The Deconstructor: Providing the Scaffolds for Students to Excerpt

The Deconstructor 1

The Deconstructor: Providing the Scaffolds for Students to Excerpt,

Describe, Analyze, Interpret and Synthesize to Form New Understandings

Kristen A. Sosulski

Columbia Center for New Media Teaching and Learning, Columbia University

Lawrence Engel

Columbia University

Gordon Campbell & Lee Davis

Columbia Center for New Media Teaching and Learning, Columbia University

The Deconstructor 2

Abstract

The Deconstructor, a technology developed at The Columbia Center for New

Media Teaching and Learning in partnership with adjunct professor Lawrence Engel,

provides students with a digital workspace to review, excerpt, describe, analyze and

synthesize texts. The Deconstructor is described in the context of work done in an

undergraduate film course at Columbia University. The technology supports a

constructivist approach to learning about film, giving students the means to form their

own interpretations and new understandings through in-depth study. Results indicate that

technology as a component supports students in forming their own interpretations of film.

Future uses of The Deconstructor are in the areas of music, anthropology, media literacy,

and journalism.

The Deconstructor 3

The Deconstructor: Providing the Scaffolds for Students to Excerpt Describe, Analyze,

Interpret and Synthesize to Form New Understandings

The purpose of this paper is to familiarize readers with the Deconstructor: An

Online Film Analysis Tool, an educational technology intervention that offers a

constructivist approach to learning about film. Through active engagement with film

scenes, students examine and deconstruct the content from multiple vantage points as a

method to develop new and multiple understandings. The Deconstructor is described in

the context of a an introductory course at Columbia University’s School of the Arts’ film

department from fall 2002 through spring 2004.

Theoretical Framework

Deconstruction as a Learning Activity

In the context of this study, deconstructing is a technique to critically analyze

compositions such as films, texts, music, or images. The activity of deconstruction entails

investigating the constructs of a composition as a method to form new insights into the

original work. This involves identifying the elements that construct the whole text and

discovering the relation of the sub-parts to the complete work. This type of activity

assists students in building analytical frameworks through deconstructing artifacts and

describing the grammatical elements used by the author to construct the text.

Deconstruction is a type of authentic activity that can provide learners with

primary source materials. When students are given the same materials used by experts,

they can gain experience, exposure and context by engaging real-life activities in contrast

to artificial and de-contextualized activities. Authentic activities are designed to assist

The Deconstructor 4

learners to confront the gaps in their knowledge rather than having the knowledge

interpreted for them (Lajoie, 2000).

Difficulties for Novices

However, it can be especially difficult for novices to engage in authentic

deconstruction activities without the language of the domain and the framework for

analysis. An understanding of the language of the discipline must precede critically

examining the elements of a composition. Without this grammatical understanding,

novices cannot accurately identify and describe specific details of the text nor construct a

framework for analysis.

Should the activity of deconstructing be reserved only for advanced learners and

experts? Authentic activities are important for learners in providing experiences and to

place their learning in context (Brown, Collins, & Duguid, 1989). As educators, we argue

that this type of constructivist investigation is invaluable for novices if they are provided

with the appropriate supports and resources to engage in this type of study. Students

should be provided with opportunities to construct their own knowledge “rather than

having the teacher interpret the world” (Lajoie, 2000. p. 188) for them.

Scaffolding Novices

Scaffolding provides students with supports in a learning environment and frames

the interactions with the content (Young, Nastasi, & Braunhardt, 1996). With the

appropriate scaffolding, novices can begin to construct their own knowledge of an

unfamiliar domain. Types of scaffolding that can assist in the activity of deconstruction

are timely explanations, observations, examples, guiding questions, and visualizations.

The Deconstructor 5

Various scaffolding techniques can enable novices to interact with authentic resources in

an expert-like activity.

Prior to engaging in the sophisticated analysis of a text, it is necessary that

students understand the language and vocabulary used in the discipline. Furthermore,

students need to acquire and apply their knowledge of these terms while actively

engaging in the deconstruction process. The terms should be coupled within the context

in which they are applied (Cognition and Technology Group at Vanderbilt, 1997).

In addition to situating the vocabulary in the context of use, the learning of

analytic skills can be scaffolded by the teacher modeling the use and application of the

language when describing the content under examination. Observing an expert

deconstruct a composition can help students understand the framework from which the

expert analyzes the text (Lajoie, 2000). The expert can scaffold learners by providing

multiple examples and recommendations when first learning and less as the learners

acquire their own analytic skills (Bransford, Brown, & Cocking, 2000; The Cognition

and Technology Group at Vanderbilt, 1997; Vygotsky, 1978, 1986). Examples establish

a shared understanding and communication amongst students (Cognition and Technology

Group at Vanderbilt, 1997).

