Data Sampling

ACCA F2 - Management Accounting

Sampling determines characteristics of (whole lot) population. A small portion chosen from the population for studying its properties is called a sample. Number of units  in the sample is known as sample size. Sample stands for best possible estimate of population parameter. It helps generate information about characteristics of population with less use of time and resources.

Methods of sampling depend very much on the objective and scope of enquiry, nature of population, size of sample, available resource etc. Sampling methods can be classified broadly into two types.

Random sampling (Probability sampling) is the method of drawing a portion (sample) of population so all possible samples of fixed size have same probability of being selected.

Types of Random sampling:

Ø  Simple random sampling: Sample units are selected in such a way that each and every unit of population has an equal chance of being selected. Sample units are selected with replacement or without replacement. E.g. 10 numbers selected from 0-1000

Ø  Stratified random sampling: Where population is heterogeneous with respect to variables or characteristics under the study, the population is divided into various groups or strata based on certain characteristics so that various strata are not overlapping. Then simple random technique is applied. E.g. 10 letters selected 5 from Capital letters and 5 from Small Letters

Ø  Systematic random sampling: Only first sample is selected at random and remaining units are automatically selected at fixed equal intervals from one another. E.g. (59,159,259……,959) 10 numbers selected from 0-1000 where 59 is first random selection.

Ø  Cluster sampling: Population is divided into groups, called cluster in such a way that the characteristics with the cluster are heterogeneous and between the cluster are homogenous so that number of sampling units in each cluster should be approximately same and then a sample random sampling of clusters is selected. E.g 10 sets of letters (Caps and Small) are selected from ((A,a),(B,b)……..(Z,z))

Non-random sampling (non-probability sampling) depends on judgement or convenience of the investigator. Experienced and skilled investigator applies this method to yield valuable result.

Types of Non-random sampling

Ø  Judgement sampling or Purposive sampling: Items in sample depends entirely upon the judgement/discretion of investigator.

Ø  Convenience sampling or Chunk sampling: In this a small collection of  data (one register out of available) is selected form whole lot.

Ø  Quota sampling: A sampling method in which researchers are given quotas to be filled from the different strata and within the pre-assigned quotas.

Standard Error: Standard deviation of the sampling distribution of sample statistic is known as its standard error of the statistic.

If random sample is drawn with replacement from large population the standard error of sample mean for large population is given by:

Standard Error of mean = Standard Deviation /(sample size 1/2)    ……… sample with replacement