There Was a Great Collision in the Stock Market: Middle School Students, Online Primary Sources, and Historical Sense Making

Frances Jacobson Harris, Librarian and Associate Professor of Library Administration, Univeristy Laboratory High School, University of Illinois at Urbana-Champaign

This paper is based on a presentation at the Treasure Mountain Research Retreat IX, "Information-Rich Environments: Blessing or Curse," Brown County, Indiana, November 2001.

Students now have online access to rich collections of unmediated primary historical sources. This paper describes the results of an exploratory study of eighth-grade students and their use of online primary sources as part of an oral history unit on family farming. To understand connections between historical and present farming issues, the students analyzed photographs from the Depression-era Farm Service Administration photograph collection and wrote multigenerational family stories based on these images. Their work was analyzed for patterns in how they used the bibliographic information and how much evidence of higher level thinking skills was revealed in the work as well as their ability to demonstrate skills of observation and interpretation and to reflect a historical perspective in fiction writing. The results reveal a number of issues that have importance to the secondary social studies and information literacy fields and suggest several avenues for further research.

Troubles with Textbooks |  The Importance of Personal-Social Context |  Use of the Arts in Learning History |  The Project |  Data Analysis |  Analysis of Visual Observation Task |  Analysis of Story Task |  Conclusion |  References

Primary sources can easily mystify the novice learner (Afflerbach and VanSledright 2001). Many K–12 educators assume that only very advanced students of history can be turned loose in an archive and make meaning of what they find. It has hardly mattered, however, because student access to primary sources has traditionally been very limited. Historical archives do not typically open their holdings to visiting troops of middle school students, whose teachers can hardly justify a labor-intensive field trip when they do not feel sure of the learning outcome. But the Internet has transformed these possibilities. Anyone who can click a mouse now has access to a mass of primary historical material. K–12 teachers can introduce their students to such compelling resources as the Library of Congress (2002) American Memory collections (, Calvin College’s (Bytwerk n.d.) German Propaganda Archive (, or the Smithsonian Institution’s (2002) Library and Archival Exhibitions on the Web ( As K–12 teachers inevitably turn to rich resources like these, they must determine how best to incorporate primary sources and provide the meaningful context students need to enhance their historical comprehension and sense making.

Troubles with Textbooks

There are a number of sound justifications for the use of primary sources in the K–12 setting. One argument is that a dependence on textbooks can compromise student learning. McKeown and Beck (1994) observed that teachers often assign textbook readings without contradicting student assumptions that the textbooks are objective and omniscient conveyers of these events. Textbook accounts also typically fail to represent uncertainty in historical problems and may not completely know the facts or may partly reconstruct them from indirect sources; in addition, controversy may exist, meaning that textbooks may tell different versions of the same sequence of events (Britt et al. 1994).

McKeown and Beck (1994) noted another problem with textbooks and student learning, which is that their authors often assume students have more prior knowledge than they actually do. Students’ lack of adequate context results in a shallow understanding of the historical phenomena. McKeown and Beck (1994) draw on Kintsch’s (1986) distinction between learners’ ability to recall "textbase" information as opposed to their ability to form a "situation model" of events. In other words, students may learn enough from reading a text to reproduce what they have read, but they have not learned in such a way that they can use the information from the text to interpret or reconstruct it (Kintsch 1986). In addition, many textbooks are often not engaging and suggest no voice in the way trade books do. But textbooks are traditionally at the center of the classroom curriculum, endowing them with an undeniable authority and a “beyond reproach” objectivity (McKeown and Beck 1994). Students are unaware of the role of the historian or the textbook as interpreters of historical fact, rather than mere relaters of memory. Garner and Gillingham (1998) commented that textbooks are seldom read selectively in schools, that instead a start-to-finish treatment is more common. They described the “textbook-as-tyrant trap,” where the curriculum and pace are driven by the book in a transmission model of teaching.

In studying historical problem solving methods and skills, Wineburg (1991) asked high school students and professional historians to look at a range of conflicting written and pictorial sources about a single historical event and construct the true history. The students gave most credence to the version of history supplied by the textbook, in contrast to the historians who were immediately inclined to examine the credibility of the sources, whether primary or secondary, and gave little attention to the textbook treatment. The historians also relied on two other criteria: corroboration, the act of comparing sources to determine reliability; and contextualization, situating the claims of a document in the particulars of the historical events. Students, on the other hand, viewed the textbook as a primary source offering unbiased factual information. For historians, the question was not if a source was biased, but how its bias influenced the quality of the information.

Rouet et al. (1998), who studied college students’ ranking of primary and secondary historical information, confirmed this tendency of students to give high marks to the credibility of textbooks. But the students in their study also recognized the value of primary documents. The researchers found that proper preparation helped student understanding:

In particular, when students are told explicitly that different points of view exist about the issue at hand, and when they are told this before engaging in the documents, they can come to discern some sources as more useful and trustworthy than others (Rouet et al. 1998, 104).

These researchers noted that aside from the information that is contained in a primary source document, the reader must look at the subtext—the other attributes that affect the meaning and significance of the document. Subtext elements include the author’s identity, date of publication, and whether the document is private or public, official or unofficial, refereed or not. For a photographic image, the viewer needs to add intended audience, the purpose of the photographer, any possibility of staging, and so on. Their research points to a progression in the development of historical thinking. Student comprehension proceeds from sequential thinking (i.e., “first this happened, then this happened, then this happened”) to a more robust view of historical events that takes into account the simultaneity and complexity of historical processes. With experience, coaching, and domain knowledge, students can learn to contextualize, scrutinize the nature of a source, and use a broader common sense.

