Survey Calculator Definitions

Survey definitions used with our survey calculators:

Cluster Sampling

Gives more accurate results when the group can be easily separated into heterogeneous clusters. Some examples might be:

  • Separating students by classroom.
  • Separating residents by neighborhood.

Confidence Interval

This is the range of possible population values. If the population values range between 10 and 20 percent, then the confidence interval would be "10% to 20%". If the population values range between 100 and 200 people, then the confidence interval would be "100 to 200 people".

Elsewhere, "confidence interval" is often used to denote the single number that this website identifies as the margin of error. To reduce confusion, this website will use "confidence interval" for a range of possible population values, and "margin of error" for the width of a range of possible population values.

Equations used to compute the confidence interval are found
at confidence interval calculations.

Confidence Level

This is the success rate of the calculator's results. In other words, if the confidence level is 95% in the confidence calculator, then 95 times out of 100 the population percentage, if it could be measured, would be within the confidence interval, and 5 times out of 100 the population percentage would be outside the confidence interval.

Convenience Sampling

Ease of sample selection and data collection

Judgment Sampling

Ease of sample selection and data collection

Margin Of Error

This specifies the width of a range of possible population values, where the range of possible population values is called the confidence interval.

If the confidence interval ranges from 10 to 20 percent, then the margin of error is half of the difference, or 5%. If the confidence interval ranges from 100 to 200 people, then the margin of error would be half of the difference, or "50 people".

Elsewhere "confidence interval" is often used to denote the single number that this website identifies as the margin of error. To reduce confusion, this website will use "confidence interval" for a range of possible population values, and "margin of error" for the width of a range of possible population values.

Equations used to compute the margin of error are found at confidence interval calculations.

Population Percentage

This is the percentage of the population that would choose a particular answer. Often this value is unknown, so it cannot be used to compute the sample size before the survey has started. There are two workarounds for this problem.

  1. The first solution, which is used by the Simple Sample Size Calculator, is to simply use a value of 50%. As demonstrated by the Animated Confidence Interval Calculator, that value represents the "worst case" confidence interval. You will not get into trouble by using this first approach.
  2. The second solution is possible when the percentage is known approximately, in which case the approximate value may be entered into the Advanced Sample Size Calculator to compute a more accurate sample size. After a sample has been taken, the sample percentage may be used as the population percentage.

Population Proportion

This is the population percentage represented as a decimal number (divided by 100).

Population Size

This is the number of people in the entire community who have opinions that the survey team would like to measure. Only some of these people will actually be surveyed, and out of those only some may actually respond to a particular question.

Sample Percentage

This is the percentage of the sample group that chose a particular answer.

Sample Proportion

This is the sample percentage represented as a decimal number (divided by 100).

Sample Size

This is the number of people who respond to a particular question. For more accurate survey results, this number should be as close to the
population size as possible. Unfortunately, issues such as time and logistics usually reduce this number.

Equations used to compute the sample size are found at sample size calculations.

Sampling

Selecting a representative subset of the whole population that will be studied in a survey.

Simple Random Sampling

Can be used in any situation, as long as a list of every member is available.

Stratified Sampling

Gives more accurate results when the group can be easily separated into homogeneous strata. Some examples might be:

  • Separating people by age.
  • Separating people by income.

Systematic Sampling

Can be used in any situation, as long as a list of every member is available.