Expected number of people who read documents they are asked to sign = E = 50% of 50 = 25
Observed number of people who read documents they are asked to sign = O = 7+15 = 22
Standard error = sq rt (npq)
n = 50
p = 50% = 0.5
q = 1 - p = 1 - 0.5 = 0.5
Standard error = sq rt (50*0.5*0.5)
= 3.54
Difference /S.E = (O-E)/S.E
=(22 - 25)/3.54
= - 3 / 3.54
= - 0.85
Critical value at 5% level of significance = -1.96
Since the difference (-0.85) is greater than the critical value at 5% level of significance (-1.96) the NULL Hypothesis can be rejected. In other words the sample data failed to support the hypothesis that fewer than half of all people read the documents that they are asked to sign.