[ Back to Top]

The Importance of Personal-Social Context

Levstik and Barton (1996) did not think that Wineburg’s (1991) expert versus novice paradigm, in which he compared the document evaluation strategies of historians and students, was very useful. They were more interested in the cognitive processes students use to make sense of history, which they found generally involve the mediation of a student’s social context. Personal history, family background, socioeconomic status, and ethnicity may be just as important factors for historical sense making as prior schooling. Fasulo, Girardet, and Pontecorvo (1998) observed that the nine-year-old children in their study defined Viking artifacts in terms of themselves, overlaying their own perspectives and sensibilities on the past. These behaviors confirmed that personal experience effects the way we see the past and determine what is significant. Yet the children also recognized the Vikings’ place in time. Revealing a presupposition that the passage of time is equal to progress in civilization, they expressed appreciation for what the Vikings were “already” able to do (Fasulo, Girardet, and Pontecorvo 1998). The children created bridges between personal perspectives and historical perspectives.

Greene (1994) found that novice historians are also more skilled at writing problem-based essays requiring synthesis of a finite amount of information than traditional history reports based on extensive content knowledge. This difference is due to students’ lack of disciplinary knowledge, leaving them better positioned to write analytically by drawing on a finite collection of resources and personal experiences. Therefore, he recommends that educators “…give students occasions to write informally as a means for helping them to explore their ideas and for acquiring new knowledge” (Greene 1994, 166). This use of writing as a technique for learning content knowledge adds to the repertoire of strategies for teaching history.

[ Back to Top]

Use of the Arts in Learning History

Since children are concrete thinkers, relating strongly to material culture and the personal-social context, it can be argued that history instruction should include artifacts that can be seen (Levstik and Barton, 1996). But how can children then articulate their understanding of the artifacts and other historical information they are exposed to? One constructive way is through the arts. K–12 teachers have long employed the arts in teaching social studies and language arts subjects (Henry 1995). Researchers have validated the learning outcomes of this practice. Epstein (1994) notes that “…by representing history and other conceptions of knowledge through paintings, poems, stories, or songs, students can communicate the lifelike qualities of human experience” (136). Levstik and Barton (1996) found evidence to suggest that some students remember historical information better through art experiences because of repeated practice and multiple performances, as in the case of drama. They argue that learning through the arts provides children with a context in which they are active problem solvers. Furthermore, story telling, or narrative, has a powerful influence on historical thinking and can be accomplished through reading and writing historical fiction and nonfiction (Levstik and Barton 2001).

At upper levels, however, use of the arts in the teaching of history is not as common, often seen as less academically rigorous. But we do know that various other implementations of constructivist learning, such as historical simulation exercises, can be powerful motivators and effective learning tools (Zola and Ioannidou 2000). Equally significant, for those students who are not as adept at traditional forms of assessment such as multiple choice tests and essays, “…stories or songs enable them to succeed in representing what they have come to know through forms they have the talent to manipulate well” (Epstein 1994, 140). The arts provide these students with an alternative method to demonstrate their historical understanding and competence.

Primary sources, in combination with artistic representation, have been shown to be valuable tools in helping students situate historical social context. What happens when online primary sources are used in inquiry-based classrooms? In transmission-based instruction, teachers can easily update their lessons by integrating primary source materials into PowerPoint presentations rather than passing an artifact or a book around the classroom during lectures. But what happens when students are asked to search an electronic archive and select their own primary sources for interpretation? Are they able to draw appropriate inferences and connect the threads of evidence they uncover? Can they use creative modes of expression to communicate their understanding of historical social context? Bass and Rosenzweig (1999) advocate the novice in the archive approach to inquiry-based learning, particularly in the case of digital materials. They observe that:

The unique opportunity with electronic, simulated archives is to create open but guided experiences for students that would be difficult or impractical to recreate in most research library environments. It also frees students and teachers from their traditional dependence on place for first-hand social, political, or historical research. Or perhaps more importantly, it means that students can more readily compare their own community with others, more distant (Bass and Rosenzweig 1999, 49).

These questions concerning the use of primary sources resulted in the following effort to explore the issue in more depth.

[ Back to Top]

The Project

This paper looks at a group of fifty-three academically gifted eighth-grade students and their use of historical photographs for a creative writing social studies assignment. These students were preparing for an oral history unit on family farming. Although their school and town are situated in the middle of Midwestern farm country, most of the students had little or no connection to farming in their personal lives. No ready textbook was available. Here was a situation in which teaching materials could be culled and assembled by the collective efforts of the teacher, the school library media specialist, and the students. The project offered a good opportunity to conduct informal action research on middle school student use of primary sources in an inquiry-based history classroom. The research was conceived as an effort to improve teaching while observing and reflecting on how students develop history sense-making skills through engagement with primary sources and creative writing.

In brief, the project was sequenced as follows:

  1. The teacher taught lessons on the recent history of farming in the United States, starting with the Dust Bowl era and the government programs of the New Deal and continuing to the state of present-day family farming.
  2. The teacher supplemented her lectures with a wide variety of reserve readings about modern farm life culled from newspapers, magazines, and books. Students completed a small questionnaire after finishing each reading.
  3. The school library media specialist introduced students to an online archive of Dust Bowl–era photographs from the Library of Congress American Memory collections, and taught searching and visual literacy skills.
  4. Using a photograph from the archive, students wrote short stories bridging the history between the government programs of the Depression era and the current status of family farming in the United States.
  5. The classes took a field trip to visit area farms.
  6. The students conducted oral history interviews with area farmers. They recorded the interviews and edited them. A small cadre of students then produced a thirty-minute radio program for broadcast on the local public radio station.

The focus of this particular research is on the third and fourth steps of the project, during which students used the archive and wrote their short stories. The creative catalyst for the stories was provided by the black-and-white documentary photographs from the Farm Service Administration–Office of War Information (FSA–OWA) collection on the Library of Congress (1998) American Memory Web site (, which contains more than 112,000 photos, including such well-known images as Dorothea Lange’s Migrant Mother. As the school library media specialist, I taught a lesson on using the collection and spent several days in the computer lab with the students as they searched, selected photographs (see sample 1), and wrote their stories. My lesson included instructions in how to search the collection by subject and key word and how to analyze a photograph. Student also were taught how to view neighboring photographs by clicking on the option to display images with neighboring call numbers, which brings up thumbnail images of the photographs that appear on the same negative strip. In many cases, the surrounding images help build a sense of context. I modeled all aspects of the search and selection process, then led the class in group analysis of a photograph. Together, we first described what we saw, without attempting interpretation. Then, using those visual details, we developed inferences and interpretations that might describe more fully what was happening in the photograph. After this lesson, the students—working in pairs—selected a photograph, conducted the visual analysis exercise, created their own title for the photograph, and then wrote a story that would take into account a person in the photograph and the succeeding two generations. The students chose their own partners, which resulted in all same-gender pairings. The entire project took about a week of class time.

