anxiety. Rejection or revision of their articles because of the English grammar was reported by 43.5% of the doctoral students, and 33% elected not to attend international conferences and meetings due to the mandatory use of English in oral presentations. Finally, among the translation/editing services reviewed, the cost per article is between one-quarter and one-half of a doctoral monthly salary in Colombia. Of particular note, we identified a positive correlation between English proficiency and higher socioeconomic origin of the researcher. Overall, this study exhibits the negative consequences of hegemony of English that preserves the global gap in science. Although having a common language is important for science communication, generating multilinguistic alternatives would promote diversity while conserving a communication channel. Such an effort should come from different actors and should not fall solely on EFL researchers.

Fig 1. Time invested writing a scientific article in English (EN) as a secondary language and in Spanish (ES) as a first language.

An ANOVA analysis was performed to compare the variables obtaining an F-value = 7.095 and p-value = 0.00951 **. The dotted line represents labor hours per month.

Fig 2. Cost of translation and editing services of scientific articles from six publishers.

The Y axis is the price of the service in US dollars, the X axis represents the type of service, the standard or premium service corresponds to the delivery days. The dotted line represents an average Ph.D salary in Colombia ($ 947).

Fig 3. Reading comprehension, interpretation of figures, scientific terminology in English (EN) vs. Spanish (ES).

A Poisson regression was used to analyze these discrete ordinal variants (Qualification from 1 to 5). A Chi-squared test was performed between languages for each category: interpretation of figures (Z-value = 0.756, Pr (Chi) = 0.09754), understanding of scientific terminology (z-value = 0.143, Pr (Chi) = 0.4619) and reading comprehension (z-value = 1.427, Pr (Chi) = 0.01209 *).

Fig 4. Difficulty of writing the different sections contained in a scientific article in English (EN) and Spanish (ES).

A Poisson regression and Chi-square test was carried out: Introduction (z-value = 9.325, Pr (Chi) = 0.0158 *), methods (z-value = 3.046, Pr (Chi) = 0.07057), results (z-value = 4.899, Pr (Chi) = 0.04397 *), discussion (z-value = 11.732, Pr (Chi) = 0.02384 *), and conclusion (z-value = 7.688, Pr (Chi) = 0.03956 *).

Fig 5. Anxiety level when making oral presentations in English (EN) vs. Spanish (ES).

A Poisson regression was used to analyze discrete ordinal variants (Anxiety level from 1 to 5). A Chi-square test was carried out (z-value = 8,882, Pr (Chi) = 0.005419 **).

Fig 6. Relationship between socioeconomic status and English proficiency.

Principle components representing socioeconomic status and English proficiency are significantly correlated (R2 = 0.1548, adjusted R2 = 0.1368, F = 8.605, p-value = 0.005168 **).