Showing posts with label how. Show all posts
Showing posts with label how. Show all posts
Sunday, April 16, 2017
Matchmaking how does it work
Matchmaking how does it work
Imagine that you have four daughters and you want to marry them. You carefully select four guys and let each of your daughters spend some time with each guy. Then all the daughters and all the guys mark their potential partners in the order of their priorities. How do you use this ranking to match your daughters to the guys to create four stable couples? Couple is considered stable if each of your daughters is happy with her partner and she wouldnt rather be with some other guy who simultaneously also prefers her to his current match.

Or take another matching game. Four medical residents need to be matched to work in the hospitals. Each of the residents spent some time practicing at each of the four hospitals in the program. Then all the residents and all the hospitals mark each other in the order of their preferences. Your task is to use this rating to create a satisfactory match, sending one resident to each of the four hospitals. Each resident should be sent to a hospital where he is happy. It may not necessarily be her or his first choice but there should be no other hospital that he would rather be at that simultaneously also wants him more than the current resident you sent there.
Try playing such match-making and when you are done, watch this video. Click to see Harvard Postdoc Emily Riehl explains the optimal matchmaking algorithm for such situations. You can share this with your 10+ year old kids.
You can also see the explanation of the same algorithm applied to the National Resident Matching Problem via this link.
Note that while this algorithm is simple and is always guaranteed to converge to the optimal result, the algorithm is not symmetric. The optimal solution for girls is different than the optimal solution for boys, and optimal solution for residents differs from the optimal solution for hospitals.
Top image by imaekelley, distributed under CCL.

Or take another matching game. Four medical residents need to be matched to work in the hospitals. Each of the residents spent some time practicing at each of the four hospitals in the program. Then all the residents and all the hospitals mark each other in the order of their preferences. Your task is to use this rating to create a satisfactory match, sending one resident to each of the four hospitals. Each resident should be sent to a hospital where he is happy. It may not necessarily be her or his first choice but there should be no other hospital that he would rather be at that simultaneously also wants him more than the current resident you sent there.
Try playing such match-making and when you are done, watch this video. Click to see Harvard Postdoc Emily Riehl explains the optimal matchmaking algorithm for such situations. You can share this with your 10+ year old kids.
You can also see the explanation of the same algorithm applied to the National Resident Matching Problem via this link.
Note that while this algorithm is simple and is always guaranteed to converge to the optimal result, the algorithm is not symmetric. The optimal solution for girls is different than the optimal solution for boys, and optimal solution for residents differs from the optimal solution for hospitals.
Top image by imaekelley, distributed under CCL.
Available link for download
Labels:
does,
how,
it,
matchmaking,
work
Friday, April 7, 2017
LSG Webinar How great content systems and communications create provable learning results
LSG Webinar How great content systems and communications create provable learning results
- How a portal approach enables rapid, targeted access to learning
- Why a communications campaign is essential for adoption
- How elearning speeds up time to competence
- Calculating and demonstrating cost savings
- Showing provable increased performance by learners
OK, Im interested particularly in how people are improving business performance with learning.
So Steve asks: What do you think a learning function should measure and report on?
Some interesting answers from the group:
- Improvement in ability to carry out tasks
- increase in meeting targets
- that skills base increases
- improvement in performance
- Expectations of stakeholders (ROE noit ROI)
- were they appropriate to the role
- impact of learning - efficiency, effectiveness
- Return on expectation
- Should be based on business reqs and Learning outcomes that result from the lna
- behavioural change
- Committment
- What the buisiness expects
The Company and their Challenge
Ford was founded in 1903 - one model and one colour of the car. Now they obviously create so many different kinds of vehicles. The profits they make are related to the technological advancements for their cars and even the extras generate huge profits for the business. Ford Europe is the subject of this case study - they have 8000 dealerships with 25000 dealers. So thats huge scale!
Jacob - Head of training at Ford had a challenge. Traditionally, people would go to Henry Ford College. People join, and after 3 months of trial you get hired by the dealers (youve sold from Day 2, btw). After this you complete the minimum training and then by month 37 youre fully trained. Unfortunately by month 37, 40% people left or dropped out. So their problems were:
- 40% staff turnover
- People never trained in Ford Brand
- People rarely trained in all products
- People never had any tests on their competence.
So Jacobs objectives were:
- Increase vehicle sales
- Increase profits - Do more for less
The most important part of executing the project was in the space of Change Management, apart from all the L&D aspects.
What they did and how they did it
The big thing for the team was that elearning was a new way of learning and they had to construct different messages for each group of people in the corporate heirarchy and I think this is key as is the delivery of the message. One size fits all rarely fits anyone. Steve is showing how they used a standard look and feel and how they varied the media to suit different people and different purposes. It seems like a very professionally executed change programme. I see greenbacks floating -- this is bound to be costly, but Im sure its well worth the cost. Ok - time for 1 more demo!
Steves now showing another of these introductory videos explaining the benefits of the elearning programs. The look and feel looks really polished - its more of the same though. The idea is to communicate effectively.
The results
They also released 24 different language versions, so people could access training in a language that made sense to them.
Now Ford is looking for ways to make the learning modules even shorter and to increase the number of delivery platforms to take the materials to mobiles, tablets, etc. They also have a request to move materials to a format where sales people can use these with customers as well! Sounds like theyre loving it, arent they?
Metrics
93% find the training useful. 81% people find it useful to sell vehicles before the launch event. Most importantly, amongst two people selling cars for 5 years, the person who had done elearning outperformed the people who hadnt done elearning almost consistently. More importantly the people who scored highly on the elearning did better on sales than the others. Thats a very powerful statistic to send out to management to talk about the success of the program. Attrition had gone down tremendously as well!
In summary, this is a hugely successful program from Fords perspective. The top tips;
- Pursue business metrics.
- Dont underestimate Change Management.
- Fun, engaging, interactive content is critical.
- elearning really works!
© Sumeet Moghe
Available link for download
Monday, March 27, 2017
Look How Far Hes Come!
Look How Far Hes Come!
Our Baby L was born exactly 1 year ago. 6 lbs. 15 oz. Such a sweet little man.
One week later he became sick and was admitted to the NICU. Eleven gut-wrenching days. But God sustained us and never left us. And despite the fact that his MRI showed 3 brain bleeds, despite the fact that he has overall low muscle tone, despite the fact that our therapists didnt even expect him to crawl by age one...
Yup! Thats Baby L standing all by himself 2 days before his first birthday! God is good!!!
L1 waits expectantly.
Hmmm...seems interesting.
Oh, yeah! This is definitely my favorite...his first bite. I laugh out loud every time I see this picture.



