The Impact of not Carrying out Some Part of Interviews on the Validity of Statistical Inference in Social Research. TelephoneI nterviews and CATI in View of Systematic Errors

• Authors: Jabkowski P.
• Languages of publication: PL

Abstracts

EN

The article concerns the question of inaccessible units set in the perspective of the theory of statistical conclusion. The focus is on consequences of omitting of non-sampling errors resulting from the lack of knowledge of distribution of the opinions of people who were selected but not included in the sample. It has been shown that non-sampling errors resulting from not carrying out of part of the test (if they are not restricted efficiently) disturb the results to such an extent that drawing any reasonable conclusions becomes practically impossible. The article consists of three parts connected with each other. In the first the problem of not carrying out of a part of the test have been formulated and statistical theoretical bases for further arguments have been given. In the second part procedures of evaluation of non-sampling assessment errors resulting from not carrying out of assumed test at the beginning of the test have been derived. In the third part a critical analysis (in the light of the problems discussed) of two research techniques - telephone surveys and questionary surveys - has been provided.

Keywords

EN

NON-RESPONSES; NON-SAMPLING ERRORS.; REPRESENTATIVENESS; SAMPLING ERRORS

Discipline

• SOCIOLOGY: SOCIOLOGY

• Publisher: Institute of Philosophy and Sociology, Polish Academy of Sciences
• Journal: ASK. Research and Methods
• Year: 2007
• Volume: 16
• Pages: 67-86
• Document type: ARTICLE

Contributors

author

Jabkowski P.

• P. Jabkowski, Uniwersytet im. Adama Mickiewicza, Instytut Socjologii, ul. Szamarzewskiego 89c, 60-568 Poznan, Poland

Identifiers

CEJSH db identifier

08PLAAAA04508713