Browsing All Posts filed under »How-to«

Sampling until you get all six

June 4, 2013


A common problem goes like this: There are six different Famous Statistician cards, one (randomly) in each box of Chocolate Sugar Bombs. How many boxes do you need to buy to get all six? This is a good three-collection simulation in Fathom, and uses two little-used features, sample until and the uniqueValues( ) function. Make […]

Using summary tables in simulations

August 17, 2012


In the apstat community thingy, Rudy Medina posted a problem, and Paul Myers posted a solution. I’ll show another solution, and maybe veer off into philosophy. Here’s the problem (from 16 August 2012): A bus stop has 7 stops and 4 passengers.  If every passenger is equally likely to get off at any stop, what is […]

Streak Lengths

August 14, 2012


We humans see streaks as indications of non-randomness, but streaks do occur in genuine random phenomena—usually more than we think. There are famous activities about this, like the one where you ask students to (secretly) create “realistic” sequences of 100 heads-or-tails coin flips along with genuine sequences from a fair coin. The wise, omniscient teacher […]

Sampling Without Replacement: Lottery Problem

June 14, 2012


Yay! A request! From Ben Ceyanes on the APstat listserve: Ok, So I am brand new to fathom and to AP Stats this year and I am getting frustrated trying to figure out how to simulate a problem on fathom.  The problem says that a person buys 5 lottery tickets with 6 numbers ranging 1 to […]

Paired Tests in Fathom

June 8, 2012


This is one of those, “oh, yeah” and slap your forehead things. Suppose you have data on which you want to do a paired test. The prototypical paired test is a pre-post test of some kind. Let’s call these scores pre and post. The basic idea is this: compute the pre-post change and test whether […]

Two-tailed tests, one-tailed tests

June 8, 2012


No one is making requests, but there have been searches on this site about tests with different numbers of tails. So here, without ado, is how you specify whether a Fathom test should be one or two-tailed, and if one-tailed, which one: First, check out this post about Fathom’s test objects in general. The key […]

The Key to the Birthday Problem: uniqueValues( )

May 30, 2012


It’s a famous probability problem: given N people in a room, what’s the probability that at least two of them share a birthday? It’s famous because the answer is often surprising: a remarkably small number (23) will give you a probability over 50%. How do you simulate it in Fathom? The key is going to be […]