|
| 阅读上一个主题 :: 阅读下一个主题 |
| 作者 |
 大王，这是你想知道的。整篇太长。 |
 |
Imbecile
[博客]
[个人文集]
游客
|
|
|
作者：Anonymous 在 罕见奇谈 发贴, 来自 http://www.hjclub.org
3.3. Different types of thinkers
The results above are in line with the assumption that developmental possibilities may be differentially realized, at least to a certain extent, in the two cultures due to systematic differences in the experience each culture provides to its members. Thus, it is justified to examine if there are different types of thinkers in the two cultures. Mixture modeling was used to answer this question (Muthen & Muthen, 2001), as it enables one to specify models in which one model applies to one subset of the data, and another model applies to another subset. That is, this method enables one to specify if there are different groups of participants in a sample that are characterized by systematic differences in the combination of performance on the various measures of interest. For the sake of this analysis, we used two measures for processing efficiency (that is mean speed and mean control of processing), three measures for working memory (that is, mean phonological, mean visuo/spatial, and mean executive memory), and three measures for thinking (that is, the logit scores for verbal, quantitative, and spatial reasoning).
The modeling here used a stepwise method, i.e., the model already presented above was tested under the assumption that there are two, three, and four classes of persons. The best fitting model was the one involving four classes (model AIC=3750.567 for 43 free parameters, entropy=0.837). The means and variances of each of the eight abilities across the four classes are shown in Table 1 and Table 2 shows the distribution of the four classes across the two ethnic groups and the four age groups. It can be seen that most of the participants belonged to class 1 (93 participants, 39.9% of the total, average class probability for members of this class is 0.94) and class 2 (95 participants, 41.4% of the total, average class probability for members of this class is 0.8 . Only a minority of the participants belonged to class 3 (35 participants, 15.0% of the total, average class probability for members of this class is 0.92) and class 4 (10 participants, 4.3% of the total, average class probability for members of this class is 0.95).
Table 1.
Estimations of means and standard errors of tasks by four classes
Tasks Class 1 Class 2 Class 3 Class 4
Mean S.E. Mean S.E. Mean S.E. Mean S.E.
Processing efficiency
Speed 0.67 0.02 0.78 0.03 0.95 0.02 1.23 0.17
Control 0.77 0.02 0.91 0.03 1.12 0.05 1.61 0.06
Working memory
Phonological 4.65 0.10 4.12 0.06 3.74 0.11 3.62 0.22
Visual 5.15 0.14 4.15 0.24 3.01 0.26 2.62 0.52
Executive 3.20 0.12 2.67 0.06 2.43 0.09 2.19 0.32
Reasoning
Verbal −0.02 0.14 −1.24 0.24 −2.38 0.19 −1.8 0.56
Quantitative 1.08 0.26 −0.12 0.27 −1.93 0.32 −1.66 0.33
Spatial 2.56 0.25 0.91 0.22 −0.82 0.49 −1.45 0.79
Table 2.
Distribution of participants in the four classes across culture and age
Culture Age Class
1 2 3 4
Chinese 8 1.7 13.7 6 1.7
10 11.1 14.5 – –
12 17.9 7.7 – –
14 25.6 – – –
Total 56.4 35.9 6 1.7
Greek 8 – 2.6 15.5 6
10 0.9 14.7 7.8 0.9
12 6 19 0.9 –
14 16.4 9.5 – –
Total 23.3 45.7 24.1 6.9
Class 1 is clearly the class of the efficient thinker. In this class, processing efficiency is very high, working memory is very powerful and thinking operates at the highest level possible. Moreover, it may be said that the thinker belonging to this class is a visualizer. This is suggested by the fact that visuo/spatial memory is distinctly robust in this group and spatial reasoning operates at ceiling. In as far as the other two reasoning domains are concerned, the mean logit score of the class indicates that persons in this class can solve up to level 3 quantitative problems and up to level 2 propositional reasoning problems. Class 2 includes the standard thinkers, so to speak. That is, the state of all processes is satisfactory but considerably lower than that of class 1. Moreover, no process seems to have a special character in this class. In as far as reasoning is concerned, problems up to level 2 can be solved in all three domains. Finally, classes 3 and 4 are very similar in as far as reasoning is concerned. That is, persons in both classes could basically solve level 1 problems in all three reasoning domains. However, it is notable that the mean logit attainment of class 3 was somewhat lower than that of class 4 in all three reasoning domains, despite the fact that both processing efficiency and working memory were considerably stronger in class 3 as compared to class 4. Thus, it is as though persons in class 3 have not yet transformed their processing potential into the reasoning skills and abilities that are commensurate to it.
How are these classes distributed over ethnic and age groups? It is interesting that all classes included participants from both ethnic groups. This finding suggests strongly that the two groups are not so deeply different so as to belong to disjoint classes. However, naturally enough, based on the results of the analyses presented earlier, there are differences in the ethnic composition of the four classes. Specifically, the majority of Chinese belonged to class 1 (56.4%), a considerable number belonged to class 2 (35.9%) and very few belonged to class 3 (6.0%) and class 4 (1.7%). The majority of the Greeks belonged to class 2 (45.7%). About the same number of participants belonged to classes 1 (23.3%) and 3 (24.1%) and few belonged to class 3 (6.9%).
Moreover, it must be noted that class membership is systematically related to age. That is, class 1 included participants mostly belonging to the two older age groups (that is, 2.2%, 15.1%, 30.1%, and 52.7% belonged to the four age groups, respectively). Class 2 was more evenly distributed over the four age groups (that is, 20.0%, 35.8%, 32.6%, and 11.6% belonged to the four age groups, respectively). Almost all of the participants belonging to class 3 came from the two younger age groups (that is, 71.4%, 25.7%, and 2.9% came from the first three age groups, respectively). Finally, 90.0% and 10.0% of participants belonging to class 4 came from the two youngest age groups, respectively. Thus, it seems that these children either have not yet acquired the processing potential needed to deal with the reasoning problems presented to them (class 4) or, if they have (class 3), they have not yet transformed it to reasoning skills and operations.
Attention is drawn to the distribution of the persons belonging to the class of visualizers in the four age groups in each cultural group, because it is very informative for the differences between the two cultural groups in regards to visuo/spatial representation and reasoning. Specifically, of the total of the Chinese visualizers, 2%, 11%, 18%, and 26% belonged to the four age groups, respectively. Of the total of the Greek visualizers, 0%, 1%, 6%, and 16% belonged to the four grades, respectively. That is, while the two cultural groups have almost the same number of visualizers until the third grade of primary school (that is until about the age of about 8 years), they deviate thereafter. It is reasonable to assume that this differentiation is more related to education rather than to other factors.
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6W4M-4F0GR5W-1&_coverDate=03%2F01%2F2005&_alid=313206621&_rdoc=1&_fmt=&_orig=search&_qd=1&_cdi=6546&_sort=d&view=c&_acct=C000049301&_version=1&_urlVersion=0&_userid=955653&md5=3a64617812b9a6397c1912a9a8a4880b#SECX23
作者：Anonymous 在 罕见奇谈 发贴, 来自 http://www.hjclub.org |
|
|
| 返回顶端 |
|
 |
| |
|
|
您不能在本论坛发表新主题
您不能在本论坛回复主题
您不能在本论坛编辑自己的文章
您不能在本论坛删除自己的文章
您不能在本论坛发表投票
您不能在这个论坛添加附件
您不能在这个论坛下载文件
|
based on phpbb, All rights reserved.
|
