The Art of Giving Feedback

(Photo Credit: Veer)

Giving feedback is truly an art that one needs to master. Before I go ahead, let me make it clear that I am referring to feedback given for assessments. We had this really interesting #KernLearn session on Twitter last week on Designing Assessment in eLearning. Given below is the transcript.
----------------------------------------------------------------------------------------------------------
rnarchana 1. Do assessment questions aid learning? How can they be designed so as to enrich the learning experience? #KernLearn
partvinu #KernLearn yes, they do, the strategies are dependent on multiple variables, formative or summative, blooms level, audience maturity etc
shana1729 @rnarchana 1) IF designed better, assessments enrich, otherwise kill learning. understand learner motivation n design acordingly. #KernLearn
shana1729 @rnarchana how to design assessment in elearning- a guide: http://bit.ly/1x8C7 #KernLearn
shana1729 @rnarchana A) Numerous no. of qn kills interest B) design in innovative way- dare 2 move out of traditonal methods, yet learnable #KernLearn
shana1729 @rnarchana C) Make assessments more visual, but avoid unnecesary images. In elearning, a dry screen de-motivates learner.. #KernLearn
shana1729 @rnarchana D) Challenge the learner (if the profile of learner fits it) E) Make it game based F) Design around a story..#KernLearn
shana1729 @rnarchana F) Don't let learner think that "I am being examined" #KernLearn
shana1729 @rnarchana #KernLearn 2. I think max is 10 if it is designed in traditional methods - mcq, mmcq etc..
geetabose 1. Assessments in refresher courses: used an exercise to help learners recall skills & clearly identify areas to brushup & focus. #KernLearn
geetabose 1. What was special abt this exercise - it cud guide learners across levels based on response & help zero in on thr focus area. #KernLearn
rnarchana 2. Assements in eLearning: How much is too much? Any thumb rules you follow? #KernLearn
geetabose @rnarchana Clearly depends on the key testing points identified and their impact on learning outcome... 1-2/ LO is a good measure #KernLearn
geetabose RT @shana1729: F) Don't let learner think that "I am being examined" #KernLearn | this is an interesting point...
rnarchana 2. Learner will think its too many only if he does not see the purpose. If designed well, a quesn will have him wanting more. #kernlearn
geetabose For adult learners, their motivation is high if they know upfront they r being examined & know the consequences @shana1729: F) #KernLearn
rnarchana 2. 2 many/2 little is relative. How many testing points do U have 4 the learning objectives is. imp & R they motivating enuf #KernLearn
partvinu #KernLearn 3 screens 1 AQ, and final assessments generally 1 per objective, or 2 per topic
geetabose RT @rnarchana: 2 many/2 little is relative. How many testing points do U hav 4 learning objectives is imp & R thy motivating enuf #KernLearn
geetabose Most critical aspect of assessments that aid learning is the 'diagnostic' feedback that the learner receives - #KernLearn
rnarchana Assessments should have rewards and punishments. Increases impact. These should be designed based on the learner's motivations. #KernLearn
rnarchana RT @geetabose: Most critical aspect of assessments that aid learning is the 'diagnostic' feedback that the learner receives - #KernLearn
rnarchana Designing assessments is not about tricking the learner into making errors. It is about making them think rather than guess #KernLearn
geetabose Use assessments creatively- enable learners to demonstrate knowledge, explain reasoning, develop argument or evaluate a situation #KernLearn
geetabose What's the most creative use of assessment that you have seen? Share some examples? #KernLearn
vijeesh @rnarchana Assessment Qs: Response and analysis based assessment Qs can work wonders #KernLearn
vijeesh @rnarchana 2. 10 interactive and motivating Qs per half an hour #KernLearn
vijeesh 1. Most important! RT @shana1729: @rnarchana F) Don't let learner think that "I am being examined" #KernLearn
rnarchana An example for interesting assessments: http://tr.im/OpPJ Check Employee Security, Front Line Loss Prevention, #KernLearn
rnarchana Examples of interesting assessment in ELearning http://tr.im/OpQB Check examples of scenario based learning #KernLearn
rnarchana Examples of interesting assessments in eL http://tr.im/OpRE Check Pre-Op intro and Main Surgery #KernLearn
varmarashmi RT @rnarchana: 2. 2 many/2 little is relative. How many testing points do U have 4 the learning objectives is. imp & R they motivating enuf #KernLearn
varmarashmi RT @geetabose: RT @shana1729: F) Don't let learner think that "I am being examined" #KernLearn | this is an interesting point...
varmarashmi RT @geetabose: Most critical aspect of assessments that aid learning is the 'diagnostic' feedback that the learner receives - #KernLearn
varmarashmi Assessments enhance learner involvement and interaction (also visible in twitter sessions) #KernLearn
-----------------------------------------------------------------------------------------------------------------