Domains such as literature, film or history are composed of multiple and complex

structures. These structures are best understood when examined from various vantage

points (Spiro, Feltovich, Jacobson, & Coulson, 1991). For example, the same text can be

analyzed from many different viewpoints such as, historical, cultural, political, or

technical. In these ill-structured domains, multiple readings of a text from different

perspectives can reduce oversimplifications and misunderstandings (Spiro, et al., 1991).

The Deconstructor 6

Providing multiple examples may offer a more cohesive understanding of the domain,

where one example for a complex concept may limit the understanding and even be

misleading. Multiple examples may prevent students from generalizing from just one or

two instances (Spiro, et al., 1991).

Furthermore, a set of questions can guide the analysis and shape the way novices

focus their attention. Texts can be viewed from many different perspectives, making it

difficult for novices to know which elements to focus their attention. The questions slow

the process down to a pace more suitable for beginners allowing them to simultaneously

learn and apply the framework.

Lastly, visualizations of analyses can provide opportunities to review the global

and local properties of an artifact (Snir, 1995; Lajoie, 2000). Specifically, visualizations

are helpful in illustrating the interconnection among the elements that comprise the text.

The Need for Scaffolding in a Deconstruction Activity

The need for providing the scaffolding in the authentic deconstruction activity

arose from an instructor’s desire for his undergraduate students to critically examine the

structural elements of film in an introductory film studies course, Introduction to the

Study and Theory of Film. The course introduces undergraduate students to the history

and theory of cinema and the various ways to analyze and understand the underlying

structure of film. Students study the aesthetic components of film, the language used to

identify and describe these components, and the ways they shape narrative and dramatic

elements of film. In this course, the instructor follows a methodological approach to

analyzing narrative films. This approach deconstructs the structural design of film

(Sharff, 1982) as a method to determine the impact of that design on the viewer, which

The Deconstructor 7

"involves presenting chiefly linear information (the story) through a battery of shots"

(Sharff 1982, p. 6). As evidence, the instructor deconstructs a film scene, shot1 by shot,

identifying the grammatical strategies employed by the use of the camera and subsequent

editing in presenting a cogent narrative to the viewer.

A class goal is to help students see film differently from they way they normally

see movies at the theater. Rather than watch, students are asked to analyze. By reviewing

the chain of shots students have the opportunity to study the anatomy of film separately

from the narrative or story context to better understand how a reader becomes engaged in

the text. Students look for the patterns, progressions, and the ways the cycle of familiarity

and change lead certain films to carry greater “weight” to the narrative, to develop a more

influential text that connects with the reader.

In class, students were given a spreadsheet template that provided a framework to

analyze each shot of in a film scene (see Figure 1). Students would rely on VCR counters

or stop-watches to time each shot and either trace the TV screen or draw the key frame2

for each shot. This was an inordinately cumbersome process that took many hours of

work. Students often chose not to pursue this kind of detailed analysis. Even after

collecting data, students had to plot and graph the data in order to decipher patterns.

The complexity of this approach and the lack of expertise prevented the

introductory students from easily engaging in their own deconstructions of film scenes as

1 A shot is a single graphic arrangement, connected to another shot through cut, camera movement, actionwithin the frame, or some form of special effect..2 A key frame is a special-purpose marker that denotes a shot.

The Deconstructor 8

a method to explore the underlying structural elements that provide a narrative discourse

for the audience.

Figure 1. Spreadsheet scene analysis template distributed to students.

To address the need for students to easily construct their own interpretations of

films, The Deconstructor was created to enable film students to effectively manipulate,

describe and analyze films. The Deconstructor was developed by The Columbia Center

for New Media Teaching and Learning (CCNMTL) in partnership with Professor

Lawrence Engel’s theory of film. CCNMTL is dedicated to working with Columbia

University faculty to “Advance the purposeful use of new media and digital technologies

in education” (Columbia Center for New Media Teaching and Learning, 1999).

CCNMTL is comprised of an administrative staff, web developers, programmers, and

educational technologists. Educational technologists typically work with faculty in

implementing project ideas, such as the Deconstructor. CCNMTL approaches the study

of purposeful uses of technology in higher education through a design research

The Deconstructor 9

framework (Edelson, 2002; Bereiter, 2002; Collin, Joseph & Bielaczyc, in press).

CCNMTL and Professor Engel are the principle investigators in this study.

In this paper, we will be presenting the findings of this educational intervention

implemented in an introductory film course at Columbia University in 2002.

Additionally, we will share preliminary findings from more recent implementations. The

focus of this paper is based primarily on the fall 2002 implementation within the course:

Introduction to the Study and Theory of Film. Other course implementations include

Documentary Tradition and Film Production.

