E-Learning Design
A 15-Minute Intro Course to DSLR Camera

Project Timeline
From Oct 5 to Dec 17 at Carnegie Mellon University, Pittsburgh, PA

Team & Advisor
Individual Project | Dr. Ken Koedinger

Key Words
E-Learning Design | Online Education | Course Design | Iterations | Photography


Before you read about my design process, please feel free to explore the mini DSLR course through the link on the screenshot!


I. What and Why

Why did I decide to do an E-Learning unit on DSLR Cameras?
I have been a photography fanatic starting from five years ago when I bought my first iPhone. After gaining more interest in photography, I got a DSLR camera and took a course in DSLR photography to equip myself with more technical skills. After that, I have been practicing DSLR photography for a few years (travel photography and taking portraits for friends and clients).

Photography and documenting through photos has become an accessible everyday practice in many people’s lives. By take photos that record a moment to its full potential, or tell a story that you want to be understood, some skills are a plus. Although cameras in our phones are decent enough, learning how to use a DSLR on manual mode is helpful in circumstances where phone cameras reach their limit (regarding its flexibility and capability).

Learning Goals



Who are the learners I am designing the course for?
The project will be aimed at adult learners who are interested in photography and would like to understand technical variables in a DSLR camera. By taking this course, students would be able to practice concepts of Aperture, Shutter Speed & ISO on (any type) of a DSLR camera.

This course module does not require a DSLR camera (or any camera). What the students need is a computer device with access to the internet (and your motivation!).


II. Cognitive Task Analysis (CTA)


Empirical Cognitive Task Analysis: I conducted two contextual inquiry sessions and three think-aloud sessions with five photography experts. Two out of the five experts are photography instructors with combined over 60 years of teaching experience. Two other experts are graduates students with extensive traveling and portrait photography experience. One last expert acquired her degree in photography .

In the contextual inquiry sessions, I inquired the photography instructors about:

1. Their understanding of what a good photography assessment entails
2. What they think is important in an online learning course of photography
3. How I can scope the lesson to be specific
4. Their goals for students when they teach
5. Challenges of learning photography

My expert interviewees: Sarah Meghan Le & Sardi Klein

Since this task (taking a photo of a child on a swing) is not easily observable with an expert on the scene, I decided to conduct think-alouds while cueing recall of past experience (Contextual Inquiry Principle slides, Slide 4). Cueing questions include: “what if the child is blurry?” “what if it’s noon time and very bright?” “what if the background is not as blurred out as you expected?”

The other three CTAs are empirical CTAs -- think alouds with experts. Aria, a traveler who takes nature photography, says that she would start with putting the DSLR on Shutter-first mode (movement mode). Helen, another travel photographer, says that she would pan the kid on the swing. Kim, a photographer with academic background in photography says that she would choose the lens that gives her a shallow depth of field (opens up the aperture) and it’d be easy to do so.

Cognitive Model Redesign (back to theoretical CTA): Based on the two contextual inquiries, I selected a task that touches upon most knowledge components that I want to introduce in my mini photography course: taking a child on a swing with the background blurred out.

Confirmed Features of Model
Based on feedback from two respondents (experts), I was informed that my model had a lot of notable mistakes. My theoretical model was wrong/incomplete in the following notable points (feedback from two photography experts on what a revised model can be):

Feedback from Sardi Klein:
The model should start with installing batteries, or address that it is a given; also add turn on camera. If camera has a viewfinder, I should set the diopter to you where the eye focuses (should google that) Set MODE--Manual-Program--A or S, and set the ISO and White Balance.
Regarding terminologies, the focus button is the SHUTTER RELEASE button - press down 1/2 way to LOCK in the focus on the most important part of the picture - don’t let go -keep pressed in! Also, I should call “F” “ F-Stop” not “F” or call it “The Aperture.”

When taking the photo, I should set that side of camera first (shutter speed) -then change the other side (F stop) to get the light meter to to the center of it's scale- Zero, Then press down all the way and take photo.

Feedback from Sarah Meghan Lee:
Some of my model isn’t correct as far as the photography vocabulary photographers use. (The “F,” is actually “F-stop or aperture.” Also,  It isn’t a “focus button,” it’s “the shutter button.”)  It also does control the autofocus when the photographer is using autofocus, but that isn’t its primary function. The F-number doesn’t make sense, again the aperture.  