While there are several thoughts here that interest me, the one that mattered most is the one shared by Geeta: Most critical aspect of assessments that aid learning is 'diagnostic' feedback that learner receives.

Tell me something...

1. While designing feedback, do you ever think, 'How will my learner react to the tone of the feedback?'
Why is this important, you ask? What is the standard feedback that you use "That is correct. or That is incorrect? I will tell you what I think is wrong with these (I could be biased as I hate conventional ways).
a. It just sounds too robotic! The learner will find no joy in getting it right. On the other hand, if he does get it wrong, it is going to have 'zero' impact on the learner.
b. It sounds condescending. If your learner is the sensitive/timid/introvert/self-respecting type, he is going to be offended. Learner motivation dips and you have lost him.

What can you do instead?
Think about the tone. Think about the feeling that you want to evoke in your learners. Use direct feedback to make them feel good (You are right! or Absolutely! or We agree!). Use milder (but still effective) ways to tell him he got it wrong (Oops! or Nope. or Really? or Are you sure?). Please do not reprimand. We want to encourage him to learn and not make him feel small.

2. Do you ever think about what you want the feedback to do?
For example, how does the following feedback help?
Feedback for incorrect answer: That is incorrect. Rakesh falls in the low income group.

This learner has got this assessment wrong. This means that 'something' went wrong. She has not understood what you have explained previously. (Remember that is also not her fault) This is your opportunity to ensure learning is checked. This is your chance to set things straight. Here, you can explain why Rakesh falls in the low income group. If you don't, the learner is not going to go back and check (not even the studious ones go back). He will move ahead and your learning objective is already in danger.

What can you do instead?
Do not underestimate the role of feedback. It can play a very crucial role in checking misconceptions/understanding. We all know that assessments are designed to check understanding. But checking does not mean identifying whether they got it right or wrong. By checking, I mean providing them the rationale. Tell them why they went wrong. It is not important to prove them wrong but to make them understand why it is wrong. Even if the learner gets it right, the rationale reinforces all that she has learnt. When she reads it, she feels good that she also had the same logic in mind. If it was good guess work, at least she can read why it is right.

3. Feedback can be more than just text!
Text is just one way of giving feedback. Coming back to a point that I have touched briefly, what is the desired impact? Do you want to the feedback to have a high impact? Make the learner think? Use more than just 'That is correct. or That is incorrect.'

What can you do instead?
Audio: You can use audio to let them know how they have done. Audio is a very powerful medium of getting feedback across. You must have seen how games such as Mario Brothers, Pacman, Tetris, and others use audio to indicate whether your move was a good one or bad.
Visual: Show the consequence of an action as a visual. For example: image of a happy or a angry customer.
Rewards and punishments: Add a challenge and raise the stakes. This ensures high impact. Some simple ways to do this could be points system, a (virtual) pat on the back, a big jump ahead, happy ending, and so on.

3. Choose my feedback strategy based on my learner.
But most importantly, chose your feedback strategy based on your learner (or as we say at Kern, learner persona). Why? Imagine this.
Your learner has to take your course because your research shows that this is a skill that he needs help with. But your learner believes otherwise. He thinks he knows everything he needs to about the topic.
So what will you do? I would suggest you go for rewards and punishments. If you go by the conventional method, he will think he knows better and that the exercise itself must be flawed. So, challenge him. Trust me this works like a dream. At the end of the course, the learner still feels good about himself and he has also learnt a lot.

