1 Nature, Timing and Extent

　　1.1 Nature

　　Substantive analytical procedures are generally more applicable to large volumes of transactions that tend to be predictable over time and are often used in conjunction with a strong control environment and where audit evidence has been obtained from testing the effectiveness of controls (e.g. computer information systems). See Session 16 for further detail on substantive analytical review.

　　Tests of detail on transactions are used to obtain audit evidence regarding the assertions related to transactions (i.e. COCOA—completeness, occurrence, classification, cut-off and accuracy). These will be on manual systems or CISs where it is easy to trace a transaction through the system.

　　Tests of details of transactions trace a transaction through a system, for example to ensure that a despatch is correctly recorded as a sale or that a purchase entry recorded in the daybook is supported by a purchase invoice, goods received note and purchase order.

　　If no reliance is placed on controls, the level (i.e. sample size) of transaction testing will be high. If reliance can be placed on controls, the level of transaction testing will be lower (and usually conducted as a hybrid test (see s.3) alongside the control testing).

　　The auditor may assess that audit risk can be reduced to an acceptable level by reliance on controls and substantive analytical review, rather than transaction testing (as above).

　　Tests of detail on balances are used to obtain evidence about the assertions related to account balances (i.e. CARE— completeness, accuracy and valuation, rights and obligations and existence).

　　1.2 Timing

　　Substantive transaction procedures may be carried out at an earlier date than the entity's year-end and the final audit (i.e. at an interim audit). If carried out at an interim date, further tests must be carried out to cover the remaining period. These tests (substantive and/or tests of control) must be sufficient to ensure that the risk of misstatement does not increase during this period.

　　Unlike tests of control, where prior-year audit evidence may be relied on under certain circumstances, prior-year substantive evidence will be insufficient to address a risk of material misstatement in the current period.

　　1.3 Extent

　　Generally, the greater the risk of material misstatement, the greater the extent of substantive procedures.

　　For any one substantive procedure, the extent of testing usually relates to sample sizes (e.g. increasing the extent means increasing the sample size).

　　However, the extent of substantive procedures may also be considered in terms of: selecting large (e.g. material) or unusual items from a population; or stratifying the population into homogeneous subpopulations for sampling.

　　It is not unusual in many substantive testing approaches that all items greater than the materiality level (taking into account performance materiality) are selected for testing. If an error is found, that error is likely to be material to the financial statement assertions.

　　2. Hybrid Approach

　　In some cases, a single test approach can be used as a test of control and a substantive procedure at the same time. This is called a dual-purpose or hybrid test.