The visual analysis portion required students to answer the following questions:

  1. What is happening in this picture?
  2. What are the circumstances this photo represents?
  3. How are the people dressed? (Be specific in your descriptions.)
  4. What can you observe from the expressions on their faces, posture, position in the photo, etc.?
  5. Describe the setting. What do you notice about the room (furniture, walls, etc.)?
  6. Is there anything interesting or surprising about the situation in the photo? Given what you know about the era, how might you explain it?

The questions in the visual analysis portion of the assignment prompted two different types of thinking: questions 1, 3, and 5 called for straightforward, observation-based description. Questions 2, 4, and 6 led the students to analyze and interpret what they observed. One purpose of these questions was to hone students’ visual literacy skills by directing them to carefully read the photograph as a piece of primary historical evidence. That process was modeled and practiced during class. Another purpose of the questions was to help structure student thinking as a way of preparing them to conceptualize a bigger picture. They needed to think deeply and analytically about the photograph before launching into the creative writing portion of the assignment. Their task for the story was to invent plausible, causal relationships that would take place from the time of the photo to the present. In a sense, the students were supplied with an out-of-context artifact, then were charged with supplying a meaningful, historically appropriate context for it.

In their stories, which had to be at least two pages long, the student pairs were to address how the person in the photograph survived the Depression, indicate if (and how) the next generation(s) stayed on the farm, and describe how the legacy of the farm stayed (or did not stay) with the succeeding generations. They were given two class days to work on their stories, and were allowed to use out-of-class time as well. The actual assignment parameters were quite open ended. Students were not required to include a specified number of historic facts or events in their stories. Instead, they were instructed to write stories that reflected the historical context and realities. The purpose of the exercise was to help students personalize the scope of farming history so they would be more receptive and engaged participants in the oral history portion of the project. Their final product for this portion of the unit included a copy of the photograph ( sample 1) with its new title, a copy of the bibliographic information page ( sample 2), answers to the visual analysis questions, and the story.

As discussed earlier, textbooks appear to speak with a single voice, portraying events from a tidily edited vantage point. A variety of perspectives contribute to any single historical event, which means there are also multiple possible representations of the event—whether contemporary to the event or created after the fact. Each individual source reflects one view, one glimpse, and must be coupled with other sources, “because a situation model cannot be formed by reading a single document” (Rouet et al. 1998, 98). A situation model is a representation of the situation described by a text and not the characteristics of the text itself (Kintsch 1986). Multiple texts (or sources) contribute to the development of much richer situation models, and thus, deeper historical understanding. For able readers, the practice of reading from multiple texts—even those with several challenging features—provides the conditions that can promote historical thinking and more complex understanding of the past (Afflerbach and Vansledright 2001). This project was designed to expose students to just such multiple historical perspectives. Students acquired background information from the teacher and from a wide variety of short readings from which they could select, including news articles, tables of statistics, short fiction, and trade publications. Exposure to the breadth of the FSA–OWI collection was another way to supply students with multiple perspectives. Searching the photograph collection exposed them to contemporary visual records, adding further dimension to their knowledge base and giving them the means to formulate a robust situation model. At the same time, they would have control over the content they would ultimately work with.

[ Back to Top]

Data Analysis

Twenty-seven stories and their accompanying visual analysis sheets and bibliographic records were made available to the researcher. Because this project took place in a naturalistic classroom situation, without the parameters that would normally be established for a controlled study, it is not possible to make global generalizations or conclusive recommendations from quantitative analysis of the student work. Instead, the findings presented here can serve as a set of tentative observations that suggest questions for further empirical study.

The student-created artifacts were analyzed on a number of levels. The visual analysis task was studied to obtain a sense of how students differentiated between observational tasks and interpretative tasks, and to understand the students’ varying visual literacy skills. The stories were examined for the presence of certain types of historical representation and recurring characteristics and themes in writing style and historical sense making. The intent of the analysis was to let the stories speak for themselves, not to test a hypothesis or anticipate trends. This was accomplished by using a naturalistic method generally based on interpretive traditions (Miles and Huberman 1994) and grounded theory methodology (Glaser and Strauss 1967). In this case, that meant employing an iterative, quasi-ethnographic technique of reading the texts and identifying themes, then rereading the texts to verify themes and identify new ones, and so on. Although in-depth statistical analysis is not appropriate given the study’s limitations, some aggregate data is available to shed some light on the overall picture.

[ Back to Top]

Analysis of Visual Observation Task

There is a well-known cliché that says, “Pictures don’t lie.” But neither do they tell the entire truth. Fasulo, Girardet, and Pontecorvo noted that photographs appear to possess tricky features (1998). A viewer has to consider the photographer’s intentions in terms of a notion of relative objectivity and be aware of techniques that color the meaning of a photograph, such as the framing of the subject and the use of light. Finally, photographs capture a single instant; there is no before and after, no around (Fasulo, Girardet, and Pontecorvo 1998). Students seemed to struggle in their efforts to separate the observation tasks in questions 1, 3, and 5 from the interpretive tasks in questions 2, 4, and 6 (see table 1). Only 63 percent of the observation task questions were answered in strictly descriptive terms, while 91 percent of the interpretive questions were answered appropriately. Unfortunately, the language of question 1 contributed to the students’ confusion. “What is happening in this picture?” should have been worded “What do you see in this picture?” In general, though, students found it difficult to simply describe without judging or drawing inferences.