Everything else is alright, and it's hard to explain in a diagram like this really, because there are other factors, and it’s not so absolute as this.  in other words, do they want it more blurry?  It is blurry, but perhaps they want it more blurry.  Also, it depends on if the kid is swinging fast or slow. And, it’s going to be hard for beginners to autofocus on a moving object (the kid swinging).  You may want to pick a subject that is not moving, because a moving object is a rather difficult subject to photograph.  Also, it is better to include photographs in the the e-learning unit, examples, so that there is an example.


CTA models before & after revision:

Main takeaways from CTAs
I have overestimated the details in a single task in DSLR photography. Initially, I was hoping to use the task “taking a photo of a child on a swing” as a guiding task for the learners while I scaffold for them during the course. However, after working on multiple CTAs, I have seen how a photography task can be so “unstructured” and dynamic. Different photographers take different approaches, which will show differently if put in a CTA model. The steps also highly depends on the cameras they use too, which entails different levels of complexities given different types of cameras. However, if taking each of the steps out, we see how different photographers need to follow the principles (though they might take different paths there). For example, if the photo is too bright for them, then they either lower the ISO, or speed up the Shutter Speed, or close the Aperture (by increasing the F-stop). This stage of the research pointed me to just looking at steps in taking a photo -- using small pieces of knowledge components following the Contrasting Cases Principle to guide my design.

Improvements to Assessment Design
1. From the CTAs, I was informed that the cognitive model of my designed task is too complicated for a novice of a DSLR camera. As a result, I would need to break the model down to several chunks to guide my assessment.
2. I deleted components/KCs that are unrelated to DSLR-only topic, questions on students’ motivation and interests for example.
3. I decided to add several more questions on conceptual skills and procedural skills.


IV. Assessment & Instruction Design

Overview of Knowledge Components
The instruction for this assessment is an alternating sequence of direct instruction, checkpoints and practice with hints. resulting in a combination of sense-making, induction, and memory/fluency building. It takes approximately 5 to 10 minutes to walk through the direct instruction pieces; the embedded checkpoints and practice questions may take anywhere from 5 to 10 minutes depending on comfort with the material.

Topics in order

1 What is a DSLR camera?
2 ISO, Shutter Speed, Aperture, Depth of Field
3 Something fun to practice in the moving: panning

As shown in the table, I would like to introduce my instructions by first addressing why using DSLR camera is an appealing skill to acquire despite the current highly-developed technology in cameras on mobile phones. Then, in part two, the three basic main operations of a DSLR camera will be introduced: Aperture, Shutter Speed, ISO, and Depth of Field. After this “heavy” section on a lot of technical terms, last but not least, I would like to borrow a well-made youtube tutorial video to introduce a DSLR-specific shooting technique, “panning,” to ignite learners’ interest in DSLR camera operations.  


V. Design Principles

Principle: Contrasting Cases
Rationale: Based on theories in cognitive science, using contrasting cases helps learners catch the key differences between two photos, and guide them to think about what might have caused the difference. Below are a few examples:

Contrasting Cases Example 1


Contrasting Cases Example 2


Principle: Multimedia
During the CTAs, the experts reflected that they think videos are really helpful for online photography instructions. I am also including a video for the last part of my instruction, “panning a moving object.” I selected one video from youtube based on its video quality, clear instructions, and fun tone


Multimedia Example 1:


Multimedia Example 2: Diagram


Principle: Segmenting
Rationale: Since people people can actively process only a few pieces of information in each channel at one time (Limited Capacity). For taking a photo with a cell phone and a DSLR (on manual mode), the cognitive load is on two different levels. For a novice in using DSLR cameras, thinking about multiple operations (ISO, shutter speed, and aperture) simultaneously is challenging. I plan to segment these concepts in chunks before exposing learners to thinking about these concepts together.


Principle: Practicing
Rationale: Learning occurs when people engage in appropriate cognitive processing during learning, such as attending to relevant material, organizing the material into a coherent structure, and integrating it with what they already know (Active Processing) This course, as an intro to a lot of concept knowledge in DSLR, is not offering students opportunities to practice hands-on photography. As a result, it is an important question to think about how to motivate learners to find relevance in the course to field practices.

There are some other principles that I learned to follow after testing my unit with my first four learners, which will be explained below.

VI. Feedback from 4 Users during Unit Pilot Testing

Problem 1 from feedback: “The pre-quiz is way too hard for me. I am not the right person for the lesson, right?”

Iteration: Use Personification Principle to guide the learners that the pre-quiz being “too hard for them” does not mean that the lesson is not the right level for them, but to test their current knowledge for the lesson. In the future, I might even add a few words to encourage learners that even feel that the course is “too easy.”


