American Association for Public Opinion Research

A RejoinderAuthor(s): Robert WilliamsSource: The Public Opinion Quarterly, Vol. 14, No. 3 (Autumn, 1950), p. 611Published by: Oxford University Press on behalf of the American Association for Public OpinionResearchStable URL: http://www.jstor.org/stable/2746031 .

Accessed: 05/12/2014 06:50

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

American Association for Public Opinion Research and Oxford University Press are collaborating with JSTORto digitize, preserve and extend access to The Public Opinion Quarterly.

http://www.jstor.org

This content downloaded from 128.235.251.161 on Fri, 5 Dec 2014 06:50:20 AMAll use subject to JSTOR Terms and Conditions

FROM POQ READERS FROM POQ READERS FROM POQ READERS

be used in market surveys? Can we solve the problem by running from it, by ignoring it, or by reducing survey standards? Mr. Williams' case history raises this question by implication. The student of marketing may be left with the impression that the difficulties of substitute respondents spring from prob- ability sampling. This would be an un- fortunate conclusion to draw from this excellent case study and, undoubtedly, was not intended by the author.

The purpose of this note is to em- phasize that the problem of the uncoop-

be used in market surveys? Can we solve the problem by running from it, by ignoring it, or by reducing survey standards? Mr. Williams' case history raises this question by implication. The student of marketing may be left with the impression that the difficulties of substitute respondents spring from prob- ability sampling. This would be an un- fortunate conclusion to draw from this excellent case study and, undoubtedly, was not intended by the author.

The purpose of this note is to em- phasize that the problem of the uncoop-

be used in market surveys? Can we solve the problem by running from it, by ignoring it, or by reducing survey standards? Mr. Williams' case history raises this question by implication. The student of marketing may be left with the impression that the difficulties of substitute respondents spring from prob- ability sampling. This would be an un- fortunate conclusion to draw from this excellent case study and, undoubtedly, was not intended by the author.

The purpose of this note is to em- phasize that the problem of the uncoop-

A REJOINDER

To THE EDITOR:

It would be unfortunate if the paper under discussion left the implication that the problem of the unavailable re- spondent might be solved by running from it, ignoring it, or by reducing sur- vey standards. On the contrary, the pa- per sought to shed some light on those factors which have to be considered in deciding what is to be done towards a solution.

The problem of the unavailable re- spondent is present in all survey work. A quota sample falls short in dealing with it in other ways. Both approaches have limitations; neither has the com- plete answer.

The paper under discussion deals with a further set of problems raised by the probability sample. How far does one

A REJOINDER

To THE EDITOR:

It would be unfortunate if the paper under discussion left the implication that the problem of the unavailable re- spondent might be solved by running from it, ignoring it, or by reducing sur- vey standards. On the contrary, the pa- per sought to shed some light on those factors which have to be considered in deciding what is to be done towards a solution.

The problem of the unavailable re- spondent is present in all survey work. A quota sample falls short in dealing with it in other ways. Both approaches have limitations; neither has the com- plete answer.

The paper under discussion deals with a further set of problems raised by the probability sample. How far does one

A REJOINDER

To THE EDITOR:

It would be unfortunate if the paper under discussion left the implication that the problem of the unavailable re- spondent might be solved by running from it, ignoring it, or by reducing sur- vey standards. On the contrary, the pa- per sought to shed some light on those factors which have to be considered in deciding what is to be done towards a solution.

The problem of the unavailable re- spondent is present in all survey work. A quota sample falls short in dealing with it in other ways. Both approaches have limitations; neither has the com- plete answer.

The paper under discussion deals with a further set of problems raised by the probability sample. How far does one

erative and unavailable respondent is primarily an interviewing problem. The survey plan can only point out which persons should be interviewed-it is up to the interviewer to follow that plan as closely as possible. A plan such as is used in probability sampling which sheds light upon interviewing failures and thus points the way to more accurate research should be encouraged.

ALFRED N. WATSON Assistant Treasurer The Curtis Publishing Company

erative and unavailable respondent is primarily an interviewing problem. The survey plan can only point out which persons should be interviewed-it is up to the interviewer to follow that plan as closely as possible. A plan such as is used in probability sampling which sheds light upon interviewing failures and thus points the way to more accurate research should be encouraged.

ALFRED N. WATSON Assistant Treasurer The Curtis Publishing Company

erative and unavailable respondent is primarily an interviewing problem. The survey plan can only point out which persons should be interviewed-it is up to the interviewer to follow that plan as closely as possible. A plan such as is used in probability sampling which sheds light upon interviewing failures and thus points the way to more accurate research should be encouraged.

ALFRED N. WATSON Assistant Treasurer The Curtis Publishing Company

need to go towards the goal of theoreti- cal perfection? How far may one de- part from an accepted theoretical design and still retain the advantages of a prob- ability sample?

The practical necessity for dealing with this set of problems arises from the considerations of quality and cost. Without reference to costs one could al- ways try to do a perfect probability sam- ple. Without reference to quality one could depart considerably from an ac- cepted theoretical design. Since, how- ever, we cannot dismiss these considera- tions we need data to help us decide what modifications may and may not be made. The paper is intended as a contribution towards these decisions.

ROBERT WILLIAMS New York City

need to go towards the goal of theoreti- cal perfection? How far may one de- part from an accepted theoretical design and still retain the advantages of a prob- ability sample?

The practical necessity for dealing with this set of problems arises from the considerations of quality and cost. Without reference to costs one could al- ways try to do a perfect probability sam- ple. Without reference to quality one could depart considerably from an ac- cepted theoretical design. Since, how- ever, we cannot dismiss these considera- tions we need data to help us decide what modifications may and may not be made. The paper is intended as a contribution towards these decisions.

ROBERT WILLIAMS New York City

need to go towards the goal of theoreti- cal perfection? How far may one de- part from an accepted theoretical design and still retain the advantages of a prob- ability sample?

The practical necessity for dealing with this set of problems arises from the considerations of quality and cost. Without reference to costs one could al- ways try to do a perfect probability sam- ple. Without reference to quality one could depart considerably from an ac- cepted theoretical design. Since, how- ever, we cannot dismiss these considera- tions we need data to help us decide what modifications may and may not be made. The paper is intended as a contribution towards these decisions.

ROBERT WILLIAMS New York City

PROSPERITY AND POLITICAL VICTORY PROSPERITY AND POLITICAL VICTORY PROSPERITY AND POLITICAL VICTORY

ToTHE EDITOR:

In the Summer i950 issue of this journal, Wilkinson and Hart give the results of a study undertaken "with a

ToTHE EDITOR:

In the Summer i950 issue of this journal, Wilkinson and Hart give the results of a study undertaken "with a

ToTHE EDITOR:

In the Summer i950 issue of this journal, Wilkinson and Hart give the results of a study undertaken "with a

view to demonstrating the relationship between prosperity and political vic- tory." Having sought persistently for valid proofs and found none, they are

view to demonstrating the relationship between prosperity and political vic- tory." Having sought persistently for valid proofs and found none, they are

view to demonstrating the relationship between prosperity and political vic- tory." Having sought persistently for valid proofs and found none, they are

6II 6II 6II

This content downloaded from 128.235.251.161 on Fri, 5 Dec 2014 06:50:20 AMAll use subject to JSTOR Terms and Conditions