Analysis
Scores for each of the 33 items for each questionnaire were initially recorded in
Microsoft Excel, where tables and data were first organized in tabs corresponding to
independent groups. The Likert scale on the FLCAS was reversed for negativelyworded
questions. Raw data was then copied and pasted into SPSS for final analyses.
Various tests were run to examine linearity, normality, and homoscedasticity of
variables. Initial comparisons between means were analyzed using one-way ANOVA
and independent t-tests. However, ANOVA was considered to be a more robust test of
the data.
ANOVA tests were run to examine differences between means in FLCAS scores
between all groups, and for differences between means in the immersion group. Pearson
product-moment correlations were calculated to assess relationships between FLCAS
and English exam scores for all groups together. An additional Pearson correlation test
was run to examine relationships between variables in the immersion group.