If your learner persona is a shy type who is afraid of getting it wrong, keep your strategy fairly simple. Encourage him all the way and he will do his best to learn. Therefore, chose your feedback strategy based on who your learner is. Understand their motivations and design your feedback accordingly.

Next time you are writing feedback for an assessment, ask yourself 'What do I want to do with this feedback?' (I pray the answer is not 'nothing' :)) Let us work on mastering the art of giving good feedback. Do share instances of how you use (un)conventional feedback to teach better.

Understanding Your learners: All things that you can do wrong

Jai and Prachi live in Bangalore with their 2 year and 2 month-old son, Druv. Druv is an active kid. He loves watching Hanuman and Jungle Book. He repeats words as soon as he hears them. He stays at home with his grandparents and parents. They typically converse in their mother tongue. He is the darling of the house. Thinking it is time, the parents start looking for a school for their son. They research the best schools in the area. They visit a few and select one. They approach the faculty for admissions. A faculty member mentions that the kid is a little younger and therefore, the parents will have to meet the headmistress. An appointment is made with the headmistress. The parents and Druv finally meet the headmistress. The headmistress takes a few plastic fruits and asks Druv to name them. Druv gets Apple and Banana right. She then shows him a watermelon. Druv has never seen a watermelon before and therefore, refuses to answer her. She continues to ask him the name of the fruit. Finally, Druv gives in an says "Stop it!" and talks in his own language (child talk). The headmistress says, "Your son does not know English. How will he fit in class?" She asks Druv's mother to ask a few questions in their mother tongue. Druv promptly answers all these questions. The headmistress watched him closely. She then tells the parents that their son may have a speech problem. She says that she has been in this industry for a long time and has seen several 2 year olds. Children of his age should be able to speak a complete sentence. The fact that Druv was unable to do so and was blabbering shows that Druv has a speech problem. She suggests that they take him to a spastic society and get some tests done. She also suggests that they come back with the test results. She mentioned that this school also took in special student and therefore, she can help them. The parents were shocked and quite stunned to react. The parents were later extremely angry and offended. They checked with Druv's pediatrician. She was equally angry. She explained that kids at the age of two connect a few words together and cannot be expected to talk fluently.

VIBGYOR High is the name of this school. This is a real incident. The parents discussed here are people very close to me. Druv is like my very own son. I also have a two year old son. He does not talk in English and cannot talk complete sentences. He only connects two or three words together. So what was my reaction to this? Why am I sharing this here? I am absolutely appalled by the headmistresses behavior. Why?

1. You do not judge kids based on their behavior the very first time you meet them: Children are different at home and extremely different in front of new people. The headmistress who has seen several 2 year olds should have known this. Before you reach such conclusion, you must spend some time with the child.

In our realm: Do you make rash judgments about your learners based on their behavior the very first time you meet them? There is always more that what meets the eye and it is your duty to understand the complete picture before you describe a learner. Remember how offended the parents were. If you jump to the wrong conclusions about your learners, you are going to offend the company.

2. Stress interviews are not for learners: At home, you may teach your kid names of cars rather than fruits. How can the headmistress expect the kid to name the fruits without understanding what he already knows? Is it a crime for the kid to not know what a watermelon is? She should have made the kid feel comfortable, rather than forcing him to answer. Is a stress interview really necessary for kid?

In our realm: Do you make your learners feel comfortable when asking them questions about their work? Do they feel like they are being 'tested' or judged. There is no doubt that it is exactly was this headmistress was doing. Judging the kid...

3. Do not expect the learner to 'know': Any kids goes to school to learn languages, numbers, shapes, colors, etc. The headmistress had no right expecting the kid to know English and names of fruits.

In our realm: Do you assume that your learners must know certain things? When you meet your learners, you must be free of preconceived notions of how they must behave, they must know, and what they need to learn.

