Among all school subjects, world history bears the greatest responsibility for providing students with a world-wide perspective. It prepares them to trace the origins of the pluralistic society of the United States; to understand the role of the United States in the world arena; to become world citizens who operate successfully on the national and international scene; and to make informed decisions which affect our relations with other nations and progress toward world peace.
A number of studies indicate, however, that students shy away from world history.1 If students do not feel that world history is particularly valuable, interesting, or enjoyable, learning is adversely affected. We are faced with a dilemma: "How to teach (the subject) to the tremendous number of students who do not want to learn it, who can find no reason whatsoever ... to study it."2
To address this dilemma, experts have proposed improving the teaching of world history by changing various aspects of curriculum design. There have been suggestions to reorganize the content of world history, to revitalize the methods and techniques used to teach it, and to diversify classroom activities and materials. A case can be made for each of these changes.
Another approach that addresses several problems faced by world history is what Stavrianos calls the "Flashback Approach."3 This approach is closely related to the epistemology of history as taught in schools, and could help improve the rating of world history among students.4 Testing the effectiveness of using this approach to teach world history is the focus of this study.
The Flashback Approach
According to Stavrianos, the Flashback approach combines alteration in the curriculum content with changes in instructional methodology. It is based on teaching the past by starting with and emphasizing the present. It begins by describing and analyzing existing conditions, and centers on issues and questions of concern to students before tracing their historical evolution through the ages. The Flashback approach touches lightly upon the events of ancient times, and stresses the more recent past and its relation to the modern world.5
In the Flashback approach, the past is used as a background to the news in order to clarify present day problems, institutions, or situations.6 However, to understand contemporary problems, students need not only a background on events and personalities, but also an understanding of the evolution of society, economy, and culture. Consequently, the curriculum organized on the Flashback model becomes an integrated social studies curriculum rather than a world history curriculum in the narrowest sense of the words.7
Beginning with current questions and dealing with them at a personal level makes history, and especially world history, more relevant and interesting to learners. By allowing students to reconstruct past events to address present concerns that they feel are connected to them in time, space, need, or interest, we exploit the psychological potential of the curriculum, while at the same time enhancing its motivational possibilities and, consequently, learning. The Flashback approach facilitates this process.8
Classroom observations, as well as research on courses, programs, textbooks, and topics organized on the Flashback principle in the social studies, have suggested that the Flashback approach has been successful in creating a positive change in students' attitudes toward social studies and in enhancing their cognitive gain in its content. Teacher satisfaction with and praise for the approach have also been reported.9 However, much of the research to date has been anecdotal, incomplete, or based on personal opinion. There has been a definite need to investigate the effectiveness of the Flashback approach in teaching world history through a systematic, controlled experimental study. The research reported here was part of such an effort.
The research project studied 250 students in ten classes at the ninth and tenth grade levels. The detailed methodology of the project is described in an appendix at the end of this article. The ninth and tenth grade levels were chosen because world history is taught at these levels in the state where the experiment was conducted. The ten classes represented ten different schools, each from a different school district. All ten districts had similar characteristics.
To ensure comparability, the schools were matched as far as possible in size, socio-economic status, ethnicity, and teaching policies. Classes were matched to the extent possible as to grade level, size, student age group, gender, ethnic distribution, exposure to current events, the time of day world history was taught, course offerings-and the teacher's gender, experience in teaching social studies, other school duties, and discipline policies. Classes were also matched on the basis of their achievement on an NAEP (National Assessment of Educational Progress) test, to make sure that their students had comparable levels of knowledge before the experiment took place. Although thirty teachers volunteered their classes for the experiment, the number of classes included was limited to ten. This decision allowed reasonably close matches to be made between pairs of schools, which the inclusion of all thirty in the study would not have permitted.
The world history unit taught focused on the Persian Gulf. It was taught as the Persian Gulf War ended. The study examined the impact of the Flashback method on both the academic achievement of students whose teachers used it and students' attitudes toward world history. To measure achievement, a specially designed test on the material taught in the unit was administered to students whose teachers used the Flashback approach and to students whose teachers used the traditional Chronological approach. The varying impacts of the Flashback and conventional approaches on the attitudes of students toward world history were measured through the level of agreement with positive and negative statements about world history presented to students in a questionnaire. This questionnaire was presented before and after the unit, and was designed to measure attitudinal changes.
