Can language policy outcome be predicted?: A matrix approach to language policy analysis using Hong Kong from 1997 to 2010 as a test case

This study used Hong Kong as a test case to examine whether a self-devised, theory-informed language policy (LP) analysis tool can be used to make reliable outcome predictions, especially in multilingual environments. A so-called Ferguson Tollefson Matrix (FTM) was developed in this study to facilitate LP analysis and outcome predictions. The FTM has been designed to systematically break down a LP into its education and political components, therefore highlighting its strengths and weaknesses. The FTM is significant in at least two ways. First, as an analytical tool, it can make before-the-fact predictions on language policy outcomes - facilitating meaningful policy debates and refinements prior to implementation. Second, as a policymaking tool, the FTM can be used to support the design of pedagogically sound language policies and thus has the potential to reduce the risk of political controversies. Hong Kong was identified as a candidate for testing the FTM's reliability because, since 1998, she has gone through substantial changes in her LP, in particular over the medium of instruction (MOI) for secondary schools. The Chinese medium of instruction policies have not been popular with the majority of parents and the government has since announced a policy revision, the Fine-tuning policy (FT) taking effect from the school year 2010-2011. The timing of this policy revision vis-Ã -vis this study provides a unique opportunity to test the FTM's predictive capability and reliability against a real case. Based on the literature, historical data and others' empirical studies, this study developed and tested the reliability of the FTM before using it to analyse and make predictions about the FT. To establish a proxy-outcome, the study surveyed one quarter of Hong Kong's local secondary schools to identify what changes in practice in MOI were planned. The findings show that the FTM's predictions were in fact met, thus giving an affirmative answer to the thesis question. A discussion on the implications of the FTM's potential as an aid for language policymaking concludes the study. All rights reserved.