Questionnaire Design, Interviewing and Attitude Measurement - Free ebook download as PDF File .pdf) or read book online for free. PDF | Petra Lietz and others published Questionnaire Design in Article (PDF Available) with 1, Reads . and attitude measurement. face interview ( CAPI) situation as this is the highest standard of interview practice in. Questionnaire Design, Interviewing and Attitude Measurement, London, Pinter. Pp £ paperback, £ hardback. ISBN (pb),
|Language:||English, Spanish, Dutch|
|Genre:||Politics & Laws|
|Distribution:||Free* [*Registration Required]|
recommendations about optimal questionnaire design based on the common of the survey, as it was described to the respondent prior to the interview; . The American National Election Study surveys have measured citizens' political point on the rating scale that most closely matches that attitude (see. To give you an overview of questionnaire design, including: . sources of measurement error. Interviewing and Attitude Measurement (New Edition). Get this from a library! Questionnaire design, interviewing, and attitude measurement. [A N Oppenheim] -- Research methods and evaluation in the social.
Each of these points will be further discussed throughout the following sections. Figure 4. It emphasises that writing of the questionnaire proper should not begin before an exploratory research phase has been completed. The first of these is to articulate the questions that research is intended to address.
The second step is to determine the hypotheses around which the questionnaire is to be designed. It is possible for the piloting exercise to be used to make necessary adjustments to administrative aspects of the study.
This would include, for example, an assessment of the length of time an interview actually takes, in comparison to the planned length of the interview; or, in the same way, the time needed to complete questionnaires.
Moreover, checks can be made on the appropriateness of the timing of the study in relation to contemporary events such as avoiding farm visits during busy harvesting periods. Preliminary decisions in questionnaire design There are nine steps involved in the development of a questionnaire: 1. Decide the information required. Define the target respondents. Choose the method s of reaching your target respondents.
Decide on question content. Develop the question wording. Put questions into a meaningful order and format.
Check the length of the questionnaire. Pre-test the questionnaire. Develop the final survey form. Deciding on the information required It should be noted that one does not start by writing questions.
The first step is to decide 'what are the things one needs to know from the respondent in order to meet the survey's objectives?
One may already have an idea about the kind of information to be collected, but additional help can be obtained from secondary data, previous rapid rural appraisals and exploratory research.
In respect of secondary data, the researcher should be aware of what work has been done on the same or similar problems in the past, what factors have not yet been examined, and how the present survey questionnaire can build on what has already been discovered.
Further, a small number of preliminary informal interviews with target respondents will give a glimpse of reality that may help clarify ideas about what information is required. Please direct all enquiries to the publishers. Printed and bound in Great Britain by Biddles Ltd www. This book is not for them. It is intended to help all those who, for one reason or another, have to design a questionnaire.
It could serve as a textbook for research students and undergraduates in social psychology and sociology and for practitioners of G r k e t research.
It could also help doctors, personnel officers, civil servants, criminologists, social anthropologists, teachers, and many others whose curios- ity or need for information may impel them into the quicksands of social research. Questionnaire design cannot be taught from books; every investigation presents new and different problems. A textbook can only hope to prevent some of the worst pitfalls and to give practical, do-it-yourself kind of information that will point the way out of difficulties.
It is not meant to be a work of scholarship, nor is it exhaustive. The number of references has been kept to a minimum, though some annotated readings have been provided with each chapter. Throughout, clarity and basic ease of understanding have been my guidelines; in each case, an attempt has been made to present the main lines of the argument and, while not shirking weaknesses in underlying assumptions, to omit some of the more esoteric controversies.
Preface to the new edition At the suggestion of many colleagues the book has been greatly expanded, as well as revised and updated. There are two new chapters on research design and sampling, two new chapters on interviewing, a chapter on questionnaire planning, a chapter on statistical analysis and a special chapter on pilot work.
Other chapters have been changed and expanded, and a special set of Workshops has been introduced to offer practical experience. From having been a specialized text dealing with the design of questionnaires and attitude measures, it has broadened out and become a general survey research handbook. It still aims to reach those who have little or no background in the social sciences, and those who lack the facilities which many Western academics take for granted, as well as research students starting their first project and the junior staff of social research agencies.
Teachers of courses in survey methods may find the book particularly useful because of the wide variety of examples spread through the text. It is my hope that many will find this not just a book to read but to keep.
In trying to make this text clearer, more accurate and better-written, I have sought the help of colleagues and graduate students.
I owe much to their incisive critique and generosity of spirit and so does the reader. The responsibility for the book's remaining faults must be mine.
Bram Oppenheim I wish to thank a number of people for their kind permission to use material London, January from their works, as follows: A. Baldwin, J. Kalhorn, and F. Himmelweit, A.
Oppenheim, and P. Michael Shepherd, A. Cooper, A. Brown, and G. The Open University. Course DE. You may have jotted down some questions and handed these to forty-two pet owners in your neighbourhood or you may have conducted forty- two interviews.
Assuming that your questions are well formulated and that all your respondents have answered truthfully, would you be able to draw any conclusions from your findings? Probably not, unless by 'conclusions' you mean those which apply only to these particular forty-two people. Such very limited conclusions are likely to be of little interest since it would be quite unwarranted to draw any wider inferences from the responses.
