In order to draw correct conclusion, it is necessary that the sample should be reliable. Unrealiable sample gives incorrect data and the conclusions drawn may not show the features of the population. Reliability of sample depends upon a number of factors.

1. Size : Choice of an appropriate sampling method increases the reliability of data.

2. Sampling errors : Sampling errors reduce the reliability of sample. A sampling error arises due to drawing of inference about the population on the basis of a few observations. There are two types of sampling errors : 1. Biased error, 2. Unbiased error.

3. Non sampling errors : They are more serious than sampling errors because a sampling errors because a sampling error can be minimised by taking a larger sample. It is difficult to minimise non-sampling error, even by taking a large sample. Some of the non-sampling errors are :

(a) Errors in data acquisition

(b) Non response errors

(c) Sampling bias.