Methods of Inquiry and Data Sources

Educational Intervention

In fall 2002, seventy-four students in the Introduction to the Study and Theory of

Film course used the Deconstructor. The structure of the course requires students to

attend a four-hour lecture per week in addition to a one-hour discussion section. The

lecture is taught in a seventy-five seat screening room where students watch and discuss

films. During the film viewings, the instructor provides commentary to focus the

attention of students on particular details. Students are required to keep journals where

they record the details about the films they view. Attendance is crucial in the course,

since each film is shown once. While students have opportunities for discussion in the

lecture, the main discussion takes place in smaller groups within the discussion sections.

The seventy-four-person class was divided into three discussion sections; each led by a

different teaching assistant. Each teaching assistant designs his or her curriculum for the

discussion section with direction from the instructor. Students are evaluated based on

The Deconstructor 10

their journal entries, attendance in lecture and the discussion section, a midterm exam and

final project.

The class intervention took place during lecture and in discussion sections where

the students could work with the Deconstructor and begin to apply the professor's

methodological approach to looking at film. Students were required to use the

Deconstructor for the midterm exam, in the discussion sections and as an option for the

final project.

Functions, Features, and Scaffolds

The functional aspects of The Deconstructor allow students to perform a four-step

process for scene analysis: careful review of the scene; determination of the cinematic

variables in use; dissection of the scene into its series of shots during which each shot’s

duration and variables under analysis are documented); visualization of the data and

using the data and its visualizations to articulate the results that concentrates on the

relationships among the shots and their connection to the plot.

Students begin their work in The Deconstructor by going to the World Wide Web

and entering the address for the Deconstructor. Since the Deconstructor is a tool that

allows students to employ a method of analysis previously not encountered, an

introductory page with a letter from Professor Engel welcomes the learner, introduces the

tool and its purposes. The introduction provides a brief preview of the work to come, and

provides support from the instructor. From this page, students can access information on

how to use the site, which includes a glossary of terms, an opportunity to send feedback

or comments to the researchers, and the login page to access the main functionalities of


the Deconstructor. These options continue to be available to students throughout their

work in the Deconstructor.

After reading the introduction to the Deconstructor, students can login with their

unique user name and password to access the core components of the technology. The

Deconstructor allows three levels of access: student, teaching assistant, and instructor,

each with different access privileges. Individual account access provides students,

teaching assistants and the instructors with a personal space to manage their analysis of

film scenes. With each user having a personal space, their analyses can be stored and

available at anytime via the Internet. The personal space is referred as My Databoards

and gives students a personal space to view, edit or delete their scene analyses. In

addition to being a place to store the scene analysis of students, the My Databoards page

provides an example of a deconstructed scene. Students can study the choices made in the

analysis and modify them.

The teaching assistants and instructor have access to all student work and the

ability to view and comment on student work. As a way to provide apprenticeship to

students, teaching assistants and the instructor can provide feedback to students by

annotating their scene analyses within the tool.

After students login to the Deconstructor, they are automatically taken to the

View/Select Scene screen. Before students can begin their analysis, they must first

choose a scene to analyze. There are over seventy scenes available through the

Deconstructor film library for students to investigate. Students may request other scenes

to be made available to them. However, instructors are given permissions to add more

content to the film library. This feature is not available to students due to the labor of the


support staff to compress3 the video clips, insert the time code track to display the hours,

minutes, seconds, and frames and label the clips accordingly.

Figure 2. View/Select scenes page.

It is important for students to have opportunities to deconstruct films from

different genres and time periods to create a powerful frame of reference or schema on

the ways films are constructed (Spiro, et al., 1991). The scenes available to students range

from classic Hollywood to documentaries. All the video clips are used under the fair use

clause described in the University’s copyright guidelines. Each scene may be previewed

3 For the film scenes to be easily view over the Internet, they are compressed to a frame rate of 15 framesper second (fps) and encoded in QuickTime format at 240 x 180 pixels. The average frame rate for video(VHS or DVD) is 30 fps.

QuickTime Icon


from this View/Select scene page by clicking its corresponding QuickTime icon (see

Figure 2).

Figure 3. Databoard View. An analysis of shot using various descriptors such as shottype, shot angle, and shot perspective.

After reviewing a film scene on the View/Select page, students create a

Databoard to organize their analyses. The Databoard is an organizing unit that

distinguishes one shot from another. The Deconstructor allows students to deconstruct

many scenes, each scene one at a time and one shot or sub-shot4 within that scene. The

4 The functionality of the Deconstructor allows for sub-shot data which will expand the number ofdataboards to exceed the strict number of shots in the scene. This occurs when a single shot is a complexone. The complexity lies when either through subject movement within the frame or through cameramovement or both, the image changes dramatically in type or kind over the length of the shot. It is as if


Databoard window allows the student to enter data on a single shot or sub-shot (see

Figure 3). There will therefore be the same number of databoards within a scene analysis

as there are shots within the scene.