The Flashback Approach
Teachers were assigned the Flashback or Chronological approach randomly after being matched in pairs according to similarities in their: (1) preparation and experience in teaching world history, (2) experience in teaching other subjects, (3) teaching styles, (4) discipline and classroom management, (5) gender, and (6) ethnic background. All teachers were trained during the fall semester to teach the unit, and all taught it at about the same time in the spring. The experimental unit itself was developed by the researcher, two doctoral students, and two teachers participating in a pilot study. The length of the unit was four weeks.
The Flashback approach first described and analyzed existing conditions in the Persian Gulf, then traced their historical evolution through the ages. The present served as a starting point to raise questions and discuss current conditions that the past helped explain. This approach was used to teach five classes.
Teachers began the unit by showing news clips from the height of the Persian Gulf War. They then helped their classes define important questions to be addressed throughout the unit. The issues were predetermined and pre-planned to ensure similarity of content coverage in all classes. Teachers used a combination of lecture, group investigation, individual and group projects and reports, discussion, role-playing, media analysis, and debate to teach the unit. Media resources and experts were also drawn upon heavily.
The assigned tasks during this unit were embedded in the context of the war. The issues raised during the introductory activity directed the orientation of the unit throughout the four weeks, with each topic starting with and remaining centered around a current issue. At various points, students were called upon to explain historical materials to their colleagues. Teachers concluded the unit with a simulation of a United Nations conference to end the Persian Gulf War. All parties were represented at the conference, and were coached to argue for their best interest.
The Chronological Approach
The second approach involved a straightforward chronological survey. Teachers started by covering the earliest periods in Persian Gulf history, and moved forward to the 1991 Gulf War. Teachers covered the same content and used the same techniques and materials as were employed in the Flashback groups. Students in both types of groups were actively involved in their own learning. The difference in the Chronological approach involved the organization of the content strictly in terms of time sequence rather than around current issues, and the resultant lack of articulation between past and present. The reason for using the same content, strategies, and materials in both groups was to make the Flashback approach, rather than any other factor, the key difference between the two groups of classes.
The Statistical Analysis
The statistical results suggested that the Flashback approach had stronger effects than the Chronological approach in enhancing students' cognitive outcomes. The results as to whether the Flashback approach also had a positive effect on student attitudes toward world history were mixed.
The descriptive statistics comparing average achievement scores for each class showed higher scores for the Flashback than for the Chronological classes.
Analyses of variance (ANOVA) and covariance (ANCOVA) showed significant differences between the two groups. The ANOVA analysis showed a difference significant at the .05 level, while the ANCOVA analysis using the scores on the NAEP test as a covariate was significant at the .01 level.
Using the individual as the unit of analysis, both a t-test and ANCOVA were significant at the 0.0001 level, providing additional evidence of the effectiveness of the Flashback as compared to the Chronological treatment.
A multiple regression analysis focusing on the individual as the unit of analysis also showed a positive correlation between the Flashback approach and achievement scores, though with a weaker level of significance (p=.14). In this analysis, NAEP scores and the degree of interest in current events were used as control variables (the latter to determine whether student familiarity with current events affected student performance in world history). Also used as a control variable was a measure representing interaction between the Flashback treatment and interest in current events. Both NAEP scores and the degree of interest in current events were positively and significantly associated with achievement scores. The interaction variable had an insignificant relationship with achievement scores.
Regarding changes in student attitudes toward world history as a subject, the results were mixed. Comparisons of class averages showed a gain in positive attitudes in both groups, though the gain was higher in the Flashback group. But neither the variance (ANOVA) measure nor the covariance (ANCOVA) measure showed any statistically significant differences between the two groups.
When the individual was used as the unit of analysis, both the t-test and ANCOVA showed a difference significant at the 0.003 level, favoring the Flashback group. However, a multiple regression analysis at the individual level, using the same independent variables as those in the multiple regression examining student achievement reported above, found no variable to have a significant correlation with positive changes in interest.
Follow-up interviews with four students per class were also conducted by three investigators (the researcher and two doctoral students) to investigate the effects of the Flashback approach. The interviews suggested that both groups developed a positive attitude toward world history, but there seemed to be more gain in the Flashback group.
In the Chronological group, students expressed appreciation based on the novelty of the topic ("hot" and controversial, current, and provoking concerns about U.S. involvement overseas), the materials, the activities, the techniques, and having experts in their classes.
