sampling and sample size

SAMPLING DAN SAMPLE SIZE

BADAN PUSAT STATISTIK

Yosep, SST., MIDECKasi Analisis Statistik Lintas SektorEmail: Yosep @bps.go.id

Bandar Lampung, 15 Desember 2012

menurut Guy (1981) ada empat kelebihan sampling statistik;

a. Memungkinkan auditor menghitung reliabilitas sampel dan risiko berdasarkan sampel.

b. Mengharuskan auditor merencanakan sampling dengan lebih baik (more orderly

manner) dibandingkan dengan sampling non statistikc. Auditor bisa mengoptimalkan sampel size, tidak overstated atau understated, dengan risiko yang hendak diterima terukur secara matematis.

d. Berdasarkan sampel, auditor bisa membuat statement yang obyektif mengenai populasi sampel.

Hall et al. (2002) menyebutkan bahwa pengadilan federal di Amerika Serikat sesuai dengan Federal Judicia Center 1994 memutuskan akan menerima bukti sampel, tergantung dari apakah fakta atau data sampel tersebut merupakan “tipe sampel data yang digunakan oleh ahli dalam bidang tertentu untuk membentuk opini atau menarik kesimpulan atas subyek tertentu.”

Dengan demikian bukti sampel yang dihimpun oleh auditor layak dijadikan bukti di pengadilan. Dan ini merupakan tantangan bagi profesi untuk meningkatkan kualitas pengambilan sampel.

Compliance Testing Sample Size Table

Desired Level of Assurance (Remaining Risk of Material

Noncompliance)

Minimum Sample Size

0 exceptions expected

High 60

Moderate 40

Low 25

The following table provides suggested minimum sample sizes associated with high, moderate, and low remaining risk of material noncompliance for populations of 250 items or greater. The remaining risk of material noncompliance is an indicator of the desired level of assurance. A high remaining risk of material noncompliance indicates that a high level of assurance is desired to meet the audit objective.

Small Population Sample Size Table

Frequency and Population Size

Sample Size

Quarterly (4) 2

Monthly (12) 2–4

Semimonthly (24) 3–8

Weekly (52) 5–9

Testing Small Populations

Some significant controls or instances, or both, of compliance do not occur frequently. The following table provides suggested minimum sample sizes in testing small populations subject to controls and compliance requirements.[1] Small populations, for purposes of the Chapter, are defined as populations of fewer than 250 items.

[1] The table is adapted from the table Small Population Sample Size Table in the AICPA Audit Guide Audit Sampling.