Choice "D" is correct. Statistical sampling helps the auditor to measure the sufficiency of the audit evidence because the auditor can quantify the audit risk, thus assisting in limiting it to an acceptable level.
Choice "a" is incorrect. Nonsampling errors relate to the improper evaluation of evidence by the auditor and are not dependent on the sampling method used.
Choice "c" is incorrect. Audit risk can be reduced to a relatively low level with non-statistical sampling. Materiality is not affected by the sampling method used.
Choice "b" is incorrect. The use of statistical sampling does not necessarily minimize the failure to detect errors and fraud.