The interviewees in the Flashback group expressed the same ideas, but also praised the Flashback approach itself as the major reason for their change of attitude. Sixteen out of twenty students interviewed who were exposed to the Flashback approach felt that the approach itself contributed to their attitude change. Some of their comments were:
"Starting with the present and going back to the past to bring it to the present is very meaningful; it relates past to present."
"You feel you are learning about something; you understand; you relate better; you understand why and you feel what history is like."
"You learn to deal with problems and understand them better."
"You understand the background of the current issues, the news, and the implications and complications of the U.S. involvement in the world . . . by asking beforehand the questions to be studied."
"How can you fall asleep in such a unit? There is too much suspense and curiosity to know not to be motivated and involved."
"This is nothing like history."
An attempt was also made during the interviews to check for student retention, comprehension, and mastery of skills learned in the unit. The consensus among investigators was that students in the Flashback group-more than those in the Chronology group-were able to construct hypotheses and support them logically with facts and evidence; consider and evaluate arguments; put issues in context and into perspective for their colleagues; pursue lines of discussion logically; unpack concepts into their components; link current issues to other pertinent issues across time, space and disciplines; and apply concepts to new situations.
This study has a number of limitations due to its small sample size (ten classes). Even after these limitations are considered, the trend throughout the analyses suggests that the Flashback approach manifested a stronger effect than the Chronological approach in enhancing student cognitive outcome; it may also have had some influence in changing students' attitudes toward world history, though the evidence for this is mixed.
The Flashback approach may be considered a viable and valuable alternative to traditional chronological lectures for specific units or sections of courses in world history and in social studies. One of its advantages is that it does not require special material and equipment or other extra expenditures to be implemented. The simplest use of the Flashback approach consists of reversing the order of lectures. Instead of starting from the earliest times and proceeding chronologically to the present, teachers start with the present and select evidence from the past to explain the present in their lectures. Another engaging use of this approach would be to center the world history curriculum around current issues that are relevant to student needs and interests, and to use the past as a background that can help students to understand contemporary problems.
A sample of ten 9th and 10th grade classes, forming two groups of matched pairs, was assigned randomly to the two approaches (Flashback and Chronology).
Variables on which classes were matched included grade level, size, student age group, gender, ethnic distribution, exposure to current events, the time of day world history was taught, course offerings, the teacher's experience in teaching social studies and other school duties, gender and discipline policies. All students in the best matched classes also took an NAEP test (National Assessment of Educational Progress), and class means were calculated. NAEP is considered by the literature as a good indicator of student cognitive knowledge (or lack of it). Then, the five best matched pairs of classes (on all of the criteria) were selected and divided into two groups. Group assignment was done as follows: class means were rank-ordered from lowest to highest, and then randomly assigned, one class from each pair to the Flashback group and the other to the Chronology group. Classes represented low to high ability students and various types of teachers.
The study had a randomized Block Design, which is also known as Matched Pairs Design. Statistically speaking, this design had one entry per cell. Because intact classes were used, the class mean was the unit of analysis. The subjects (classes in this case) were matched on the blocking variable (NAEP) and were randomly assigned to treatments. This model was a Mixed Effect Model where class was a random factor nested within treatment and NAEP, and treatment, with its two levels, a fixed factor. The blocking variable (Mean NAEP) was crossed with treatment.
The unit was taught at the same time, and lasted for four weeks in each class. Standardized procedures were followed in teaching the unit to both groups. On the last day of the unit, the post-attitude questionnaire was administered first, followed by the final achievement test, to prevent any test influence on student post-attitude.
Two weeks after the unit ended, the investigators interviewed four students from every class to compile more data on student cognitive and attitudinal gain. The students were selected by dividing the posttest scores in each class into high and low scores, then randomly selecting one boy and one girl from the higher level, and another boy and girl from the lower level.
The researcher used the NAEP test as a basis to match the classes involved in the study. The test included twenty world history multiple choice, objective items. The individual items were selected from a national pool, and were valid and reliable at a high level. One college professor and two doctoral students helped choose the items. An Alpha reliability test administered after the pilot study was high enough to permit the use of the test in the study (.81). Standardized procedures were used for administering and scoring the test. Two graders (doctoral students) marked every test, and the researcher served as a third grader. The highest possible test score was twenty, the lowest, zero.