One pair of students wrote this observation about a photograph: “They only have enough silverware for the parents, so the children eat with their hands” instead of “The parents are eating with silverware, and the children are eating with their hands” ( sample 3). Levstik and Barton (1996) observed that children connect what they see to patterns from their own lives or from other familiar sources, a process called intertext. The authors identified family story telling, family activities (including trips to historical sites), popular culture, fiction and nonfiction trade books, and school instruction as types of intertext, or referents children use to make sense of historical information. Intertext best supports children’s understanding of material culture, the objects and tools of everyday life, rather than more conceptual aspects of historical change over time. This silverware example illustrates an intertext phenomenon. In our students’ experience, not having enough silverware for all family members would have been the only plausible explanation. It may also be that students thought they were not doing enough by simply describing what they saw.

These students were probably also caught between the instructions to remain descriptive, their own need to make connections to familiar signposts in their lives, and a developmental inclination to go to the next step. Wigfield, Eccles, and Pintrich (1996) noted that changes in cognition and achievement occur during early and middle adolescence:

. . . the most important changes to note are the increasing ability of children to think abstractly, to consider the hypothetical as well as the real, to engage in more sophisticated and elaborate information-processing strategies, to consider multiple dimensions of a problem at once, and to reflect on oneself and on complicated problems (151).

Accordingly, when asked to be specific in describing how the people in the photograph were dressed, one group wrote: “His shirt is very baggy, as though it was not made for him.” The first half of the sentence meets the description requirement; the second half of the sentence interprets and illustrates. The reader can visualize just how baggy the shirt is.

Students sometimes went beyond interpretation, leaping to conclusions or embellishing in some way, as in this example: “The little boy looks sort of lost and spacey in this picture. He is also kind of hiding, perhaps from reality.” This type of description may be a characteristic of adolescents’ ability to consider a great deal of information, but still need experience and practice to use it (Wigfield, Eccles, and Pintrich 1996). The academically gifted students in our sample were no different in this regard. In general, student descriptive skills ranged from being very flat to highly complex and detailed, reflecting varying rates of cognitive development. There was also some indication of gender difference in this skill, with boys appearing to use fewer adjectives, though more rigorous study would be needed before a definitive claim could be made of this observation.

In responding to the interpretive questions, many student groups drew overreaching and rather absolutist conclusions. However, others showed signs of higher order thinking, demonstrating awareness of the subtext (Rouet et al. 1998), the photographers’ intentions, and the tricky features of photographs (Fasulo, Girardet, and Pontecorvo 1998). These student pairs used qualifying vocabulary like “probably,” “seems to be,” and “appears” in their analyses. The most sophisticated students mentioned an awareness of the photographer and his or her possible influence on the subjects’ demeanor in the photograph:

These people are showing hardly any emotion, and the children don’t even seem to care that their pictures are being taken.

Since they don’t have a frown on their faces, but not any smiles either, we can conclude that they are unsure of how to react. This is probably the first time they have been photographed.

Several student groups reached conclusions that could be considered reasonable based on the evidence at hand, even though a more experienced viewer might draw different inferences from the same evidence:

The people are dressed shabbily. This picture was taken in January, yet all of the children are wearing short sleeves.

In this particular case, the family is standing next to a wood-burning stove. It is unlikely that the students in this cohort would have direct (or even indirect) experience with the heat such a stove generates in a small space. The same students drew another interesting conclusion:

She is also wearing shoes and socks, which shows that she is not from a very poor family. Shoes would definitely come after food.

Question 6, which asked students to describe anything interesting or surprising in the photograph, elicited some of the most compelling responses:

None of the subjects in the picture are looking at the camera. Film was fairly expensive in those days, so this kind of surprised me. But it probably makes sense, since the photographer is trying to capture normal farming life during the Depression.

Here is another example of students, who are accustomed to smiling for the camera, inserting their own sensibilities into the historical scene. If these contemporary student writers were the participants in this scene, they would likely try to support the photographer’s intent of capturing normal farming life during the Depression and not pose. But the cost of the film most likely had no actual import for the subjects of the photo, who were probably never going to see the results. It does seem significant, however, that the students had an awareness that the presence of the photographer might influence the situation. With some coaching or classroom discussion, the students might note that the photographer was a stranger to the household, perhaps representing outside authority, and that his subjects might have been reluctant to make eye contact with an outsider.

Other student groups recognized and wondered about extremely subtle cues:

What is surprising in this picture is the contrast between the family’s appearance and the house and conditions that they live in. I think a logical explanation for this is that although the family is not very wealthy, ultimately judging from their house, they want to look their best for the government representative.

And they further speculate:

There is a fire insurance banner on the house, and yet the house would probably burn down with a small flame. Why provide insurance? This possibly was part of a New Deal package, or maybe the poster was found by the man somewhere and used to repair or cover something. Also peculiar is that FDR’s New Deal promised prosperity, but this man was still on the lines of poverty.

These two excerpts reveal a nuanced understanding of the human response to historical-social factors.

[ Back to Top]

Analysis of Story Task

The stories were first examined to determine how much, if any, of the bibliographic content from the archive was used. Next, the stories were assessed for the nature of history learning that was expressed.

Use of Information from the Bibliographic Record

It is worth taking a moment to comment on the online environment in which the students were making their photo selections, as it contains historical cues we anticipated the students might draw on for their stories. Students generally searched by keyword (using such terms as “Illinois” and “farming”) or browsed the list of subject headings (they were prompted to start with the “Illinois” headings). They also scanned through adjacent pictures similar to the way a library user would scan a shelf of books in a particular section. All groups had selected their photographs by the end of the first day in the computer lab. The bibliographic record for each photograph in the FSA–OWI collection generally includes a caption, typically assigned by the photographer, as well as other salient information, such as the date the photograph was taken, the location, and the name of the photographer. The captions are sometimes lengthy (see sample 4), providing helpful explanations of the activity in the photo. We were curious to see if students would incorporate any information from the bibliographic record or if they would choose to devise their own explanations.