4. Handle learners with kid gloves: The headmistress should have been more sensitive to the parent's feelings. She should have shared her thoughts on the kid's speech in a very tactful and sensitive manner. You cannot just declare such things out loud. If you want the parents to take you seriously and not get offended, you must ensure that you have enough facts and that you consider their feeling before giving them that information. This headmistress was hardly professional. She neither took the time to understand the kid better, nor did she handle the parents well.

In our realm: Do you identify learning gaps after a detailed analysis and backed by research? Or do you general push a list of courses that you think may solve a surface level issues? Be sensitive to your learner's and client's needs. Treat them with care and more importantly, with respect.

Please do not put your kid in VIBGYOR High. If the headmistress is like this, imagine how the staff would be. All children (learners) are different. Each has a different learning curve. Some kids start walking earlier, some starts talking later, and so on. It would be wrong to pass judgments without really understanding them. Someone I know started talking only when he was 5, does this mean he that he had a speech problem? I don't think so. He speaks fine. If headmistresses and teachers don't understand that each child is different, I dread the day when I have to put my son in school! Finally, I really don't understand why schools want to take only the so called 'smart' children. What is great in taking the cream and showing to the world that you get the best grades? Take all children and do the same and that would be truly great! Do we have the option of making just the smart learners take the course? No. This would be crazy! Infact, our focus is on the so called low performers. At this point, I must warn you: When you request your client to share names and numbers of learner who you can talk to, they may give you their best performers. Ensure that this is not the case. You want to talk to all 'types' of learners and not just the management favorites.

My last and final point: During learner analysis, please do not be this headmistress! Acting as she has, you will only damage the learner and the learning. Understanding your learner requires higher EQ. Be human...

Hate to Rework?

What is your attitude toward rework? How often have you heard or said 'I already changed that! You want me to work on that again?' Rework can really demotivate you, especially when you are new to the field. You will see rework as you not getting it right the first time or reviewer not appreciating the effort you put in. Rework can also be very mechanical. You just need to add a line here, replace an image or remove word there.

When is it ok to rework several times?

a. When you are exploring a new approach: When you are trying something new, be prepared to rework. You are bound to realize that there are loose ends; things you had not thought of earlier; newer ideas that you think will work better; that old ideas do not have the necessary impact and so on.

How to tackle: Ensure that you are working in a group. Extra heads help identify the issues at an earlier stage. Have smaller milestones and frequent reviews.

b. When you have time to improve the quality: You know that your design is good but not great. The reviewer points out relevant value adds that can make your design great. Be open to this.

c. When you want to ensure you get it right: When is it important to get it 100% right? It is important to get it right when:

• the training is in line with a primary role/skill
• you want to being a attitudinal/behavioral change
• you need solutions to bring about a mind shit
  
• you have promised results

d. When the reviewer is adding value: It is important to allow the reviewer to add value to your work. If it can be improved, then why not? Be open to newer ideas.


When is rework frustrating?

a. When you don't see eye to eye with person giving the feedback: The person giving you feedback may be a client, an authority, a SME and so on. But, never fix something because you are being told to do so. Ensure that you are convinced that the change is a value add and will aid learning further.

How to tackle: Have long drawn discussions about why this change needs to be made. Share your concerns. Ensure you reach a consensus.

b: When there is no clarity on what the reviewer is thinking: This can be very frustratedly. If you continue blindly fixing, you will have several version before things are clear in the head of the reviewer.

How to tackle: Have detailed discussions to understand what the reviewer is trying to say. Ask the right questions to ensure that the reviewer thinks further. Do some research and share information with reviewer. Also, double check by restating what you have understood and what you are going to change.

c. When new things creep up at every round of review: This happens to most of us. Where the reviewer is pressed for time and therefore, scans through the storyboards and shares a top level feedback. This never gets over till they actually sit down and go through it.

How to tackle: If possible, arrange a meeting with the reviewer. Take them through the storyboard and fix issues in front of them.

d. When you are pressed for time: If you are pressed for time and the review cycles are just not getting over, there is a major problem. The possible reasons:

• Time-effort allocation for this project was incorrect.
  
