If this seems a little murky, Jeff Sauro has an awesome online Introduction to the Margin of Error Tutorial.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Almost there! You’re probably familiar with polls where results are reported like this:
“95% of Americans prefer bananas to apples, +/- 3%.”
That “+/- 3%” is the Margin of Error. Another way to say this result is:
“Between 92% and 98% of Americans prefer bananas to apples.”
All you do is add that Margin of Error to your Adjusted Success Probability to create the Confidence Interval.
| Calculated Data | Symbol | Formula | Value |
|---|---|---|---|
| Margin of Error | m | SEx x t* | 0.095 x 2.262 = .215 |
As we noted earlier, the higher the Confidence Level, the bigger the Margin of Error. We chose a 95% Confidence Level. If we had chosen a 99% Confidence Level instead our Margin of Error would have been bigger than .215.
Also, the larger the sample size, the lower the Margin of Error. That is, the more participants, the lower the Margin of Error. Our sample included 10 participants. If we had 15 or 20 participants, our Margin of Error would have been less than .215.
It’s important to note that in usability we will likely ALWAYS have a small sample size so our Margin of Error will always be pretty large. We talk more about that shortly.