Thirteen of the twenty-seven stories included factual information from the bibliographic records. Students used the same family names (if available), the same geographic locations, and referred to the same government programs (e.g., FSA support, a WPA job, etc.). In seven stories, the students did not use any of the bibliographic information. The general framework of their stories fit the subject of the photograph (for example, the story would include a father and son struggling against the wind in a dust storm), but none of the specific details from the bibliographic record were incorporated. In five stories, the students appeared to pick and choose. They selected some of the information from the bibliographic record (for example, a location or family name), but dropped other pieces or invented details that contradicted the caption. One story was turned in without the bibliographic information and one story’s photo contained no caption.

When students were searching for a photograph to select, many of them used the collection’s display images with neighboring call numbers feature. We were interested in finding out if the students took advantage of these adjoining images to ferret out further background information to use in their stories. Direct evidence pointed to only two student groups, who mentioned the accompanying photos in the visual analysis exercise. The students apparently used this feature largely as a navigation tool to browse for photos. In the final analysis, it appears that students may have used data elements from the bibliographic records as anchors or to add a touch of verisimilitude, but apparently preferred to invent the specific circumstances of the stories themselves.

History Learning Outcomes

The students’ stories were examined for two types of learning. First, what historical information were students able to convey through their stories? What did the stories reveal about farm life during the Dust Bowl period? Were the students able to assemble plausible sequences of events spanning three generations? Second, what meaning did students make of this period and these themes in American history? What did they understand about human conditions under these circumstances and over this period of time?

Historical Information

Because the students were not assigned fixed rubrics that would demonstrate their knowledge of farming history facts, it is not useful to draw fixed conclusions about the quantity of history that was learned from the story writing task. However, there are certain events that are so strongly associated with farming during the Depression that their omission might indicate a lack of fundamental historical knowledge. Each story was examined for mention of at least one of the following historical benchmarks: dust storms, the stock market crash of 1929, or the role of the federal government (for example, a relief program, a government agency, or President Roosevelt). Of the twenty-seven stories, only five did not mention at least one of these factors. However, it is hard to say that these five stories were devoid of historical verisimilitude, their authors having focused more on the personal than the political—which could indicate that they did not see a connection between the two. Sixteen of the stories incorporated the role of the federal government, either by invoking specific names (such as the Agricultural Adjustment Act, the Farm Security Administration, or Hoover) or making general references (“things are better with money from the government”). It is interesting to note that in six of those stories, the government is portrayed in a negative fashion. For example, in one story, members of a family must leave the farm because they cannot pay their taxes. In two of those stories, the government is an agent of both good and bad—in one story, first giving and then taking away land, and, in the other, providing a WPA job that does not pay a living wage.

The stories also contained historical error, which can be divided into two types. The first type—of which there were only a handful—reflects a faulty recollection or a misunderstanding of material that was covered during the unit. For example, one story described a family that took a train to California after losing its farm. Although there probably were some displaced farm families that could afford train travel, the predominant picture presented to the students was of families without resources, who would be more likely to struggle across country in rickety automobiles. The far more prevalent kind of historical error was a type based on lack of information, in which students appeared to speculate without possessing adequate background in the subject. In one case, a family in the dry Bakersfield climate raises rice. In another, a fifty-year-old father enlists in the army (when he would have been rejected based on his age).

Ten of the stories contained what Levstik and Barton (1996) termed ahistorical elements. These are historical connections or links learners weave that do not rely on historical information. In their study, Levstik and Barton observed young children construct contexts they derived from their own life experiences and observations, rather than from historical information they were given. This technique helped the children make sense of events in a context they could understand. Our students were older and had more of a sense of “history-ness,” but many still apparently needed to view past events in a context that made sense to them. In these ten stories, students created context by overlaying their own contemporary standards on historical time frames. For example, several stories featured girls who inherited and ran the family farm—a scenario relatively rare during the nineteen-forties and fifties. Other stories included first generation Dust Bowl daughters who went to college, even agricultural school. A pair of student authors wrote a story about an African-American family that succeeded beyond probable possibility for the time period; in yet another about an African-American family, race played no part in their fortunes whatsoever—an equally unlikely scenario. The students who wrote a story about a family with ten children observed that, “There was a much higher birth rate because there were no abortions then.” Another group described a woman who “… didn’t have any children as she felt that it was a way of turning women into incubators.” In both cases, modern sensibilities were being applied to Depression-era situations.

Unfortunately, the first draft of these stories was also the final draft. A simple solution would have been to require subsequent drafts or assign students to critique one another’s stories. As it was, many of the stories suffered from narrative lapses of various sorts. Most of the stories were full of detail about the first generation, the one represented by the photograph. After this heavily invested opening, the writers often seemed to run out of steam and rush through the next two generations. As a group, these students provided few descriptions of modern farm life, with its advances in technology and the growing predominance of corporate farming. It could be that the photograph was a concrete catalyst that sparked better attention to detail when describing the Depression period. The students had no visual prompt to help them ground the later periods of the story with realistic detail. In addition, more in-class time was spent learning about the Depression period than the modern period. The major source of information about modern farming up to this point had been the assigned readings, and the students had not yet conducted their interviews with actual farmers. This would suggest that students learned more from the teacher-led instruction than they did from the independent reading. It might also indicate that the visual images left more of an impression on the students than did the texts.

Other narrative lapses may be due to naïveté. In one story, the authors described a fifty-six-year-old mother and a sixty-year-old father who have five children ranging in age from four to sixteen. In another, a mother, described by the students to be in her late fifties, has a three-year-old. Old mothers were a common theme, perhaps because many of the women in the pictures look old by modern standards. Sometimes students simply did not seem to be thinking through the task with care. In one story, a journey to California took five years, from the beginning to the end of World War II. A character in another story fell in love September 10, 1937. By September 13 he had invited his paramour to for dinner “quite a few times now,” and asked her to marry him by September 24. Two years later the couple had three children and had established a farm with soybeans, corn, cattle, and dairy products. Again, second drafts would have helped with problems like these.