• The reviewer has great expectations.
  
• You are just not cut out for this work.
  

Attitude to Rework

Why did I feel the urge to blog on such a topic? This is because I used to hate rework myself. But, over the years, I would like to believe that I have checked this attitude. Two months back, I worked on a project that required extensive rework at several phases. I realized that (though I was frustrated at times) every time I reworked the product looked better. It is highly satisfying when you look at rework from this point of view.

Rework is probably as important as writing the storyboard for the first time. Be open to it. Respect the people you work with. Remember most people want to ensure that we have a good product. Your goal should also be aligned to this. It will help if you reduce obvious errors while storyboarding/fixing. The more challenging the project, the more the chances of rework. Follow this and it will help reduce rework:

1. Ensure you understand what is required clearly before attempting to do it.
   
2. Ask the right questions so that you have all the answers.
   
3. Ensure that you are totally convinced about what you have done. If you are not, the reviewer is surely not going to be.
4. Be proactive. Do some research to get a clearer understanding if you are unable to get it from the reviewer.
   
5. Bounce ideas off peer if you are stuck.
   

Scenario Based Learning

Check out this SlideShare Presentation:

Scenario Based Learning

View more presentations from Kern Learning Solution.

Speak Out Featured in eLearning Learning

I am absolutely thrilled to announce that Speak Out is now a part of eLearning Learning. For those of you who have not heard of this site, I recommend you visit it now. The tag line of this site says it all: 'A community collecting and organizing the best information on the web about eLearning'. This site acts as a huge reservoir of information. The best blogs in the field of eLearning are featured here. Tony Karrer has brought together the best blogs under one roof.

If you are a learning professional, here are the reasons why you must visit eLearning Learning:

1. This site includes the latest posts by all the great bloggers (who you are probably already subscribed to). The difference being you can now go to one site and find all the new posts there.
   
2. Search has now been made easier. If you have a query, what do you do? Go to your favorite blog and hope the author has written on it? Google it? Now, all you need to do is visit this site and search. You will find relevant posts from several blogs.
   
3. Tony Karrer makes life even easier by assimilating a list of best blog posts for a particular category or month.

Quoting eLearning Learning:

The goals of eLearning Learning are:

Collect High Quality Content - The goal of a content community is to provide a high quality destination that highlights the most recent and best content from as defined by the community.

Provide an Easy to Navigate Site - End users most often are people who are not regular readers of the blogs and other sources. They come to the content community to find information on particular topics of interest to them. This links them across to the sources themselves.

Be A Jump Off Point - To be clear all content communities are only jump off points to the sources of the content.

Help Surface Content that Might Not be Found - It’s often hard to find and understand blog content that’s spread across sites. Most users are not regular subscribers to these blogs and other content sources.

Do visit it to see what this site has to offer.

7 Traits an ID Must Have

I know I have already blogged about how most people outside the industry frown on eLearning. I am ready to crib about something else now. :) Why do people think instructional design is easy?

Case 1:
Acquaintance (who happens to be a software engineer): So, what do you do?
Me: I am an instructional designer. I design learning programs for corporates.
Acquaintance: Really? I am usually free during weekends, give me some work. I would like to do something.
Me: Yeah ok.

Thinking to myself: Do you know that you need to learn how to do my work? I was trained for a whole year and am still learning about how I can work better. You think you can master this task over the weekend?

Case 2:
Acquaintance: Hi, I hear you work from home. Can you get me a job like that?
Me: I don't think so.

Thinking to myself: Do you even know what kind of work I do? How do you know whether you will like it or not? I work from home, but this is 'serious' work. (I get this very often. People think my job is a hobby and not a 'real' job.)

Case 3:
ID: I didn't realize storyboarding was so tough. I thought having good writing skills was sufficient.
Me: Well, ID is a lot more than that. (I go on to share my experiences and crib about how people think instructional design is easy.)

