Correlation Analysis using Teaching and Learning Analytics

Published: 20 May 2021| Version 1 | DOI: 10.17632/5yb5r8bz98.1
Contributor:
Pedro Alexandre Nery Prestes

Description

The data processed with statistical and artificial intelligence techniques show dichotomous results through their weakened correlations. For this context, Pearson and Spearman's correlation techniques were chosen where it was clear how the linearity of their correlations is compromised and thus undermining all the inferences that will define greater precision in the interconnections between the motiv themes and subthemes.

Files

Steps to reproduce

The data are sourced from the Brazilian government through INEP in partnership with the OECD. These raw data were cleaned up and applied using statistical and artificial intelligence techniques. The data refers to the issue related to teachers, managers and school units and thus through its variables they can infer how assertive and linear they may be, however in our analysis, it was observed that their correlations through Pearson and Spearman are mostly non-linear and dichotomous. Therefore, this applied method generated responses that allow us to visualize scenarios with themes and subthemes that are far from precise answers for possible more assertive decision making.

Institutions

Universidade Federal do Ceara - Campus do Pici

Categories

Categorical Data Analysis, Analysis of Education

Licence