Research skills for university students' thesis in E-learning: Scale development and validation in Peru

Background

The aims of the study was to evaluate the psychometric properties and to characteristics of a scale on research skills as well as to characterize the experience of university thesis students using information databases (i.e., Scopus or Web of Science) and bibliographic reference managers.

Methods

A sample of 1598 university students from five departments, acroos the different campuses of one Peruvian university. The psychometric properties of the instrument were evaluated using factor analysis and measurement invariance according to sex and age groups. In addition, an internal consistency analysis was conducted for scale reliability using Cronbach's alpha, ordinal alpha, and the omega index.

Results

Our study identified that the unidimensional model with correlated errors showed adequate indicators of reliability and goodness of fit, with the exception of the RMSEA, which shows values higher than 0.08. Measurement invariance by sex and age group was present. The majority of university students in E-learning showed use of the bibliographic manager Mendeley (52.13%), low ability to use information databases such as Scopus (22.24%) or Web of Science (17.26%), and the least skilled research skills were in formulating the problem, objectives, and hypotheses (51.63%) and carrying out data analysis and processing using statistical techniques (49.37%).

Conclusions

The conclusions of the study were that the 8-item research skills instrument presents acceptable psychometric indicators in Peruvian university students in thesis courses. Likewise, a scarce use of global databases, the approach to the research topic, and carrying out its statistical processing.

Social-Scientific Research Competency

Abstract. Although the development of research competency is an important goal of higher education in social sciences, instruments to measure this outcome often depend on the students’ self-ratings. To provide empirical evidence for the utility of a newly developed instrument for the objective measurement of social-scientific research competency, two validation studies across two independent samples were conducted. Study 1 ( n = 675) provided evidence for unidimensionality, expected differences in test scores between differently advanced groups of students as well as incremental validities over and above self-perceived research self-efficacy. In Study 2 ( n = 82) it was demonstrated that the competency measured indeed is social-scientific and relations to facets of fluid and crystallized intelligence were analyzed. Overall, the results indicate that the test scores reflected a trainable, social-scientific, knowledge-related construct relevant to research performance. These are promising results for the application of the instrument in the evaluation of research education courses in higher education.