- 著者
-
布留 武郎
- 出版者
- 国際基督教大学
- 雑誌
- 国際基督教大学学報. I-A, 教育研究 (ISSN:04523318)
- 巻号頁・発行日
- vol.20, pp.141-167, 1977-03
Hypothesis: The bulk of TV program content is entertainment, offering mainly passive pleasure. Although the program content varies the perceptive process in TV viewing has certain common characteristics in its structure. Children are required to perceive immediately the context developed on a TV screen with a global frame of reference but are not required to visually analyze the situations or reflect over them when the responses are uncertain. Therefore as a cumulative effect it can be assumed that TV might function to facilitate children's cognitive tempo and increase their field-dependency, nonananalytic preference or impulsivity. Method: Three kinds of tests were administered to ninetyone boys and ninety-six girls in the fifth grade from the Tokyo suburban area to assess cognitive style and one questionnaire to measure three kinds of TV viewing pattern. The fifth grade children are selected because they are of the age which receives a relatively stable habit of TV exposure (Furu '71), and in addition they still have the possibility of modifying their cognitive style in the future. The three kinds of cognitive tests included a Group Embedded Figure Test (25 items) developed by H.A. Witkin, a Group Conceptual Style Test (24 items) developed by J. Kagan, and a Matching Familiar Figure Test (10 items with 6 variants of each item) developed by J. Kagan, which was further modified for the Japanese children (Furu et al., '75). Reliability coefficient of each test was r_<tt>=0.88 (Kuder-Richardson's formula 20) on EFT. r_<tt>=0.89 (K-R's formula 20) on CST, r_<tt>=0.92 (Cronbach's α) on MFF-response time and r_<tt>=0.40 (Cronbach's α) on MFF-errors respectively. TV viewing time and program content were obtained through a weekly diary designed to complete a time table and identify TV programs watched, which were later classified into three types by a factor analysis (direct varimax method). The first factor was named "entertainment for adults" (e.g., variety shows, detective dramas, etc.); the second factor, "information" (e.g., news, newsshows, etc.) and the third factor, "entertainment for children" (e.g., cartoons, puppets, etc.). Channel preference obtained by a questionnaire was classified into two types, NHK (the public TV station in Japan, having two VHFs) and commercial TV (five VHFs). Ten months later the panel study which measured the same cognitive style and channel preference was carried out in order to investigate the causality of those two variables. Results: Among girls TV-I (entertainment for adults) or TV-III (entertainment for children) showed a statistically significant correlation with EFT respectively (p<.05, p<.05); in other words, the more they watched entertainment TV programs, the more they increased their field dependency. TV-II (information) also correlated with the reflection-impulsivity (close to 5% level of significance) which was categorized by double median split techniques on MFFT. Stated another way, they showed a tendency towards reflective perception in proportion to increment of TV-II (information) exposure. (See 表 5, page 149: 表 means Table). Boys who indicated commercial TV preference were inclined to be field dependent or nonanalyitc type, in contrast to the NHK preference group. Girls with commercial TV preference showed a tendency toward impulsivity in contrast to NHK preference group (see 表 6; 男 means boys, 女 girls. page 149). Although there was no relation betweeen the amount of TV viewing time and any cognitive style, significant correlations among them were found if a certain third variable was held constant. For example, among the lower verbal intelligence boys the more they were exposed to TV as a whole, the more they increased their field dependency, while the higher verbal intelligence group showed the reverse tendency. (see 表 8; 言語性上 means high in verbal scale 1 and下 low; 差 difference, z is tested on the difference between two correlations, page 151). Among girls the higher social maladjustment group tended to be field dependent in proportion to increment of TV exposure time, while higher social adjustment group did not show such a trend (表 9b). Extroverted girls tended to delay their cognitive responses and showed a decrease in errors in MFFT according to the increment of TV exposure time; the converse was true for the introvert group(表 9d, e). Discussion: It was assumed that children's TV viewing pattern would have a certain effect upon their cognitive styles. According to the results of present research, however, it seems more reasonable to interpret that any cognitive style children already possess determines their TV viewing patterns, since there were no TV programs which could help children to make analytic, field independent or reflective perception and which at the same time, aquire enough number of audience to the extent of influencing upon our findings. Secondly it is reasonable to consider some third variables mediating the relationship between cognitive style and channel preference. Among girls we found the intervening variable, the Family Communication Pattern correlated with both of these variables, that is, field independency-depencendency (EFT) and TV entertainment (TV-III or TV-I). See 図1^. (図 means figure, page 155). FCP is a scale which measured a certain parental attitude towards their children on a continuum between the polar of conciliatory or conservative and critical or progressive. The conservative family communication seems to lead their children to watch TV-III and also leads them to field dependency. Path analysis showed .the following results. When TV-III was taken as the criterion measure, the path FCP to TV-III had a significant B weight (.21, p<.05), but the path EFT to TV-III did not show the significence (図 2c; page 156). Next when EFT was taken as the criterion variable, the path FCP to EFT had a significant B weight (-.24, p<.05), but no significance on the path TV-III to EFT (図 2d). This implies that there would not be a causal relationship between EFT and TV-III, and FCP contributes to predict a part of total variance of EFT or TV-III. The coefficients of multiple correlation in this Figure are comparatively low (R=.25 to .32), which implies the existence of some third variable besides FCP, and also the question of FCP Test. It is also reasonable to assume the existence of some third variable intervening between channel preference and cognitive style. As far as the research data were concerned, however, FCP had little correlation with the either variable mentioned above. Therefore we connot refer to any third variable here. At any rate, the notion that cognitive style is an antecedent, and TV viewing pattern its consequence, or vice versa, cannot be concluded here. Another suggestion derived from this research is this correlation between the amount of TV viewing time and some cognitive styles were found when the subjects within a sub-sample were equalized in their attributes by adding one more attribute, that is, verbal ability or some personality aspect to sex difference. Particularly the studies on medium level such manipulation would be desirable. Prominent sex difference was found in our data. But as far as present research is concerned, it is impossible to refer to its cultural and social origin. The results of a panel study which was carried out ten months later showed the following cross-lagged-correlations as denoted in Figure (図3; page 158) (男 means boys, 女 girls, nummeral 5 or 6 indicate the school year, coefficients show zero order correlations.) A path analysis calculated from this cross-lagged-correlations indicated that field independency-dependency at the fifth grade did not contribute to the variance of channel preference at the sixth grade, nor was the converse true. However, in the case of reflection-impulsivity a path analysis showed that channel preference at the fifth grade could prescribe in part the total variance of reflection-impulsivity at the sixth grade. In this case the possible effect of the third variable can be considered but as far as present research data are concerned, such a variable was not found.