After finding some facts it is investigated that. Non-response data is preplanned to get data or information from a fixed number of the sample units Including in the case of the census and also in all the population. The accuracy that is obtained by this method depends on much more upon the prefixed number of the sample unit. After the investigation, we can found information in social science that we can get information from each and every one of the sample units and the accuracy of the sample design is maintained without the loss of any information.
But the investigation that is found in social science is that it is hardly lucky to be able to observe/interview each and every sample units. But in social investigation , the sample units are usually human beings and as such, they are not controllable by the investigation. Information from many of them even though they are selected as the sample element, could not be elected because of numerous reason. Therefore, complete information desired by the investigation could not be collected. Therefore the loss of the information is regarded called as non-response.
The non-response causes bias in the estimation of the characteristics of the study of population.
It is due to the following reason.
Non-respondent occur mainly due to.
Types of non-respondents.
There are different types of non-respondents. Kahn and Cannel (1957) identified five principal categories of non-respondents.
They are listed as the following points.
Control of non-response.
It can be decreased by the following methods or points.
The export factor technique generally moves from present to present future. It is proposed to study the influence of boarding as well as non-boarding school education upon future adjustability of the student we took as an equal number of students in two groups one groups consisting of one type of education and other consisting of another type of education. The groups are selected keeping in mind that other variables like age, economic, status, intelligence are matched in both.
After ten years (Say) we shall again try to locate those persons and see how they are placed in life. The difference between the two groups may be attributed to the nature of education received.
Random response technique.
Since most of the respondents hesitate to report their private affairs directly, a statical method was developed to estimate the proportions of respondents having a particular private activity without disclosing them. The technique is known as random response technique and was developed by Warner (1955) and improve by Abul-Ela and other (1967). In this technique, respondents choose to answer one of the two statement of which one is sensitive and related to the scope of the study and the other is the non-sensitive question and not related to the fact under study.
The choice of the statement to answer is made by means of a random device provided by the investigator who knows the probability of selection of each statement. The respondent uses a random device to choose one of the statement and gives answer ‘Yes or ‘No to the selected statement without disclosing to the interviewers which statement he/she is answering. This way the respondent maintain his her privacy.
The technique is use to the following.
Where p=Probability of choosing the sensitive question.
1-p=Probability of choosing another statement.
λ=propertional of the respondents saying yes to either of the two unrelated statement (One the and other non-sensitive).
λ1=Propertion of respondents saying yes to the sensitive question.
λ2=Probability of the event in the other question.
The investigator calculates P,λ andλ2 from the statement given in the problem and from the number of Yes responses. Then using equation (1), he calculatesλ1
I was pregnant at some time during the past 12 months and had an abortion which terminated the pregnancy …..a sensitive statement.
I was born in the month of April.
Not related to the first question and non-sensitive.
In an investigation about the abortion rate, 300 respondents are asked to reply one of above statement either in ‘yes’ or ‘no’ without disclosing which of the statement they are replying, to determine which statement to answer, following procedure is adopted.
25 red balls and 15 white are thoroughly mixed in an urn and then the respondents is asked to pick one ball at random.If the selected ball is red, he has to answer statement A, otherwise, he has to answer statement B. The ball selected is not shown to the investigator, So the secrecy of the statement answered by the respondents is kept intact. Suppose altogether 15 ‘yes’ respondents are found ( Which include ‘yes’ for either of the statement.
λ=Total number of yes respondent/Total number of respondent
The total number of ball related to sensitive statement=25
Kerlinger, F.N. Foundation of Behavioural Research. New Delhi: Surjeet Publication, 2000.
Kothari, C.R. Research Methodology. India: Vishwa Prakashan, 1990.
Singh, M.L. and J.M Singh. Understanding Research Methodology. 1998.
Singh, Mrigendra Lal. Understanding Research Methodology. Nepal: National Book centre, 2013.