Recent postsView all

Exam Season
13 Apr 19
Compound Interest
23 Dec 18
This Years projects 2018
14 Sep 18
Final Report to teachers
17 Jun 18
Problems with surveys
7 Jun 18
20 May 18
May TZ2 exams
10 May 18
Subject report and IA
1 Mar 18
New Curriculum
10 Feb 18
24 Sep 17
First Draft Feedback
1 Jul 17
Venns in the news
23 May 17

Exponential (and tooth) decay

Sunday 17 January 2016

Ever decreasing m and ms

Just a quick post to remind everyone of this classic activity. I first saw it on a workshop with Jane Forrest and new it would be a winner. It is a wonderfully simple way to look at modelling exponential decay.

In short

  • Get a large amount of m and ms,
  • Pour them out on a tray, count them and then eat all of the ones that have the 'm' showing upwards.
  • Then shake them all about, count them and eat all the ones with the 'm' showing upwards and so on.
  • Keep going until they are all gone and keep a record of the number of m and ms left after each 'shake'


With the flatter, chocolate m and m's, we can predict that the amount left will reduce by approximately 50% each time. The less regular peanut m and m's are less predictable. In any case, the small amount of data is very easy to enter using a tool like Desmos. Then the key is to ask students to think about the features of the function. Start with the general expression....

begin mathsize 24px style f left parenthesis x right parenthesis equals k cross times a to the power of x plus c end style

You may have already established in previous activities that the y intercept of the function is given by k + c, whilst the asymptote is at y = c. Students can then use the features of their data to fit a model, that they can adjust.

From an 'understanding' point of view, this is infintiely preferable to an 'auto fit' function that many packages offer.

Once the technique has been practiced it can be used on other, more complex exaples such as those here in the Modelling World Population Growth activity.

Short, sweet (literally) and powerful!


To post comments you need to log in. If it is your first time you will need to subscribe.