The achievement test administered at the end of the unit covered the material taught in the unit and measured student cognitive gain (knowledge, understanding, and skills). It consisted of 31 items reflecting various levels of Bloom's taxonomy. The possible score range was zero to 33. The posttest consisted of long multiple choice and essay items. The multiple choice items tested for knowledge and understanding. They also presented cases for the students to solve, using concepts and skills they had learned in the unit. Essay questions concentrated on higher order thinking, such as presenting an argument and supporting it, and knowledge transfer.
Standardized procedures for administering and scoring the test were followed. Teachers and graduate students helped write the test items and develop a rubric for each essay question. They also analyzed the scores and individual items. Two doctoral students graded each test according to set guidelines. The researcher and the two doctoral students were experts in grading essay and performance questions as well as in developing scoring rubrics. The agreement between graders was higher than 80%. The posttest's alpha reliability coefficient after the pilot study was (.75), high enough to permit its use in the study.
The pre/post attitude questionnaire consisted of ten items concentrating on the importance of world history, its functionality, its pleasantness, and the ability of its methods and materials to motivate students. Five positive and five negative statements were presented to students to be rated on a five-point scale.
As with the NAEP, the questionnaire items were selected from a national pool, and had a high validity and reliability. The overall test validity and reliability were enhanced by having a professor, and some of the participating teachers and graduate students, choose the items and take the test; shuffling the items throughout the test; and using standardized procedures to administer and score the test. The Alpha coefficient value after the pilot study was (.79), high enough to warrant the use of the test in the study. A sign test was used to calculate student attitudinal gain. The lowest possible score was (-20) and the highest, (+20).
Current Events Survey
The current events survey was intended to measure student familiarity with events that might have affected performance and/or attitude. Some of its components dealt with the topic covered by the unit. The survey consisted of 13 items rated on a three-point scale, with the lowest possible score 0, and the highest 26. One college professor, two doctoral students, and some participating teachers helped choose the items used in the survey. They also took analyzed student answers after the pilot study. The Alpha coefficient reliability test was high enough (.77) to insure its use in the study.
1. See, for example, R. E. Gross and L. D. Zeleny, Educating Citizens for Democracy: Curriculum and Instruction in Secondary Social Studies (New York: Oxford University Press, 1958); M. C. Schug et al., "Why Kids Don't Like Social Studies," Social Education 48, 5 (1986), 382-388; J. Zevin, Social Studies for the Twenty-first Century (New York: Longman, 1992).
2. L. W. Rosenzwig, "Urban Life and World History: Can Social History Bridge the Gap?" Social Education 48, 3 (1982), 187.
3. L. S. Stavrianos, "Main Currents in World Thought." In S.H. Engle (Ed.), New Perspectives in World History, NCSS 34th Yearbook (Washington, DC, 1964).
4. See P. Seixas, "When Psychologists Discuss Historical Thinking: A Historian's Perspective," Educational Psychologist 29, 2 (1994), 107-109.
5. For an early postwar view, see F. M. Wesley and J. C. McLendon, "Organizing World History around Current World Affairs," in Edith West (Ed.), Improving the Teaching of World History, NCSS 20th Yearbook (Washington, DC, 1949).
6. See, for example, E. West, "Selecting an Organization for a World History Course," in S. H. Engle (Ed.), New Perspectives in World History, NCSS 34th Yearbook (Washington, DC, 1964).
7. R. W. Evans, "Reconceptualizing Social Studies for a New Millennium," Louisiana Social Studies Journal 17, 1 (1990), 26-31.
8. Gross and Zeleny, op. cit.
9. See, for example, Wesley and McLendon, op. cit.; R. E. Gross and D.W. Allen, "Upside Down but Not Backwards," The Social Studies 49, 5 (1958), 180-186; Stavrianos, op. cit.; H. Black, The American School Schoolbook (New York: Morrow, 1967); G. Davis and D. Lanshey,. "Tampering with the Temporal Order," History Teacher 5, 3 (1972), 40-44; V. J. Schlene, "History Education Reform: An ERIC/CHESS Sample," OAH Magazine of History 6, 1 (1994), 59-60; S. S. Wineburg, "Introduction: Out of Our Past and Into Our Future: The Psychological Study of Learning and Teaching History," Educational Psychologist 29, 2 (1994), 57-60.
Joseph Khazzaka is Associate Professor of Education at the University of Scranton, Pennsylvania. He is a former social studies teacher at the middle and high school level.