Meaning Making

The purpose of our unit was not only to teach students to analyze historical artifacts, but also to create personal historical perspectives based on what they had learned from diverse sources. We selected historical fiction as the vehicle students would use both to develop as well as to demonstrate this historical meaning making. In order to write successful stories, our students needed an ability to comprehend—even place themselves in the midst of—the various perspectives of historical figures and events. This developmentally based skill rewards the learner by endowing the content with an intrinsic value. Robert Selman (1980) describes perspective taking as " . . . including an understanding of both the coordination of the perspective of self and other and the nature of man as a perspective-taking animal” (65). In other words, the learner must not only be able to understand the personal perspectives of historical figures, but must also recognize those figures as actors able to assess the perspectives of others as they make decisions about their actions. Selman observed that this third-person and mutual perspective taking are learned around ages ten to fifteen. In-depth perspective taking, in which the learner can place perspectives within a societal or symbolic context, begins to develop as early as age twelve.

In our setting, creative writing was selected as the most useful form for encouraging personal investment and perspective taking. We hoped that this medium would promote a sense of connection to an experience that was otherwise outside of their usual realm of knowledge. With that goal in mind, the requirements for the story-writing component were designed to give students as much latitude as possible. We did not limit the literary devices students could use to tell their stories. Most chose a first- or third-person memoir style. Some wrote diaries, while others used a flashback format. One group wrote from the perspective of a grandson telling the family story, a particularly effective technique. Many groups used dots or dashes to separate generations or mark the passage of time. We also observed, as did Epstein (1994), that a few of the students who were not otherwise the best performers on traditional academic tasks seemed to particularly excel in this atmosphere of artistic creativity.

One powerful way some students demonstrated a sense of perspective was by incorporating the photo session into the narrative, also done in some of the visual analysis exercises:

In a family photograph taken by a man named Lee Russell in May of 1938, you can see disappointment in the faces of my dad’s family.

The boy stood by his grandmother, at the doorway, facing the photographer. The woman wanted to show others what is (sic) was like living in a shack town. After shooting a few pictures on the big, clumsy camera, the photographer thanked them and went away. He was embarrassed and angry with the photographer for taking the picture. He looked down at his clothes that he wore every day . . .

Another way to ensure multiple perspectives was to have learners supply them through the discourse and negotiation that takes place during group work. Fasulo, Girardet, and Pontecorvo (1998) noted that group discussion and joint decision making are the best conditions for encouraging the display of knowledge, causal reasoning, and imaginative thinking. By exercising skills of argumentation, students must take into account other points of view. The group itself should provide the necessary scaffolding for the cognitive explorations of its members (Fasulo, Girardet, and Pontecorvo 1998). Our students were assigned to work in pairs to maximize the potential for developing the richest possible understanding of historical issues. These teams had to negotiate the tension between the photographer’s captured moment and their need to generalize to a broader historical explanation and narrative. The teacher and I hoped that the group interaction would cause productive argumentation, resulting in alternative theories and inferences about the circumstances portrayed in the photograph.

It was hard to tell from this single experience how effective the collaborations were. However, in the one case where a student missed class and had to complete the assignment alone, his story rather quickly departed from reasonable historical plausibility. In another case, the two students apparently split the story-writing task, stapling the two halves together. One partner covered the Depression period and used a story line. The other partner composed a narrative that resembled a biblical genealogy: “Jon was given two children Maya, age five, and Ben, age eight. Nicole went to college and became a biologist. She had one child, Mariel, age eighteen.” In both these cases, at least, the absence of collaboration led to a real qualitative deficit.

Most of the stories fell into one of three general affective stylistic categories: humor, melodrama, and realism. These styles were not mutually exclusive, but one typically dominated the tone of each story. The use of humor ranged from outright punning to the use of irony, or to just a simple sense of the fanciful:

Monica marries Peter Paul, who is divorced from Mary.

He considered himself an outsider, and wanted to go east, where he had heard of the “stock market of gold.” But of course, at the age of four, these were only dreams.

Joan’s daughter Faure was adopted by Pachelbel, Joan’s sister.

There are several possible reasons students chose to use humor in their writing. Some may have used it to give themselves psychological distance from the seriousness of the topic. On the other hand, humor may have been used as a device to help students feel more connected to the events, yet still have control over them. The two musical examples above could be interpreted as student-connected intertext, in which the writers import elements from other media with which they are familiar. The use of humor may simply reflect classroom culture, since the task was grounded in a literary endeavor rather than a historical one. The parameters of the story assignment were an invitation to invent, not simply report or even analyze data as in traditional assignments. It might even be said that the students were able to take advantage of a delicious opportunity to share an inside joke. In these safe circumstances, the teacher knew that they were fooling and they knew that she knew they knew. This behavior may be another demonstration of development in perspective taking (Selman 1980).

With the modern generation segments of the stories, the students seemed to feel even freer to invent, again perhaps because they had no other visual cue to anchor the narrative. In one story, a daughter becomes an internationally regarded conservationist. In another, the daughter travels to the Persian Gulf as a lieutenant. Some fabrications were written at the expense of plausibility. One story tells of a son who gets married in 1964 to a woman who turns out to be an animal rights activist and convinces him to quit pig farming, then to quit farming altogether. The couple goes on to form an organization called the “Animal Rights Society, National,” also known as ARSN. This playfulness probably made the assignment more enjoyable and engaging for many students, and gave them a chance to exhibit their cleverness. An advanced sense of humor is a common characteristic of gifted students (Holt and Willard-Holt 1995). Other explanations not considered here might also account for the imaginative spin present in the stories.

