INTENSIVE USE OF STATISTICAL SOFTWARE AS EDUCATIONAL STRATEGY TO IMPROVE DATA ANALYSIS

Abstract

A recent study uncovered errors in data analysis carried out in a significant proportion of research conducted by nursing students. Such errors suggest dubious conclusions that highlight educational strategies aimed at getting students to make correct use of statistical software and good data analysis. We think that using the statistical software intensively improves the data analysis.

The objective of study was to demonstrate that an educational strategy (ES) improves the quality of data analysis performed by nursing students.

Materials and methods: Quasi-Experimental study with repeated measurements alternating to ES. Quality of data analysis had three dimensions; General knowledge on research methodology, Integration of knowledge on research levels and Application of knowledge in data analysis. Statistical analysis plan. Measures of central tendency and variability were calculated. The 95% Confidence Intervals were obtained. Comparisons were made for k-means with one-way ANOVA, and Tukey test was applied as post-hoc. Significance level ≤ 0.05

Results. General knowledge about research methodology. Significant differences were observed between the measurements (p = 0.000). Post-hoc test tells us that there are three subsets. Integration of own knowledge of investigative levels, significant differences between measurements were observed (p = 0.001). Post-hoc test tells us that these differences allow construction of two subsets. Finally, Application of knowledge in data analysis, no significant differences were observed (p = 0.157).

Conclusion. The ES positively influenced on General knowledge and Integration of knowledge but there were no significant changes in Application of knowledge in analysis of data. Initial basic training on use of statistical software should be included in a subsequent study.