Figure 4. Storyboard View. An example of a film scene from Psycho (Hitchcock, 1960)dissected in The Deconstructor.

Students begin to dismember the selected scene into series of shots (see Figure 4)

after creating a Databoard to store their data. Students extract each shot from the scene by

several particular graphic arrangements are combined not through cuts from shot to shot but are connectedthrough camera movement.


indicating the start and end time for each shot. The shot can be previewed before the

student commits to the start and end time. Figure 4 illustrates a scene dissected into nine

distinct shots that comprise the forty-three second scene. This collection of still images

represents the visual flow of a film scene and is referred to in the Deconstructor as the

Storyboard View and is accessible by students upon demand during their analyses.

After dissecting the scene into a series of shots, the Databoard acts as a scaffold

that frames and focuses the analysis of each shot (see Figure 3). The Databoard focuses

the student analysis of each shot on two basic sets of elements; one is the static or

photographic set, the other the kinetic or movement set. The Deconstructor does not

contain variables that address sound or tone in the scenes. To compensate for the

numerous ways a shot can be analyzed, students can identify their own elements for

analysis using the Customized Analysis options. Using the Customized Analysis

functionality, students can create two new elements to use in their analysis of a shot.

When students are uncertain of the ways in which to apply a term, a definition is

provided and an example of the term applied in a visual example. This information can be

accessed on the databoard by selecting the red question mark, indicating that the student

has a question or wants information. Additionally, a glossary of terms is provided within

the Deconstructor help file. Embedding the definitions and examples in the context of use

gives the learner the option to review terms when it is most useful to them (Spiro,

Feltovich, Jacobson & Coulson, 1991).

When students identify the shot’s characteristics, each element5 corresponds to an

integer value that is assigned to the shot along with a textual description. For example, if

5 Each shot has the following structural characteristics that can be described in the Deconstructor: graphicarrangement, shot type, shot angle, shot perspective, shot camera movement, shot camera movement type,


shot type is a medium shot, the integer value assigned is 20; a close-up has a value of 25,

usually the higher the integer the stronger the element6. This is done in order to aggregate

the analysis in a visualization as illustrated in Figure 5.

Figure 5. Visualization. The representation of the angle of the shots juxtaposed with theduration of each shot.

A Visualization tool was created as part of The Deconstructor as a way to aid

students in interpreting their shot-by-shot analysis of a film scene. The visualization tool

assists students in aggregating their entire scene analysis and are based upon the analysis

entered by students in the databoards. More specifically, the tool graphs the number of

shots, duration of each shot and a third variable selected by the student, such as shot

angle or camera movement. Figure 5 depicts a scene with nine shots where the shot angle

alternates between eye-level and slightly low as illustrated with the green data points (or

camera movement value, shot subject movement, subject movement direction, shot subject movementzoom, shot subject movement zoom strength, shot entrance, shot entrance frame, shot exit, shot exit frame,and the custom fields defined by the student.6 At least compared to that element from shot to shot. For instance, the close up has a higher value than along shot so that upon visualization or graphing, the close-up will appear higher on the Y-axis than the longshot. Similarly a movement from left to right is given a positive value while right to left is given a negativevalue.

Green

Blue

Legend


lighter color). The blue data points (or darker color) indicate the length of each shot on

X-axis.

The resulting graphs provide a basis for interpreting underlying cinematic

structure of the scene, independent of watching the scene and becoming absorbed in the

story. A legend is provided to help students interpret the visualization of their scene

analysis. The graphic visualization allows students to see patterns or changes in the

rhythm and flow of the chain of shots. It is a crude score of the underlying cinematic

structure. The arrangement of shots into coherent beats distinguishes masterpieces from

pedestrian works. When a particular change occurs from one shot to another, students are

asked to look back to the film for a correlation between this cinematic moment and its

narrative state.

Implementation

However, before introducing the Deconstructor to students, it was important to

create teaching partners and train the teaching partners in using the technology. The

teaching assistants and the instructor were the obvious partners and instrumental in the

learning progress of the students. It was crucial to train the teaching assistants to the

Deconstructor to ensure they could correctly model the process the professor envisioned

for the students. It was prudent to ensure that the teaching partners could work

confidently with the technology and perceived it as a necessary component of the class

curriculum. As a strategy, the instructor provided a demonstration of the tool to the

teaching assistants. It was hypothesized that this would further situate the Deconstructor

as a real tool for use in the class environment. After the demonstration, the teaching

assistants were trained in using the technology by CCNMTL staff.


