Survey of Data Science - Margin of Error Calculator


#1

Hello,

I am currently going through the Survey of Data Science module. I am on page 3 “The data science process: Determine the Necessary Data”. I am hung up on the definition of Margin of Error. The text explains:

  • Margin of error - The furthest we expect the true value to be from what we measure in our survey.

Is margin of error just a guess?

When playing with the calculator on the same page I was surprised to find that increasing the margin of error actually decreases the number of samples required to be an accurate sample of the population. Does increasing the margin of error always result in fewer samples to be tested?

The values of the calculator are:

Margin of Error: 5%
Confidence Level: 90%
Population Size: 50000
Likely Sample Proportion: 50%
Sample Size: 269

The second calculation that had me scratching my head is:

Margin of Error: 15%
Confidence Level: 90%
Population Size: 50000
Likely Sample Proportion: 50%
Sample Size: 30

Notice that all that changed was the margin of error increasing, that resulted in a lower sample number.

I understand that the confidence level modifies the margin of error assumption. Is this just a guess as well?

Is the calculator just giving me the minimum number of samples required to meet that confidence level, assuming the margin of error to be true?


#2

Margin of error
The margin of error is a statistic expressing the amount of random sampling error in a survey’s results. The larger the margin of error, the less confidence one should have that the poll’s reported results are close to the “true” figures; that is, the figures for the whole population. Margin of error is positive whenever a population is incompletely sampled and the outcome measure has positive variance (that is, it varies).

The term “margin of error” is often used in non-survey contexts to indicate observational error in reporting measured quantities.
Wikipedia