Regional Differences Regarding Attitudes toward Taiwanese

Region is the factor assumed to affect attitudes toward Taiwanese. In order to compare the subjects’ judgments in the two regions, a mean score for each statement in subpart III of section III “Opinions about Taiwanese” was firstly derived for observation. Then the ratings of those statements were averaged to provide group scores for each category. Next, the Independent-Samples t-Test was used to examine whether significant differences exist between the judgments made by respondents from the Taipei and Kaohsiung metropolitan areas. Among the seventeen statements, the discussion about the data of statement 1 to 15 will be addressed by categories.

Because statement 16 and 17 are related to language education and Taiwanese proficiency in northern and southern Taiwan, they are more associated to the issues of language education and policy presented later on in the questionnaire. Hence the data illustration and discussion of the two statements are integrated with the results of the fourth subpart of section III respectively.

3.4.5 Attitudes toward Language Education and Policy

In the fourth subpart of section III, subjects were asked to evaluate, on a 5-point Likert scale, five statements⁵ regarding issues of language education and policy in Taiwan. In this part, it is found that the mean scores derived from the two regions were quite similar: no significant difference was indicated by Independent-Samples t-Test. A possible reason would be that some subjects did not have specific opinions on language education and policy so that they merely put “no comments” on the 5-Likert scale. Since mean scores cannot objectively reflect the distribution of subjects’ responses, the types of subjects’ responses were collapsed into three kinds of

5 The original design of the questionnaire contained 7 statements, but only statements 1 to 5 were discussed in this study.

64

comments in order to compare the differences of subjects’ judgments in the two regions: agree (strongly agree and agree), no comments, and disagree (strongly disagree and disagree). Participants’ degree of agreement (or disagreement) was shown by a percentage so that a potential bias caused by mean scores could be avoided. The presentation and discussion of the results will be displayed by groupings.

Moreover, in order to know the effects of regional differences in percentage patterns, the chi-square test was applied to examine the independence of the subjects’

judgments.

3.4.6 Stereotypical Impressions of Mandarin and Taiwanese

The last section of the questionnaire was designed to elicit subjects’ stereotypical impressions of Mandarin and Taiwanese. Subjects were instructed to evaluate the two languages on 12 bipolar attributes, each on a 5-point scale. The 12 attributes were divided into four subgroups: value, potency, activity, and complexity. The locations of the bipolar attributes were allocated randomly on two sides of the semantic differential scale to avoid subjects’ expectancy in choosing one side of the attribute. In this way, subjects had to pay attention to the description of each semantic differential scale instead of noticing the systematic way of allocating positive-negative features of the attributes.

As for the scoring, one side of the scale is assigned 5 points and the other side is 1 point. Generally speaking, a positive adjective is given 5 points and a negative one the least point. However, in some scales, it is hard to establish the positive-negative value. For example, there is no positive or negative value contained in the “slow-fast”,

“difficult-easy”, and “complex-simple” attributes. Thus, the 5 points given to the side of “slow”, “difficult”, and “complex” was merely for the ease of illustration of subjects’ evaluation of the two languages. The scoring mapping is listed below:

65

Table 3.3. Scoring Mapping of the 12 Attributes

困難的(difficult) _5_：___：___：___：_1_ 容易的(easy)⁶ 流暢的(fluent) _5_：___：___：___：_1_ 不通順的(awkward 活潑的(energetic) _5_：___：___：___：_1_ 不活潑的(inert) 草率的(sloppy) _1_：___：___：___：_5_ 嚴謹的(careful) 優雅的(elegant) _5_：___：___：___：_1_ 粗俗的(vulgar)

快的(fast) _1_：___：___：___：_5_ 慢的(slow)

親切的(warm) _5_：___：___：___：_1_ 疏遠的(cold) 複雜的(complex) _5_：___：___：___：_1_ 簡單的(simple) 強硬的(hard) _1_：___：___：___：_5_ 柔和的(tender) 笨拙的(dumb) _1_：___：___：___：_5_ 聰明的(smart)

有價值的(valuable) _5_：___：___：___：_1_ 沒有價值的(worthless) 有說服力的(persuasive) _5_：___：___：___：_1_ 沒有說服力的(non-persuasive)

The discussion will firstly address the comparison of mean scores of individual attributes in the four categories. Then, the Independent-Samples t-Test was used to examine the differences of two groups and the Pair Test was applied to see whether the differences between Mandarin and Taiwanese were significant. In order to observe the overall pattern of the four categories, the ratings of each category were averaged to provide category scores. Again, the Pair Test was used to explore the differences between the subjects’ attitudes toward Mandarin and Taiwanese; the Independent-Samples t-Test was applied to examine the significance of differences between the two groups of participants in evaluating the two languages.

6 In the attributes of “difficult-easy” and “complex-simple”, originally 5 points were given to the opposite site, but later in my analysis I manually reversed it so that it suited the description

“complexity better.

66

Chapter 4

Results and Analyses I: Language Use

This chapter illustrates the results of section II which includes self-reported language proficiency and language use of Taiwanese in three domains. Section 4.1 displays subjects’ self-reported language proficiency in Mandarin and Taiwanese. The results of respondents’ use of Taiwanese to talk with different people, in different places, and about different topics are outlined in section 4.2, section 4.3, and section 4.4 respectively.