In the first class lecture, the instructor presents a short scene and asks the students

to guess how many shots comprised the scene. Regardless of the scene chosen, students

historically underestimate7 the number of shots in the film sequence by one-third to half.

Upon review, students attend to the shots of the scene, not the plot. The students

immediately begin to see film differently. They are directed to pay attention to the type of

shots used in the scene (how many close-ups, medium shots and long shots, for instance).

Finally, students are asked to identify movement8 within the shots and amongst them.

In the second lecture, the instructor demonstrates the ways in which students

should use the Deconstructor. The instructor connects his computer to the web and

displays the Deconstructor through a video projector onto the main screen. Usually the

instructor uses a short, nine-shot sequence from Psycho (Hitchcock, 1960). The

methodology suggests that the students watch the scene several times to familiarize

themselves with the scene. Then they are asked to determine which variables the text

seems to be relying on to structure its shots. These are then the variables that students

will use to focus their analysis. In all cases, the duration of each shot must be recorded

and is used as a key base-line for visualization and analysis. After collecting data, graphs

are created to help visualize the “score” of the scene.

During the third week of the course, students attend a special discussion section

led by teaching assistants and personnel from CCNMTL held at a media lab. Working in

pairs, students are trained in using the Deconstructor. Teaching assistants follow-up with

7 Anecdotal record-keeping by the professor for over 15 years.8 In terms subject movement, or what is commonly called mis-en-scene, and cameramovement.


students to determine their success with a question and answer session. The instructor

also elicits comments and questions in the following lectures.

During the subsequent weeks students in some discussion sections, students

present on their work with the Deconstructor and their analysis. Between the eighth and

tenth week of the course students are required to work with the deconstruction for the

midterm exam. Lastly, students have the option to do a comparative scene analysis in the

Deconstructor for their final project.

Research questions

The goal of integrating The Deconstructor into the introductory film course is to

improve students' analytic skills and acquisition of vocabulary necessary for the course.

The instructor and the research group is interested in answering the following questions:

• How does The Deconstructor provide students with opportunities to engage in

their own analysis?

• Does the Deconstructor enable students to apply the language of film?

• Are students able to form interpretations from their analyses in The

Deconstructor?

• How do students understand film analysis prior to their use of the Deconstructor?

• How confident are students in applying their knowledge of film analysis?

Data Collection

In pursuit of answering the research questions, data was collected from a multiple

sources. The data sources include class observations, periodic meetings with the


instructor and teaching assistants, a student focus group, instructor interview and analysis

of student work aided in the evaluation effort.

Prior to the beginning to the fall 2002 semester, three meetings were held with the

instructor and several teaching assistants to identify implementation strategies and to

introduce them to the Deconstructor. Informal meetings were held throughout the

semester with class teaching assistants to assess their perspectives on the project in the

classroom and of students' progress. Two additional meetings were held with the

instructor to assess the validity of The Deconstructor in the classroom.

Throughout the semester, the class lectures and three discussion sections were

observed to gather insight into the ways The Deconstructor supported the curricular

goals. Working closely with the teaching assistants, each discussion section was observed

regularly with a particular focus into the insights that students gained from using the

Deconstructor.

Next, a focus group was conducted using a quasi-experimental design for one

hour on November 20th, 2002 through November 22, 2002. Twenty-two out of the thirty

randomly selected students participated in a focus group experiment. The students were

asked to respond to questions regarding graphical representations of an undisclosed

deconstructed film scene produced in The Deconstructor. The goal was to understand the

ways students were interpreting the graphical visualizations produced by the

Deconstructor and how explicit the patterns were to students. A discussion about their

reaction to the exercise and a survey assessing motivation and usage was administered. A

CCNMTL staff member who was not a principle investigator in this research experiment

conducted the focus group. This was done to help ensure external validity.


Then, student presentations, take-home exams and participation in class were

analyzed particularly focusing on how students used the language of the discipline and

how well students made use of the Deconstructor. The professor was interviewed by

CCNTML to gauge an understanding of the quality of student analysis in The

Deconstructor.

Lastly, a survey instrument was administered to students in the Documentary

Tradition fall 2003 course and the spring 2004 Introduction to the Study and Theory of

Film course after a Deconstructor training session. The purpose of the survey was to

gaining insight into ways students understood film analysis and their confidence applying

it prior to using the Deconstructor.

Results

Prior to the educational intervention, it was difficult for students in the

Introduction to the Study and Theory of Film course to analyze film in the way presented

by the instructor. This left little opportunity for students to analyze film and deduce

meaning from their analyzes.

Our results indicate that the Deconstructor successfully enabled students to

critically examine films in ways impossible prior to the educational intervention.

Individuals without prior knowledge in film analysis actively engaged in the activity of

deconstruction as a method of inquiry. Students began to isolate, describe and analyze the

elements of a film scene and construct understandings about film.

