Page 113 - Getting of the fence

Student motivation in the EFL literature lesson
5.3.3 Data collection method and analysis
5.3.3.1 Measuring level of engagement and level of importance
In order to measure the level of student engagement we used an instrument based
on the student survey of the Engagement versus Disaffection (EvsD) instrument
(Skinner et al. 2009). The instrument was translated into Dutch and adapted to
EFL literature lessons. In the process of translation four items were deleted from
the original instrument due to ambiguity. (See Appendix II for the original and
which items were deleted). The students were asked to report on a scale of 1 - 4 (1
= I disagree, 4 = I agree), as in the original instrument, on their own behavioural
and emotional engagement and disaffection during EFL literature lessons.
Behavioural engagement was measured using 5 items that tapped students’
attention and participation during the EFL literature lessons (Cronbach α = .78).
Behavioural disaffection was measured using 4 items that tapped students’ lack
of effort (Cronbach α = .76). Students’ emotional engagement was assessed using
5 items that tapped whether students felt good during the EFL literature lessons 5 and whether they enjoyed learning new things (Cronbach α = .84). Emotional
disaffection was assessed using 9 items that tapped emotions indicating boredom and discouragement (Cronbach α = .63).
In order to measure the level of importance students ascribed to EFL literature lessons, we used the underlying elements of the Comprehensive Approach to foreign language literature teaching and learning. The students were asked to indicate on a scale of 1 - 4 (1 = not important, 4 = important) to what extent they deemed each of the underlying elements important (see Appendix III). Descriptive statistics were then calculated for each element.
5.3.3.2 Calculating relationships between engagement and importance
To calculate the relationship between engagement and importance we first conducted an exploratory factor analysis on the items of the Comprehensive Approach to define the underlying structure based on the students’ answers. Secondly, we employed a correlation analysis to calculate whether level of engagement is significantly related to the ascribed level of importance. The a level was set at p < .05.
Based on an analysis of the distribution of mean difference and correlational effects observed in 91 meta-analyses and 346 primary studies, Plonsky and Oswald (2014) propose the following field-specific scale for interpreting and reporting effect sizes for correlation coefficients in L2 research which we will follow in
111