# APPLICATION OF KAHOOT! TO COMPLEMENT THEORY CLASS CONFERENCES IN APPLIED STATISTICS IN THE DEGREE OF PHYSIOTHERAPY. A PILOT STUDY

Introduction:

Until recently, traditional teaching methods were based mainly in theory class, which used as a didactic strategy the teacher's verbal presentation of a specific subject in unidirectional way (teacher-student). With the appearance of the new degrees in Higher Education, teachers need to adapt the teaching methods to the requirements of the curricula and use tools to favour motivation and knowledge assimilation. On the other hand, theory class cannot be excluded in some subjects, so additional strategies are needed to facilitate meaningful learning or skill assessment. Our aim was to test the utility and applicability of an ICT-based interactive tool (Kahoot!) to pose problems and getting immediate feedback on conceptual knowledge learning during the course of the classes of “Applied Statistics in Health Sciences” in the Degree of Physiotherapy.

Methods:

Forty-two students of the first year Degree in Physiotherapy were included. We used four questionnaires (30 items total) which were administered by the end of each thematic block in the subject. The students used their mobile devices to access Kahoot! and they answered four-choice questions in which only one was correct. They had 30 seconds to answer. The items referred to four defined sections: I) Descriptive statistics, II) Introduction to inferential statistics, III) Parametric analysis and contrast of means, and IV) Association between variables and non-parametric tests. All lectures were made by the same teacher. We measured total score, mean time to answer, correct, and incorrect answers in each of the questionnaires. By the end of the semester, the final score in the theoretical part of the subject was registered. One-way ANOVA with repeated measures (Bonferroni’s post-hoc) and bivariate correlations were used for statistical analysis (p<0.05).

Results:

We did find differences in the score among the four different tests [F(3,115)=8.37, P<0.05, η2=0.82]. Indeed, total score was lower in section II, whereas no difference was found between III and IV. When we correlated total score, mean time to answer, correct, incorrect answers and final score, we found an inverse correlation between total score - mean time to answer (p=0.021, r=-0.428), and, as expected, total score - correct answers (p<0.001, r=0.923), and total score - incorrect answers (p<0.001, r=-0.740). Of note, no correlation was found between final score and the rest of variables.

Conclusion:

Quantitative analysis by means of the use of an ITC-based interactive tool as Kahoot! enabled us to identify problems in the assimilation of concepts related to inferential statistics, which in turn will serve to improve this particular group of lessons for next courses. Surprisingly, no correlation was found between total score and the final theory score, so the use of this tool during the semester did not seem to have a significant effect on the results of the test.