The data collection was focused the ways in which the Deconstructor offered

opportunities for student to engage in a deconstruction activity, the acquisition and


application of the language of film, and ability to interpret and synthesize the relations

between the elements used by filmmakers to construct films.

Opportunities for Students Engagement

The Deconstructor offers students and faculty a means to identify, analyze, and

deconstruct film in modes never possible. Students and faculty were able to precisely

deconstruct the film scenes and isolate portions that warranted close analysis. Evidence

from examining student work done in the Deconstructor reveals the average number of

scene analyses in the Introduction to the Study and Theory of Film course was four-and-

a-half. Half of those students used the Deconstructor as the primary source material for

their final projects. Students were only required to analyze two scenes, one as a training

activity and another for the midterm exam. Out of the three discussion sections, one

teaching assistant required students to present an analysis once during the semester. This

type of integration of the Deconstructor in the students’ activities enabled students to talk

about film analysis, helping students reflect on their learning, while actually engaging in

it. This was observed in the discussion sections and the lectures.

One key attribute of the on-line tool is the enormous time savings dividend. There

is enough automation and program support for collecting, distributing and visualizing

allowing the student a more efficient and in-depth engagement in the close-study of film

scenes. On average, students from the focus group estimated that they spent over nine-

and-a-half hours using the Deconstructor. The minimum time spent was two-and-a-half

hours and the maximum was as high as thirty hours. A scene analysis done using other

methods could take twice to three times as long, if students had to locate the scene and

plot and graph their analysis without automation or scaffolds to guide the analysis.


Application of the language of film

It appears that students learn how to re-think the way they see film much more

quickly through early exposure to the Deconstructor. In-class discussion indicates that

students are far more aware of the arrangement of shots and the relations of this

arrangement to the core narrative discourse of the film. Additionally, student journal

entries support the argument; more entries relate to phrasing and categories of cinematic

structures than to story or character issues (although these are clearly important elements

of film as well).

Students used the language of film in their interpretations; however, the language

was not acquired through the use of The Deconstructor, but through the readings, lectures

and discussions. Students had problems using The Deconstructor in the beginning

because they know the film vocabulary well enough. The on-demand terms and glossary

did not provide students with adequate information on how to apply the vocabulary. This

finding could be influenced by the fact that students rarely referenced these resources.

Thus, it was concluded that students acquired and applied the language of film in

primarily through the readings and lecture not directly through engagement with the

Deconstructor.

Forming Interpretations of Film

The focus group experiment verified that students are able to interpret the

graphical representations of scenes generated in the Deconstructor. A student that

participated in the focus group experiment expressed that she “… could totally interpret

what the graph [graphical representation of the film scene] was saying. But I’m not sure

if I could offer a full interpretation” (personal communication, November 20, 2002).


While students were able to make meaning from examining the visualizations provided in

the focus group, many noted the importance of actually deconstructing a scene to

understand the principles that can be learned. The ways students interpret the graphs

without actually doing the scene analysis was surprising. Many students did not simply

interpret the numeric information presented on the graphs. Instead students correlated the

patterns presented in the graph with a story-line. The visualizations provided in the

Deconstructor are intended to isolate the dramatic and narrative content from the

structural elements of film. However, students were using a narrative as a starting point

to form their interpretations. Reflecting on the results of the focus group the instructor

discussed,

…if they [students] can get a gestalt of, or sense of the scene, and immediately

relate it to story, that’s really good. And, maybe more sophisticated than what I’ve

been doing, which is, looking at the different element and try to come up with

what the underlying cinematic structure is. [Students may not use my approach

exactly] because they are not used to it, which makes sense because they are used

to stories. So if they see this [the graphs], and connect it to beats and stuff, which

I have harped on in class, so much the better…If they can create the story, that

would be very cool. That was what I was trying to do (L. Engel, personal

communication, December 4, 2002).

The insights offered by the instructor revealed the difficultly students approach film

analysis from purely a structural viewpoint. Instead, students use a narrative or story to

scaffold their learning of this type of analysis. Students throughout their school career

read texts, watch television and films, all guided by a narrative. On the other hand, the

majority of the students in the introductory course had limited or no exposure to


analyzing film from a structural viewpoint. This insight into student thinking and

construction of meaning offers potential for future study.

Usage

Use of the Deconstructor in the Introduction to the Study of Film course truly

generated a new teaching opportunity. The instructor is able to model the various ways to

look at film and require students to engage in their own analysis to form their own

interpretations of films. Observations demonstrate that the Deconstructor permits the

instructor and teaching assistants to engage in deeper conversations regarding the value

and challenges of film analysis, since students can easily analyze many scenes outside of

class. The results of the study suggest that these possibilities are a consequence of the

way in which the digital working environment, The Deconstructor, is integrated into

course curriculum.

