Historical change in the cognitive and affective language of teaching.

This study explores historical change in the symbolic system inhabited by teachers. Every discipline has its own unique system of shared vocabulary and as the discipline changes so does that vocabulary. This study examined changes in the psychological process vocabulary used in the titles of 3.145 teaching books published from 1830 to 1999. Content analysis revealed an increase in the percentage of affective and cognitive terms. It is argued that this increase reflects the increasing integration of psychological insights into teaching.

**********

The purpose of this study is to examine if changes have occurred in the psychological process vocabulary used in the rifles of books on teaching in the period from 1830 to 1999.

In their history of teaching, Exemplars of Teaching Method, Broudy and Palmer I 1970) argued that changes in teaching methods are closely related to the larger trends of cultural change. They further identified several dimensions for understanding the changing goals of teaching. One of these dimensions is the cognitive, which they define as "linguistic and other symbolic skills needed (a) for communication and (b) for retrieving knowledge stored in the symbolic systems" (p. 41).

This study explores historical change in the symbolic system inhabited by teachers. Human culture can be thought of as a shared system of symbols that shapes both behavior and cognition (Sapir, 1949/1962). Every discipline has its own culture and its own vocabulary, Cremin (1977) has written of "an authentic American vernacular in education" (p. 83) that arose during the 19th century. Content analysis is a method for analyzing symbolic culture by making inferences from texts. In recent years, computerized content analysis has emerged as a valid technique for making such inferences. In computerized content analysis, words from a text are classified into thematic categories and the relative counts are assumed to indicate the underlying emphasis of the text. This technique has been widely used in the behavioral sciences to make inferences about both individual psychological states and about the symbolic context of human action.

This study makes use of the titles of books on teaching to make inferences about changes in cognitive language used in the teaching profession.

The Ohio Library and Information Network (OhioLINK) is an organization that links the collections of Ohio's college and university libraries. Many of these institutions began as normal schools and many continue to have teacher training programs. Thus, the books on teaching in the OhioLINK system should provide a reasonable historic record of the types of books teachers have read. While it is not feasible to subject the entire text of these books to content analysis, the titles are available in the OhioLINK catalog. The content analysis of titles is an acceptable sampling technique in content analysis because titles are a kind of metatext that represents the major themes subsumed under the title (James, 1990).

Method

Data Source and Preparation

Computer files containing the catalog entry of all books on teaching were obtained from the OhioLINK system. All information other than titles was deleted from the files and the titles were organized into ASCII text files based on decades. Non-English language titles were deleted as were titles after 1999. In the end, there were 17 files covering the period from 1830 to 1999 containing a total of 3,145 titles.

Data Analysis

A computer content analysis program LIWC (Linguistic Inquiry and Word Count, Pennebaker, Francis, & Booth, 2001) was used to analyze words in the teaching titles. LIWC classifies words into 74 categories based on a content analysis dictionary. For this analysis two psychological process dimensions were of interest: affective or emotional processes and cognitive processes. A linear time series approach was used to see if changes in these dimensions represented secular trends.

Results

The time series results (Table 1) show a secular trend towards increasing affective (r = .55) and cognitive language (r = .86) in the titles of teaching books.

Discussion

This research has shown a clear increase in the use of affective and cognitive terms in the titles of teaching books. There is however an exception to this general trend, in the period from 1830 to 1869 the percentage of both affective and cognitive terms fell. One possible explanation for this decline is sampling error; it is possible that the number of teaching books in the OhioLINK collection published prior to 1870 is simply insufficient for reasonable inference. Another explanation is that the decline represents a real change in pedagogy, perhaps the move away from faculty psychology to a more conservative pedagogy.

The lower levels of cognitive and affective language in the last half of the 19th century coincide with the dominance of the McGuffey reader in Ohio education. In his forward to a reprint of the 1879 Edition, Commanger (1962) tells us that the readers were pedagogically conservative, uninfluenced by novel educational ideas from Europe, or even those of domestic innovators such as Alcott or Mann.