There were also plenty of examples of unintentional humor, due to typing and grammatical errors or to simple naïveté. One pair of students intended a serious title for their photograph: “Life for a Tenant Farmer’s Family Living in Oklahoma During the Depression in a Nutshell.” Other bloopers would qualify for any teacher’s list of favorite student mistakes: “Especially nobody would want to buy our desecrated farm”; “The depression and dust bowel had hit his family hard”; “In a few years the meat they sell from the cattle flourishes”; “They disclosed Ralph from his inheritance.” As mentioned earlier, requiring second drafts or having students proofread each other’s stories would have eliminated many of these gaffes.

A more unfortunate consequence of the open-ended nature of the story-writing task was that some students either became sloppy in their attention to important detail or lapsed into heavy melodrama. There was often a fine line between clever humor and silliness:

Eventually, she died of a massive heart attack when her grandson asked if he could borrow five dollars.

Alexis had gone to jail for grand theft auto.

Pa came home. Alone. Pa was not injured but was shaken up about my brothers’ deaths. He quickly got back to work, but Ma was grief-stricken. She stopped eating and eventually starved to death.

The tendency toward melodrama could be considered developmental, reflecting students’ lack of life experience. These writing examples bear a resemblance to other mass media fare, such as television soap operas or Hollywood movies, that models students might imitate when they write about experiences vastly different from their own. Their cinematic story lines could also signify a need early adolescents might have to oversimplify complex or unsettling phenomena, minimizing the impact on their own comfort levels—just as the use of humor does.

The stories that were written in a realistic, almost documentary style were probably also the most historically correct. This modern farmer’s story is steeped in realism:

If it wasn’t for my wife’s job, I would never have been able to pay for the farm, equipment, seeds, sprays, and other expenses.

But realism does not preclude the inclusion of humor or the temptation to stretch things a bit:

Charles got experienced in growing corn. He grew sweet corn and field corn. He started to sell corn to chip companies, his corn was used to make plastic, and he soon married the rich daughter of a senator. They had a daughter named Nancy, who they taught how to farm.

His daughter, Nancy, was also interested in the farming business, and had her father’s ingenious instincts for making money in the farm business. She bought plows and farm machinery, and dug irrigation canals so that her crops expanded. She sold her corn to cattle feed companies, and also to make ethanol. She was also able to grow soybeans in abundance and she soon made enough money to start her own soybean oil company.

Being able to incorporate realism and humor might signify the most mature level of historical understanding.

The role of personal investment in learning was probably compromised somewhat by what students thought they were supposed to write. Even with the latitude the students were granted and the resulting wide range of writing tactics they used, the nature of formal schooling is that implicit boundaries are prescribed and outcomes are often forced. Some students clearly responded to what they thought their teacher wanted to hear. In particular, the poverty portrayed in many of the stories was relentless, as it was in the visual analysis section:

We work and we work for these rich farmer folks and then we come home, if you can call it a home, to raggedy cardboard town where people use newspapers as blankets and old umbrellas as shelter. It is a town made out of garbage. Our money is slowly running out. I have a wife and three children to feed plus myself. Life is harsh.

[ Back to Top]


Implications for Practice

The range of themes noted during this study can be summarized with a few overarching observations:

  • With instruction, the students were able to successfully navigate this online archive and make selections that met the historical criteria of the assignment. However, in their analyses, the students tended to view the artifacts from the vantage point of their personal experiences and contemporary time frame, indicating a need for further classroom modeling and discussion.
  • The students enjoyed creative writing as a technique for learning history. But their efforts needed guidance. Without sufficient background knowledge and directed feedback, they often masked the gaps in their knowledge by inserting unrelated information or inappropriate writing devices such as melodrama. However, devices like humor and irony should be explored as a means of increasing motivation and engagement.

As with any new teaching unit, we learned what we would do differently the next time. Most importantly, the students needed a more structured framework around which to build their stories. As novice historians, their knowledge was too incomplete to be able to write with sufficient credibility. Merely requiring the students to bring the generations up to the present time was inadequate. The stories needed more points of historical contact and a structure that would propel the narrative toward some sort of closure. In our teacher-expert roles, we could have provided prompts—a list of government programs, historical events and phenomena, such as dust storms—and required the students to incorporate a certain number of these elements in their stories. Students could have selected a second photo from a more contemporary archive and written their stories to connect the generation in the Depression-era photograph to the generation in the modern one. Options like these would still have allowed students to create their own sense of events, but kept them grounded in the historical content—with knowledge they would be applying. Such techniques might have prevented one group from using half of their two allotted pages on descriptions of World War II battles. Presumably all the students would have learned the history better.

Implications for Research

The use of humor in student writing is worth further scrutiny. Humor and irony may be ways for novices to mask a lack of knowledge and experience in a content domain. Or humor may be an opportunity for students to make the task their own and to demonstrate competence in a skill they do have command of. In either case, students are attracted to humor, and educators may find it well worth their efforts to incorporate humor into pedagogical practices. Similarly, students use creative writing to make a personal connection to otherwise distant historical information. It is tempting to consider other methods that would elicit the same sense of ownership and expand student learning, such as having students create contemporary artifacts to be found by future generations. But as we found out, these projects need to be carefully crafted so students have the structure they need to be successful.

This exploratory examination of student creative writing with primary sources is a beginning. More empirical studies are needed to help us understand how young adolescents learn and assimilate historical information and processes through their interaction with collections of primary sources. Some preliminary conclusions can be drawn. If we are to accept the premise that textbooks offer a limiting view of history, then we must be prepared to support students in their reading of other types of sources:

Embedded [primary] texts and sources may contribute to students' immediate reading experience and to their developing understanding and appreciation of history. Yet, without teacher assistance these texts are extraordinarily demanding of some students. (Afflerbach and VanSledright 2001, 704)

But the potential rewards of teacher-assisted student engagement with primary sources are great. Access to multiple representations of events allows students to directly observe that reality is often in the eye of the beholder. During this engagement:

teachers can help students tolerate some uncertainly amidst variant textual accounts, and they can provide them with the tools needed to evaluate accounts—assessing why bias might exist, looking for corroborative detail across accounts (Garner and Gillingham 1998, 229).

