1. Education

7 Common Graphs in Statistics

Graphs help statisticians to visualize data. But not all graphs are created equally. Different situations call for different depictions of data. Learn more about seven of the most common graphs in statistics, and when you should use them.

More on Statistical Graphs
Statistics Spotlight10

Five Number Summary

Monday April 30, 2012
The five number summary combines five descriptive statistics to give us a fuller picture than what we might have by just looking at one of these numbers. The five numbers used are the minimum, maximum, median, and first and third quartiles. A boxplot can be used to give a graphical display of a five number summary.

Paradoxes in Statistics

Saturday April 28, 2012
A paradox is a statement or situation that seems contradictory on the surface, but may reveal some underlying truth. Paradoxes abound throughout different disciplines. A few of the ones that deal with areas connected to probability and statistics are:
  • The Barber Paradox - this paradox goes to the foundations of mathematics, and what the defintion of set means
  • The St. Petersburg Paradox - shows that an analysis of a game should involve more than just the calculation of an expected value.
  • Simpson's Paradox - reveals some of the problems that arise when we lump too much data together.

Rolling Dice

Friday April 27, 2012

A standard probability problem involves the rolling of two dice, and calculating the probabilities. But what about three dice?

Calculating probabilities for three dice is a trickier problem due to a couple of factors. Rather than 36 different possibilities with two dice, there are 216 for three. Also, there are a number of different ways to use three dice to form the same sum. For example, 6 = 1+2+3=2+2+2=1+1+4, as well as varying permutations of these sums.

Chi-Square

Friday March 30, 2012
The chi-square statistic is an important tool to quantify how much an observed value in a statistical experiment varies from an expected value. The expected values are typically calculated using probability or some mathematical model. Chi-square tests are used in two-way tables or multinomial experiments

The formula to calculate the chi-square statistic is not very complicated. There are actually less steps for the chi-square formula than for the formula for standard deviation.

Discuss in my forum

©2012 About.com. All rights reserved.

A part of The New York Times Company.