Why did the percentage of affective and cognitive terms tend increase after 1870? Perhaps as educational psychology became more integrated into teacher training and incorporated more complex models of human learning perhaps this was reflected in the themes emphasized in teaching books.

Hoy (12000) has noted that the status of educational psychology in teacher preparation has periodically shifted. In some periods educational psychology has been central to teacher training while in others it has been more peripheral. More over, as Hoy notes, the content of educational psychology has also changed over time. Perhaps the shift from behaviorism to cognitive psychology (Gardner, 1985) has given educators a richer vocabulary to describe mental processes.

Further studies are needed to confirm the patterns found here. It would be useful to look at similar samples taken from other states or, if possible, a national sample. In addition, it might be interesting to look at other materials such as article titles in teacher periodicals. If the pattern is confirmed, the more difficult question of interpreting its significance must be addressed. Have the changes in thematic empha,sis been linked to real changes in classroom practice or student performance? Might schooling be a critical component of some larger model of culturally induced cognitive change?

Many researchers and theorists (e.g., Ceci, 1991, Cole, 1996) have asserted that schooling affects psychological processes and that secular psychological trends observed in large populations may be linked to changes in educational practices.

It is interesting to note the drop in the number of teaching books in Ohio colleges during the 1940's. This may be related to the effects of wartime shortages on teacher training programs. In the three years following Pearl Harbor, over 7,000 Ohio teachers left the classroom. By 1944 Ohio teachers were leaving the classroom at the rate of 10 per day (Bowers, 1944). Although there was a serious teacher shortage and fewer books on teaching being taken into the libraries (perhaps fewer being published) the trend towards increased cognitive and affective language is unaffected.

Table 1
Percentages of Words, from the LIWC Categories Affective or Emotional
Processes and Cognitive Processes in Teaching Book Titles by Decade

Decade             Number of Titles   Affective    Cognitive
                                      Processes    Processes

1830-1839                 19            2.34         2.73
1840-1849                  1            2.36         0.47
1850-1859                  1            1.68         2.52
1860-1869                 17            0.41         1.65
1870-1879                 33            1.01         2.41
1880-1889                 71            1.08         2.17
1890-1899                 88            1.72         2.26
1900-1909                 97            1.19         1.27
1910-1919                126            1.68         2.56
1920-1929                241            2.04         2.51
1910-1939                195            1.94         1.89
1940-1949                111            2.21         4.41
1950-1959                170            2.18         5.16
1960-1969                289            2.42          5.3
1970-1979                544            2.14         7.23
1980-1989                495            1.95         7.94
1990-1999                621            3.87         9.85

Total number of titles  3145
                    time series r       0.55 *      0.86 **

References

Bowers, H. J. (1944). Let us consider teaching. Columbus, Oil: F. J. Heer Printing Company.

Broudy, H. S. & Palmer, J. R. (1970). Exemplars of teaching method. Chicago: Rand McNally & Company.

Ceci, S. J. (1991). How much does schooling influence general intelligence and its cognitive components? A reassessment of evidence. Developmental Psychology, 27pp. 703-722.

Commanger, H. S. (1962) Forward to Mcguffey's sixth eclectic reader 1879 edition. New York: New American Library.

Cremin, L. A. (1977). Traditions of American Education. New York: Basic Books.

Cole, M. (1996). Cultural psychology: a once and future discipline. Cambridge, MA: Belknap Press of Harvard University Press.

Garchler, H. (1985). The mind's new science: a history of the cognitive revolution. New York: Basic Books.

Hoy, A. (2000). Educational psychology in teacher education. Educational Psychologist, 35 (4), 257-270.

James, L. (1990). Titles: psychological characteristics mid document retrievability. Retrieved from http://www.soc.hawaii.edu/leonj/ leonj/leonpsy/instructor/titles2.html

Pennebaker, J. W., Francis, M. E., & Booth, R. J. (2001). Linguistic Inquiry, and Word Count (LIWC2001): A computerized text analysis program. Mahwah, N J: Erlbaum Publishers.

Sapir, E. (1949/1962). Culture, language andpersonality: Selected Essays. Berkeley, CA: University of California Press.