At the same time, it is important not to trivialize the difficulty of crafting these open but guided learning experiences. Bass and Rosenzweig (1999) note that “The construction of effective inquiry activities demands knowledge of the topic, the documents, and the archive, as well as the craft of introducing students to the inquiry process” (49). Teachers need support to accomplish these many goals. Britt et al. (2000) have devised a “coached-apprenticeship” method teachers can use to help students read and synthesize multiple documents. In terms of the big picture, there is a growing community of educators now interested in this approach who are willing to engage in discussion and offer support (see, for example, The Inquiry Page (Bruce 1998) at and the Library of Congress (n.d.) American Memory Learning Page at

Rouet et al. (1998) argue that it is more important for students to understand why and how participants, historians, and political analysts argue about events than to come up with a single explanation of why and how those events occurred. Good historical pedagogy requires the presentation of multiple perspectives, including those that conflict with one another. Teachers can foster this deep understanding by exposing students to the mysteries of an archive and supporting their explorations:

This is about the year 1940, which is right around the end of the Depression. This picture shows that things did not go back to normal immediately and for some never again were the same.

The students in our class who made this observation about their photograph discovered powerful key concepts, having learned to identify multiple perspectives and historical realities themselves.

[ Back to Top]


Afflerbach, P., and B. VanSledright. 2001. Hath! Doth! What? Middle graders reading innovative history text. Journal of Adolescent and Adult Literacy 44, no. 8: 696–707.

Bass, R., and R. Rosenzweig. 1999. Rewiring the history and social studies classroom: Needs, frameworks, dangers, and proposals. Journal of Education 181, no. 3: 41–63.

Britt, M. A., et al. 2000. The sourcer’s apprentice: A tool for document-supported history instruction. In Knowing, Teaching, and Learning History: National and International Perspectives, ed. P. N. Stearns, P. Seixas, and S. Wineburg. New York: New York Univ. Pr.

Britt, M. A., et al. 1994. Learning from history texts: From causal analysis to argument models. In Teaching and learning in history, ed. G. Leinhardt, I. L. Beck, and C. Stainton. Hillsdale, N.J.: Lawrence Erlbaum.

Bytwerk, R. N.d. German Propaganda Archive. Accessed July 26, 2002,

Epstein, T. L. 1994. Sometimes a shining moment: High school students’ representations of history through the arts. Social Education 58, no. 3: 136–41.

Fasulo, A., H. Girardet, and C. Pontecorvo. 1998. Seeing the past: Learning history through group discussion of iconographic sources. In International review of history education, vol. 2: Learning and reasoning in history, ed. J. F. Voss and M. Carretero. London: Woburn.

Garner, R., and M. G. Gillingham. 1998. The Internet in the classroom: Is it the end of transmission-oriented pedagogy? In Handbook of literacy and technology: Transformations in a post-typographic world, ed. D. Reinking et al. Mahwah, N.J.: Lawrence Erlbaum.

Glaser, B., and A. Strauss. 1967. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

Greene, S. 1994. Students as authors in the study of history. In Teaching and learning in history, ed. G. Leinhardt, I., L. Beck, and C. Stainton. Hillsdale, N.J.: Lawrence Erlbaum.

Henry, C. 1995. Migrant Mother by Dorothea Lange. Art Education 48, no. 3: 25–28, 37–40.

Holt, D. G., and C. Willard-Holt. 1995. An exploration of the relationship between humor and giftedness in students. Humor 8, no. 3: 257–71.

The inquiry page. 1998. University of Illinois at Urbana-Champaign Graduate School of Library and Information Science. Accessed July 26, 2002,

Kintsch, W. 1986. Learning from text. Cognition and Instruction 3, no. 2: 87–108.

Levstik, L. S., and K. C. Barton. 2001. Doing history, 2d ed. Mahwah, N.J.: Lawrence Erlbaum.

Levstik, L. S., and K. C. Barton. 1996. “They still use some of their past”: Historical salience in elementary children’s chronological thinking. Journal of Curriculum Studies 28, no. 5: 531–76.

Library of Congress. N.d. American Memory: The Learning Page. Accessed July 26, 2002,

Library of Congress. 1998. American Memory. America from the Great Depression to World War II: Black-and-white photographs from the FSA–OWI 1935–1945. Accessed July 26, 2002,

Library of Congress. 2002. American Memory: Historical Collections for the National Digital Library. Accessed July 26, 2002,

McKeown, M. G., and I. L. Beck. 1994. Making sense of accounts of history: Why young students don’t and how they might. In Teaching and learning in history, ed. G. Leinhardt, I. L. Beck, and C. Stainton. Hillsdale, N.J.: Lawrence Erlbaum.

Miles, M. B., and A. M. Huberman. 1994. Qualitative data analysis: An expanded sourcebook, 2d ed. Thousand Oaks, Calif.: Sage.

Rouet, J., et al. 1998. Understanding historical controversies: Students’ evaluation and use of documentary evidence. In International review of history education, vol. 2: Learning and reasoning in history, ed. J. F. Voss and M. Carretero. London: Woburn.

Selman, R. L. 1980. The growth of interpersonal understanding. New York: Academic Pr.

Smithsonian Institution. 2002. Library and Archival Exhibitions on the Web. Accessed July 26, 2002,

Wigfield, A., J. S. Eccles, and P. R. Pintrich. 1996. Development between the ages of 11 and 25. In Handbook of educational psychology, ed. D. C. Berliner and R. C. Calfee. New York: Macmillan.

Wineburg, S. S. 1991. Historical problem solving: A study of the cognitive processes used in the evaluation of documentary and pictorial evidence. Journal of Educational Psychology 83, no. 1: 73–87.

Zola, J., and A. Ioannidou. 2000. Learning and teaching with interactive simulations. Social Education 64, no. 3: 142–44.

[ Back to Top]

Referee Record
Manuscript submitted: August 2002
Board approved: December 2002