Coming back to my original question: why do people think being ID is easy? Is ID about throwing a few slides together? Is ID about content chunking/content development? I'd say it takes alot more than that. If you are looking to take up a job as an instructional designer, you may want to check whether you have the following traits:

a. Ability to read and grasp varied subjects: Do you read a lot? Do you read all sorts of books/blogs? An ID should be a voracious reader. He must have the ability to read anything and grasp what is being said.

Why?
You will come across diverse subjects in your career as an ID. You cannot afford to be 'comfortable' with just one domain (such as soft skills/technical). You should be able to read all kinds of subjects and understand what is being conveyed. Be it rocket science or how to dress to a prom, you should be able to read it and understand it.

b. Deconstruct theory and convert into practical examples: You should be able to look beyond the theory and search for real life examples. You can pick on the SME's brain, Google it, read books/blogs, and so on. But, get the information that will help you design learning effectively. Being an ID is not about copying information from one source and transferring to another source. What you do with this information is important. Transform it, make it learnable. (Believe me, it is easier said that done)

Why?
There are several reasons. Everyone understands a concept better when you share an example. Examples help apply knowledge. What will people do with theories? They want information that they can use. The SME may not be helpful enough to understand why theory is not sufficient or he may be too lazy to do the research for you. So, get used to it. Don't depend on others, go looking for the information yourself.

c. Understand the DLC: Instructional design is not just a part of the process, it is the process. ID is not only about identifying training needs. It is about identifying what the need is. It may not be training the way you understand it. It may mean relooking at the organization structure or the policies and procedures. Even if you are going to play a small role in this process, understand how the process works. Instructional designing is no longer (if it ever was) about storyboarding. It is a lot more. ID starts when you talk to your client, do research to understand your learners, propose the solution, design the table of content, brainstorming about the mode of delivery and implementation plan, storyboard, develop the program, do a QA, test it, and roll it out. Ensure that you know about the entire process. See how you can contribute at all times.

Why?
If you are lucky, you may get to play a part in all stages of the process. You shouldn't have to learn about this later. Believe it or not, in all my projects, I am involved right from the word go till the very end. The project is mine and I know it inside out. I believe in it and know what works for it and what doesn't. I work in a team. My work is not done when the storyboarding is done. I love how my boss (Geeta Bose) puts it: This is not an assembly line where everyone plays a small part and exits. This is so true. If you are designing a part, quit now and join a company that will allow you to design the product itself.

d. Being sensitive to people's needs: Are you intuitive? Are you observant? Are you non-judgmental? If the answer is yes, you will be a good ID. You need to understand your learners. You need to observe them at work and pick up important details that the client will not share with you. Don't judge them for being the way they are. Like Abhinava (@Abhinava) says, love your learners and they will love you.

Why?
It is important to be attached, not detached. You need to understand what your learners feel and why. How can you fill this gap? Getting these answers will help you design an effective learning program.

e. Keep your feet firmly on the ground: Every day you will read about new technology, how it is great and is going to replace all other forms of teaching. Read it. Think about it. Use it. But do not replace this information with whatever you have learnt so far.

Why?
Designing effective learning solutions is not about using the latest technology or applying the latest theory. It is about identifying the right solutions given your audience. Do not look down upon any form of teaching. Your solution may lie in ILTs, eLearning, out bound training, job aids, a book, restructuring of an organization. Keep your options open and choose wisely.

f. Visualize what you read: Do you visualize what you read? Do you have an imagination? I know there are visual designers for this job. But I firmly believe that as an ID, you need to visualize what the program is going to look like. A job of an ID does not end with content chunking and organization. It also involves visualizing how each screen or idea will unfold.

Why?
Visualization is not only about graphic elements. It is about how graphics, text, audio, and other learning elements come together to make information learnable. Visualize how the complete training program will unfold. If you are clear about this, the product will be brilliant.

g. Openness to learn: Are you open to learning yourself? It is not only about making others learn. It is equally important for you to learn. Unfortunately, several people think work pressure is a good enough reason to not learn. This only means that you are not managing your time well. Invest in learning and you will do well. Network and learn from others. Find out what others are doing, listen to what they are saying.