Oh, was it a mess! I have to say a big THANKS to Chad who totally took care of clean-up.
After a quick dunk in the tub, he was ready to open presents.
What a big boy!!Available link for download
Monday, February 27, 2017
LSG Webinar How to design learning aka “Serious” games
LSG Webinar How to design learning aka “Serious” games
This is my second webinar with Patrick - the recap of the last one is here. Patrick Dunn has been designing, producing and thinking about various forms of learning technology for more than twenty years. I particularly enjoyed his last talk and Im guessing this one will be quite exciting too.
So, here are the live blogged notes.
Don starts off in his (now famous) midnight presenter voice and is introducing Patrick in the way only he can. And after Dons usual house keeping, Patrick begins. Good morning, Patrick.
Typically in the west, people spend more on games than on films. Modern games are starting to permeate peoples lives in different ways. The above video from YouTube is a nice one to watch and see how this landscape is changing.
SG = OP + LO + PC (Patricks formula to explain serious games - explained below)
Serious Games = Organised Play with Explicit Learning Objectives usually delivered on Computers
Learning Games are part of a wide spectrum of learning exercises. You mix ingredients to come up with an experience that actually solve problems. So you mix elements from traditional elearning and learning games to create learning that meets a business objective.
So some of the elements of game design/ and even a good learning experience:
- Relevant challenges
- Sensory stimulation
- Flow
- Learner Centricity
- Repetitive Rehearsal
- Personalisation
- Safe failure
- Socialisation
- Incremental challenge
- Guiding feedback
- Contextualisation
- Stories
- Emotional Engagement
- Characterisation
First things first we can categorise most of these elements into 3 areas:
- Simulation
- Sensory Stimulation
- Repetitive Rehearsal
- Safe failure
- Guiding feedback
- Motivation
- Flow
- Relevant challenge
- Learner Centricity
- Personalisation
- Socialisation
- Incremental Challenge
- Narration
- Stories
- Emotional Engagement
- Characterisation
- Contextualisation
So heres the notion thats really makes a difference:
Extrinsic Games: Youre just dressing up a dry topic to be a game.
Intrinsic Games: Youre helping learners practice a skill by playing a game or simulation.
Big questions to ask ourselves when designing the game
What game is it?
How do you play?
How do you score?
How do you win?
And heres where my webinar screen goes kaput! So all Im doing is following Patricks voice! So pardon me for being stupid with my writing. Oops and Ive lost sound too.
OK - Im back after logging in again. Patrick is now showing a bunch of learning games examples I guess - I still cant see whats on. Patricks now showing a great example from Cathy Moore.
The idea is that all successful games produce an emotional reaction. And the challenge is to cultivate these reactions from a design perspective.
- Some questions about your story:
- Whats the perspective?
- Who are the characters?
- What are the surprises and twists?
- How does the story resolve itself?
- How will learners feel when they play the game?
- How do you establish the mood?
- Most importantly how does this relate to the business objectives?
We play by exploring, calculating, shooting, picking clues, etc. So there are bunch of things to take care of when trying to plan this interest.
- Micro-macro structure?
- One thing at once/ many things at once?
- Few similar interactions/ many different interactions?
- You could try different things:
- branching journeys - like the branching simulations you can create using traditional tools;
- guided discovery - more like the controlled learning games from Thinking Worlds;
- open - more like the games that we usually play and of course you can create these using Thinking Worlds too.
You have a lot of choice when designing things - so you need to be able to nail down various aspects of game design well in advance.
Most importantly, your game needs to give you feedback on whether youre doing well or not and whether you failed or succeeded. Feebdack is crucial - you need to decide how much is intrinsic and how much is extrinsic.
Top Tips:
- Play games yourself!
- Think experience not content
- Be 100% clear of learning rationale ("Why am I designing a game?")
- Really know your learners
- Plan your architecture (macro structure) carefully.
© Sumeet Moghe
Available link for download
Subscribe to:
Posts (Atom)