Problem 2 from feedback: “In the Question, ‘as the aperture becomes smaller,’ does it mean the number (‘F-Stop’ / ‘F-number’) the indicates aperture is smaller, or the aperture part of the camera is more closed-up?”


Iteration: This was a great question that needs clarification. Two different kinds of understanding (both correct) would give two opposite answers. And if I had not received this piece of feedback, I would assume that people who answered this question wrong were because they did not understand the concept, not because my question was misleading. Considering that A is still the only correct answer, I needed to rephrase this question to make it more clear: “As the camera’s aperture closes up.”

Problem 3 from feedback: “In the instruction, you have a checkpoint question about Rule of Thirds. But this question did not show up in neither the pre-quiz or the post-quiz. What’s the purpose?”

Iteration: Well, I deleted the KC right away.


Problem 4 from feedback: “No matter what I choose, the answer isn’t right?”

Iteration: This learner is talking about a multiple-choice question with more than one answers. I fixed the logic flow so that it’s going to run well from then on.


Problem 5 from own testing: In one question that follows the Contrasting Cases Principle, the two photos are placed horizontally. When tested on the mobile phone, the image is so small that the key different to be noticed is not clear. I changed it to a vertical display.





VII. AB Instructions

Baseline Instruction:
Practice (checkpoint) questions
without hints


Treatment Instruction:
Practice questions
with hints


VIII. Data Analysis

At the beginning of the data analysis stage, I only looked at the total scores including the one open-ended question and multiple choice (MC) questions. From the following two graphs, we see that having two versions of counter-balanced pre-tests and post-tests, the results are both significant.


AB Instruction
Instruction A (control): The two-tailed P value equals 0.0027: by conventional criteria, this difference is considered to be very statistically significant.

Instruction B (innovative): The two-tailed P value equals 0.0027: by conventional criteria, this difference is considered to be very statistically significant.

Learners who went through either of the instruction types have gained significant increase in their post-tests, while the innovative instruction version gives the learners a higher average score. It is intriguing to see that the average pre-test score was higher already for learners who were later going to be exposed to Instruction B. I would like to conduct more analyses in the future.

*An interesting finding
I have designed two sets of questions targeting the same Knowledge Component (KC), Depth of Field. During data analysis, I noticed that the two questions below, using two sets of photos, clearly have two different levels of difficulty. The set of photos with a person in it, shows a higher correction level when it is answered by learners in both the pre- and post- test conditions. The set of photos with objects, however, tells a different story. Although targeting the same KC, these two questions are not close enough in terms of difficulty within an acceptable level. I would like to run more test to learn about why learners find one question with sets of photos more difficult to answer than the other question.


IX. Final Reflection

What are three key lessons you learned?
1) Being a learning designer is harder than I thought.
2) Aligning goals, instructions, and assessment is key, which helps scope down the unit too.
3) There is not an end to the process! Iterate, iterate, iterate!

What challenges did you face?
I suspected that when participants decided to work on my unit, they would either “power-through” the lesson, or give up in the middle somehow. Another challenge for me is to stop collecting data and move on to data analysis. For the next stage of this project, I would like to set better timelines and understand that I have to draw a line and move on to the next step. However, this does not mean that data-collecting needs to be permanently put to a stop, it just suggests that I need to set up cycles of iterations.

How did you overcome them and/or why do some remain?
I decided to shorten the lesson and focus on focusing. After doing so, I received feedback such as “Yulin is innovative in creating a module that is simple, and avoids overwhelming learners.” The fact that 67 participants finished my unit is encouraging too.

What are your next steps, either with respect to this project if you plan to continue it, or with respect to other projects that could benefit from this approach?
I would like to keep revise my lesson (based on data and testing!) I’d like to test the effectiveness of hint using CTAs and learner’s learning curves. I also received feedback regarding the mobile-friendly side of my unit, which encourages me to develop an app on it.

The next time you have an opportunity to begin a new project, how do you plan to proceed differently?
I would absolutely be more patient and spend more time recording & labeling my participants, e.g., which learner goes through which instruction version as well as pre-test version. It would save so much trouble in the later process!

In conclusion, I think of this testing process to be well depicted by the following two process diagrams. I test, and rebuilt (based on different rationales). I constantly evaluate, but the challenge is, is there ever going to stop? (The answers would be no, I think.)

“Design, build, test”
after Alice Agogino (1 of 3)


Thank you!

This “Big Idea” document below is a synthesis of educational guidelines and concepts from the course “Educational Goals, Instruction, and Assessment” (EGIA), taught by Dr. Sharon Carver at Carnegie Mellon University. I would use this document to guide my design in the photography course above.