Why?
It is important for you to believe in 'learning' if you are going to want your learners to do it. You have to constantly learn to ensure that you have all the information and the skills required to do your job well. What you learn may not be useful immediately, but will help you in the long run.

h. Respect yourself: Are you used to voicing out your opinions? Will you fight for what is right? I know it sounds idealistic, but it really helps. IDs must questions why. Believe in what you are doing and stop blindly following orders. You respect yourself and people will automatically respect you.

Why?
It is important to know why certain things are the way they are. You may raise questions that others haven't thought of yet. Clients will appreciate that you are truly trying to design an effective program.

I am sure there are several more things that can be added to my list. But, I have to stop somewhere. Finally, do not become an ID if:

1. You think its is going to be an easy and comfortable job. No job is easy.
2. You love technology and are dying to use it. Design for learner and not because you want to try out a fancy technology.
3. You think you know a lot and can share it. Like Abhinava says, it is never about you. It is always about the learners. You may know lots or nothing. As long as you are willing to learn and read, you will do fine.

Read Instructional Designers need more skills than just writing! to really know what ID is really about (while you are there, check out the other great posts as well). If you already in the field, read 5 things an ID can think differently about!

Have you ever got the impression that people thought your job was easy? Share your experiences here. Also, if you have anything to add to my list, feel free to do so.

Characters in eLearning

I read this interesting post on Have you thought of Character Driven Stories for Your eLearning? by Rupa (@ruparajgo). I was tempted to blog about it myself (thanks for the inspiration, Rupa).

What role can a character(s) play in your eLearning porgram?

1. Expert: This is probably the most common use of a character in eLearning.
Type 1: This type takes you through the course and is a constant feature. You can design an expert who will take the learner through the course. This expert simplifies information or provides useful tips drawing from his experience.

The character takes on the role of a mentor. The character symbolizes wisdom and is always present to see the learner through tough situations. He/she could encourage and motivate the learner through the course.

Type 2: This type "pops up" when necessary. The role could be defined such that he/she makes an entry to challenge, guide, provide useful information, and so on. This character supplements the content when necessary.

2. Peer: The learner is introduced to the world of a peer. We define the environment and then ask the learner to observe the peer in action and help when the peer gets stuck.

Peer is the 'damsel-in-distress' (not necessarily a damsel). The peer depends on the learner to help him/her through a tough situation. The learner has to make the right decisions to get the situation under control.

3. Guide: The guide is a another common character in eLearning. I have seen several online courses that have an animated character on the top left corner that lip syncs the audio (Her eyes follow your cursor. It is a little freaky.) I have never understood the value these characters add to a learning program. They are neither experts, nor peers. They are the host, who accompanies you through the course.

4. Trouble maker: This could be a boss, an expert, a peer, a competitor, or an enemy. This character challenges the learner at specific instances. Their feedback is blunt, even rude at times. They reprimand the learner if he/she goes wrong and grudgingly accept if they get it right.

The relationship is that of power. The character throws a challenge: Let us see how to get past this hurdle. The learner has to make the right choices to save face or 'win'.

Like Rupa mentions, characters make the course more lively. Remember the following when creating a character:
Give them a personality: I love building a persona for the character. Give him a name, a background, prominent traits, and so on. When creating such a character, I think to myself, what impression do I want this character to make on my learners. Do I want my learners to admire him? Do I want them to empathize with the character? Do I want them to respect him?
Ensure Consistency: It is important to ensure that the character is consistent in behavior across the course. You do not want a timid character being very bold in another scenario. The learner will get confused.
Weave the storyline well: The storyline must blend in well with your theme. Do not introduce characters in an abrupt fashion. Build a simple storyline. Ensure that the storyline is in sync with the theme. Also, ensure that the storyline will flow smoothly through the complete course. Some become too forced if not thought through.
Do not use characters as decorative elements: Characters must aid to learning and the overall learning experience. These characters are not meant to be eye candies.

Sharing some great resources on this:
Learning Agents Part 1: Why Learning Agents
Learning Agents Part 2: Learning Agents Done Well
Learning Agents Part 3: Done Poorly