The instructor and teaching assistants evaluated student work in the

Deconstructor, however, assessment took the form of written comments on the midterm

and verbal feedback in lectures and discussion sections. The instructor and teaching

assistance made little use of the feedback feature in the Deconstructor. This was probably

because the importance of student work relied more on student interpretations and

synthesis than the actual analysis from which the interpretations were derived.

As anticipated, the students utilized the graphical visualizations as primary source

documents in take-home exams. Students saved the image of the graph into word

processing document as a way to show evidence to support for their claims and the

synthesis of their analysis.


Students used the visualizations generated in the Deconstructor as presentation

material and discussion points in discussion sections. Lastly, about half the students in

the focus group reported that they were using the Deconstructor to generate material for

their final projects.

Unanticipated usage

An unanticipated use of the Deconstructor was by a student in a Film Production

class at Barnard College. This student used the Deconstructor as a construction tool. The

student used a blank clip (film scene) to create a detailed shot list. This demonstrates

potential for the Deconstructor to assist in storyboarding or creating a shot list for

filmmakers. It could allow for distinct pre-visualization of a script or concept.

Students Perceptions of the Deconstructor

In the fall 2002 Introduction to the study and theory of film, students saw the

Deconstructor as a useful tool in looking general trends within a scene. While engaging

with the Deconstructor students identify the value of the tool in helping them

conceptualize the film analysis methodology studied in the course. One student

commented:

It was helpful in that now I have a much better understanding of shot type, length,

and all of the other things we look at with the Deconstructor, but it was more

helpful in helping me understand the importance that each of these elements had

on film (Student 13, November 20, 2002).

This comment illustrates on the value the student places on language as it is applied in a

real context. More importantly, this student presents an understanding of the impact and

results of the careful arrangement of cinematic structures. To emphasis a common theme


shared by students, another student described the impact of the technology as a way to

help “… understand the complexities and planning a director must look at when

designing a shot” (Student 53, November 21, 2002). The sediments of this student

illustrate an understanding of the director’s deliberateness in constructing a single shot.

A general comment from one student described the Deconstructor as an insight

into the complexity of an ill-structured domain such as film. “The Deconstructor truly

revealed a new level of cinema that I didn’t believe existed” (Student 49, November 20,

2002). This student revealed that the Deconstructor presents learners with a new way of

looking at film. This sediment was consistent with the course goals.

In the fall 2003 Documentary Tradition course and the spring 2004 Introduction

to the Study and Theory of Film course, students saw the Deconstructor as a helpful

mechanism for film analysis, language, filmmaker’s impact on the viewer, and

organization of scene of analysis. The surveys revealed that students had a fair to good

understanding of the language of film used in structural analysis before the engaged with

the Deconstructor. Students were only neutral to confident in applying their knowledge in

the tool. A follow-up survey will be administered to students in the spring 2004 to gauge

their confidence after having some experiences with the technology. These preliminary

results reveal that students need more ways to support their learning of film analysis prior

and during the use of the Deconstructor.

Educational Importance of this Study

The first implementation of the Deconstructor in the Introduction to the Study and

Theory Film course provides evidence that students are able to successfully engage in the


active dissection, identification, and analysis of film, and form their own interpretations

and perspectives about film.

Extending Use to Other Disciplines

The Deconstructor in this study was only used in the film context, but there are

also opportunities to extend the use to other disciplines in higher education. For example,

to look at the pieces that comprise a piece of music could offer students new perspectives

on the art. Other areas that where the Deconstructor could be repurposed is in philosophy,

art history, writing courses that focus on the image, sound, text or video.

The Deconstructor can serve as a model for instructors who want their students to

look critically at compositions as a way to make interpretations and form new

understandings. This model could be especially helpful to students who are new

analyzing compositions. Text that warrant close-study may benefit from the basic

framework of the Deconstructor. If the text owns the temporal element as is the case with

film, video, and television, along with dance and music, the duration analysis alone may

be of significant help in better understanding the text. If there is movement, again

suggested with the domains listed above, the motion categories may also be of value,

although in the case of dance and music the specific categories may need to be redefined.

If there is a particular compositional or static element (perhaps with dance, or outside

these aforementioned domains, photography, the visual arts, and sculpture all have both

compositional and movement elements to them), then with some redefinition of terms

and values, the Deconstructor will allow for systematic analysis. Any discipline that has

its own language or syntax can be used like the Deconstructor to quickly bridge gaps

from novice to expert.


For instance, if a student wishes to analyze word usage in the nightly news

reporting of a particular subject, fields could be created for those words or key phrases.

