Differences in basic digital competences between male and female university students of Social Sciences in Spain
Vázquez-Cano et al. International Journal of Educational Technology in
Higher Education
Differences in basic digital competences between male and female university students of Social Sciences in Spain
Esteban Vázquez-Cano 2
Eloy López Meneses 1
Eduardo García-Garzón 0
0 Universidad Autónoma de Madrid , Madrid , Spain
1 Universidad Pablo de Olavide , Sevilla , Spain
2 Universidad Nacional de Educación a Distancia , Madrid , Spain
This article analyses the differences in basic digital competences of male and female university students on Social Education, Social Work and Pedagogy courses. The study of gender differences in university students' acquisition of digital competence has considerable didactic and strategic consequences for the development of these skills. The study was carried out at two public universities in Spain (UNED - the National Distance-Learning University, and the Universidad Pablo de Olavide) on a sample of 923 students, who responded to a questionnaire entitled “University Students' Basic Digital Competences 2.0” (COBADI - registered at the Spanish Patent and Trademark Office). The research applied a quantitative methodology based on a Bayesian approach using multinomial joint distribution as prior distribution. The use of Bayes factors also offers advantages with respect to the use of frequentist p-values, like the generation of information on the alternative hypothesis, that the evidence is not dependent on the sample size used. The results show that men have greater perceived competence in digital cartography and online presentations, whereas women prefer to request personal tutorials to resolve doubts about technology and have greater perceived competence in corporate emailing. There is also evidence that the men have greater perceived competence in developing “online presentations” than women do. Regarding to, “Interpersonal competences in the use of ICT at university”, we observed that the female students opted for personal sessions with tutors in greater numbers than the male students did.
Gender differences; Digital competences; University; Bayes factors
Introduction
Research on the possible differences between male and female university students in their
use of technology has boomed since the beginning of the century with the emergence of
of the evidence for or against any hypothesis, nor must they influence decision taking on,
for example, developing educational programmes on technology competence that are
targeted according to gender
(American Statistical Association, 2016)
.
Besides the “classic” statistics approach, researchers have also used the TAM
(Technology Acceptance Model) and UTAUT (Unified Theory of Acceptance and Use of
Technology) to measure the gender variable, with diverse results. In this article, we
propose using the Bayes factor as an alternative method of analysis to measure how far
the data support the hypotheses related to whether there are, or are not, any differences
between men and women in their application of technological competences in the
university context. Bayesian statistics offer important advantages for the classic null
hypothesis inference processes, which include: (a) the generation of information on both
hypotheses (the null hypothesis and its alternative); (b) non-dependence on the sample
plan or on the researchers’ intentions (allowing additional information to be gleaned
from the sample without the need to use procedures to maintain any error probability
constant
(Wagenmakers, 2007)
; (c) delivering interpretations that are intuitive and easy
to understand (and as likely as the data are for each hypothesis).
Bayes factor to analyse gender differences in the use of technology
The study of the educational differences has been mainly associated with the hypothesis
analysis
(Ares, 1999; Díaz & de la Fuente, 2004)
, but this statistical approach generates
numerous problems. For example, interpreting that when we reject the null hypothesis,
we get support for the (alternative) research hypothesis is not totally correct, since a
significant result does not indicate the magnitude of the effect, so the statistical
hypothesis does not report on the significance of data
(Hager, 2000; Finch, Cumming, &
Thomason, 2001)
. The interesting thing about non-categorical issues such as gender
differences is to be able to establish the magnitude of the effect. For this reason, the
use of Bayesian methods, a subject that has hardly received attention in educational
research, is recommended by eminent psychologists and educators such as
Edwards,
Lindman, and Savage (1963
),
Rozeboom, Morrison, and Henkel (1970
),
Pruzek (1997)
,
Rindskopf (1997)
and
Lecoutre (1996)
.
A fundamental difference between Bayesian and classical inference is the subjective
(and non-frequency) character of the probabilities, since the problem of repeated
sampling does not arise and it does not require the concept of sample distribution.
Subjective probabil (...truncated)