Conducting hypothesis tests for a proportion with raw data

This tutorial covers the steps for calculating hypothesis tests a single proportion in StatCrunch. To begin, load the 50 Coin Flips data set, which will be used throughout this tutorial. This data set contains a single column titled Coin flips. Each of the 50 values in this column represents the outcome of a single coin flip and is labeled either Heads or Tails. While this tutorial uses raw data, see Conducting hypothesis tests for a proportion with summary data to compute one-sample proportion results with summary data.

Performing a one-sample proportion hypothesis test

If the coin used is "fair", the proportion of heads the coin will produce over a very long run of flips should be 0.5. Do the 50 outcomes in the short run of flips in this data set suggest the coin is unfair? To conduct the appropriate hypothesis test, choose the Stat > Proportion Stats > One Sample > With Data menu option. Select the Coin Flips column. The Success input is used to define the label of the outcome of interest. In this case, set this value to Heads. Under Perform, the default Hypothesis test for p is selected. Enter 0.5 for the null value of the proportion, and click Compute!. The hypothesis test output includes the observed sample proportion of Heads, the test statistic (labeled as Z-Stat) and the P-value.

Always Learning
Pearson