For texts using words only, either prose or poetry, the Deconstructor would be a possible

tool for syntactic analysis. Word usage, imagery, and thematic elements along with

grammatical structures could all be part of the Deconstructor’s modified category set.

Improvements

For future implementations of the Deconstructor, there are a variety of issues that

require close examination. This study only reflected a portion of the questions we have.

There are plenty of opportunities for further study into the ways in which students

actually learn from the technology. The study was shortsighted in the ways that it did not

distinguish much from the lectures or discussion sections. However, evidence pointed to

discussion sections as the main source of student interaction. Futures studies will look

closer at the ways in which students are truly acquiring the language and framework for

analysis with an emphasis on developing strategies to assist in their learning.

A study of particular interest to the authors is the effects of students working in

pairs versus those working alone in a deconstruction of a scene. Preliminary evidence

suggests that collaboration could be the optimal scaffold for learning the vocabulary of

film and developing a strong framework for analysis while students work in the

Deconstructor.


References

Bereiter, C. (2002). Design Research for sustained innovation. In Cognitive Studies,

Bulletin of the Japanese Cognitive Society.

Bransford, John, D, Ann L Brown, and Rodney R. Coking (editors). How People Learn:

Brain, Mind, Experience, and School (2000) National Academy Press,

Washington D.C.

Brown, J.S., Collins, S., & Duguid, P. (1989). “Situated Cognition and the Culture of

Learning,” Educational Researcher, 1989, 18, pp. 32-42.

Collins, A., Joseph, D., & Bielaczyc, K. Design research: Theoretical and methodological

issues. To appear in Campione, J. (Ed.) Volume in honor of Ann Brown. In press.

http://www.extension.harvard.edu/2002-03/programs/cte/ext02drt.pdf

The Columbia Center for New Media Teaching and Learning (1999). “The Columbia

Center for New Media Teaching and Learning Web Site.”

http://ccnmtl.columbia.edu (1 March 2004).

The Columbia Center for New Media Teaching and Learning (2004). The Deconstructor:

An online film analysis tool (Version 1.1) [Computer software]. New York,

NY. Retrieved March 1, 2004, from

http://ccnmtl.columbia.edu/projects/engel/deconstructor

The Cognition and Technology Group at Vanderbilt. (1997) The Jasper Project: Lessons

in Curriculum , Instruction, Assessment, and Professional Development.

Mahwah: Lawrence Earlbaum Associates, Publishers.

Cuddon, J.A. (1991). A dictionary of Literary Terms and Literary Theory, third ed.

London: Blackwell.


Edelson, Daniel. (2002). Design Research: What we learn when we engage in design. In

The Journal of the Learning Sciences. Volume 11 Number 1, Jan. Lawrence

Erlbaum.

Hitchcock, A. (Producer/Director) (1960). Psycho [Motion Picture]. United States:

Universal Studios.

Lajoie, S. P. (2000) Computers as Cognitive Tools: No More Walls. Mahwah: Lawrence

Earlbaum Associates, Publishers.

Sharff, S. (1982). The Elements of Cinema: Toward a Theory of Cinesthetic Impact. New

York: Columbia University Press.

Spiro, R. J., Feltovich, P.J., Jacobson, M.J., & Coulson, R.L. (1991). Cognitive

Flexibility, Constructivism, and Hypertext: Random Access Instruction for

Advanced Knowledge Acquisition in Ill-Structured Domains, “Educational

Technology, May 1991, pp. 24-33.

http://www.ilt.columbia.edu/Publications/papers/Spiro.html

Young, M. Nastasi, B & Braunhardt, L. (1996). Implementing Jasper immersion: A case

of conceptual change. In B. Wilson (Ed.), Constructivist learning environments:

Case studies in instructional design (pp.121-133). New Jersey: Educational

Technology Publications.


Author Note

Kristen A. Sosulski Gordon Campbell, Columbia Center for New Media Teaching

and Learning, Columbia University and the Department of Math, Science and

Technology, Teachers College.

Lawrence Engel, School of the Arts Film Division, Columbia University.

Gordon Campbell, Columbia Center for New Media Teaching and Learning,

Columbia University and the Department of Math, Science and Technology, Teachers

College.

Lee Davis is no longer at the Columbia Center for New Media Teaching and

Learning.

We would like to thank Liliana Pinto for conducting the focus group and her

consultation on the educational intervention, research design and data collection. Also,

we would like to thank Phillip Yaeger, David Cinquegrana, and Ryan Kelsey for taking

the time to review and edit the proposal for this paper and the final paper. Their

suggestions were invaluable to the authors.

Correspondence concerning this article should be addressed to Kristen Sosulski

via email at [email protected].

Documents

The Deconstructor: Providing the Scaffolds for Students to Excerpt