M/A/R/C®
M/A/R/C Research Statistical Calculator Further Specific Assumptions:

Independent-groups test of 2 percentages:
1. The responses arise from binomial distributions.
2. The two samples are independent of one another.
3. n1*p1 > 5, n1*(1-p1) > 5, n2*p2 > 5, and n2*(1-p2) > 5.

Independent-groups test of 2 means:
1. The responses arise from approximately normal distributions.
2. The two samples are independent of one another.
3. The 2 variances are unknown and assumed equal.

Margin of error and sample size n calculations:
1. The responses arise from a binomial distribution.
2. The estimated proportion = 50%.
3. n*p > 5 and n*(1-p) > 5.



Statistical Calculator


Stat Testing

Is the data weighted?: Yes No
Do the groups overlap (i.e. Total vs Male)?: Yes No
Group 1 Group 2

Unweighted Base:
Weighted Base:
Effective Base:
Testing: Percentages Means

Percentages: % %
Means:
Standard Deviation:
t-value:
p-value: (2 tailed test)
Significant at: % Confidence Level




The left side of the calculator is for statistical testing of two groups on a single variable. The point estimates that are tested can either be percentages or means and can be based either on unweighted or weighted data. The two groups are allowed to overlap.

Reliability for a 50% Proportion

Margin of Error
Base:
Confidence Level: %
Margin of Error +/-: %




Base Size Calculator
Margin of Error +/-: %
Confidence Level: %
Base Size Needed:




The right side of the calculator is for estimating degree of reliability around a 50% estimate for a proportion. It either calculates the margin of error given a base size n or calculates the base size n required to attain a specified margin of error.
Both the statistical test results and the reliability results assume a two-tailed (non-directional) effect of random samples of the population under study.
Click here for further specific assumptions.