Your little survey would tell you nothing about pet owners in general, and the results obtained in your neighbourhood might be quite misleading not only about pet owners in general but even about the majority of pet owners in your neighbourhood. You might nevertheless be tempted to study your findings more closely, for instance to see if the motives of dog owners differ from those of cat owners, but again the findings would tell you nothing about dog owners versus cat owners in general or even those in your own neighbourhood.
Nor could you test out the hypothesis that a pet is often a child substitute. It would, for example, be quite unwarranted to claim that your hypothesis had been supported because two-thirds of your sample were childless. It might be that more childless people answered your questions because you happened to know them or because they had the time or the inclination to do so.
And besides, just because people are childless and have pets, this does not prove that they own pets because they are childless: associations do not establish causality. As before, it would also be wrong to draw any inferences from your findings which might apply to the childlessness of pet owners in general. Nor is it simply a question of numbers, for the same strictures would apply even if you had questioned many thousands of people. Moreover, there may be many reasons for keeping a pet; we should always avoid thinking in terms of simplistic monocausal models see below and Chapter 2.
In short, the need for an appropriate research design arises whenever we wish to generalize from our findings, either in terms of the frequency or prevalence of particular attributes or variables, or about the relationships between them. However the creation and special group, for example all the women in Dr Z's practice who gave birth to a application of measuring instruments and data collection techniques tend to be baby in year Y, or all the employees who left firm X in a given month.
But even specific to each discipline or group of disciplines. Cross-disciplinary research, in these circumstances research design problems arise. Quite probably, we wish such as doctors conducting social surveys of their patients' smoking habits, to study these particular mothers, or job leavers, because we want to generalize familiarity with measuring techniques in more than one discipline.
But planned architecture of inquiry. In later chapters we shall deal with measure- unless our study is designed appropriately and is aimed at the correct target ment and instrument building, that is with research techniques such as scales and population, we will be unable to draw any such conclusions or comparisons. However, as we have indicated before, these two aspects of There may also be non-response problems see Chapter 7 so that some of the research are often interlinked.
The design of the research will determine whom necessary information will be missing, and the remainder of the responses may we should question and what questions we should ask, while technical and be biased. These matters will present us with further research design problems.
The need for good design becomes even more acute when we wish to undertake a more ambitious study. Suppose, for example, that we are asked to evaluate some social intervention such as the use of road paints to warn First steps in survey design motorists to slow down before an intersection or the institution of a telephone helpline for children who are being abused.
We might be asked to say, on the Too often, surveys are carried out on the basis of insufficient design and basis of survey research, whether such interventions are achieving their goall planning or on the basis of no design at all.
How would we tempting activity to which a questionnaire opens a quick and seemingly easy go about this? How would we try to make sure that our conclusions are valid and avenue; the weaknesses in the design are frequently not recognized until the could form a sound basis for generalization and further action?
Survey literature abounds with It might be helpful at this point to make a rough-and-ready distinction portentous conclusions based on faulty inferences from insufficient evidence between research design and research techniques, although each influences the misguidedly collected and wrongly assembled. Not everyone realizes that the other to some extent. We often find that, more general conclusions from it. Thus, the research design should tell us how as the research takes shape, our aim undergoes a number of subtle changes as our sample will be drawn, what sub-groups it must contain, what comparisons a consequence of greater clarity in our thinking see Chapter 2.
Such changes will be made, whether or not we shall need control groups, what variables will may require a new and better design, which in turn will lead to a better need to be measured when and at what intervals , and how these measures will specification for the instruments of measurement.
Research design is concerned with making our problem researchable by research process, though the plan may have to be changed later. A social setting up our study in a way that will produce specific answers to specific research study may last from a few months to many years, but most surveys go questions.
Good research design should above all make it possible for us to draw through the same stages or cycles of stages.
Research techniques, on the other hand, are the methods used for data I. Deciding the aims of the study and, possibly, the theories to be investigated. Shall we gather our data by interview, by telephone General aims must then lead to a statement of specific aims, and these or by postal questionnaire? How shall we measure attitudes, downloading should be turned into operationalized aims; that is, a specified set of practical behaviour, social integration, conservatism or friendship patterns?
Can we put issues or hypotheses to be investigated. This should lead directly to a questions together in groups to form inventories and scales see Chapter 9?
How shall we questions, scales and indicators will have to be formulated. Essentially, research techniques are concerned with 2.
Reviewing the relevant literature; discussions with informants and interested measurement, quantification and instrument building and with making sure organizations. Preliminary conceptualization of the study, followed by a series of exploratory T h s distinction between research design and research techniques holds true or 'depth' interviews; revised conceptualization and research objectives see for all scientific disciplines.
Moreover, the principles of research design are below. Deciding the design of the study and assessing its feasibility within the 8 Quesfionnaire Design, Infemiewing and AtfifudeMeasuremenf Infroduction to survey design 9 limitations of time, costs and staffing.
Abandon the study at this point, or statement of the study's objectives, preferably with theoretical underpinnings. Deciding which hypotheses will be investigated. Making these hypotheses question both owners and non-owners? What about ex-owners? And what specific to the situation that is making the hypotheses operational. Listing should we ask them about: their income, the availability of computers at their the variables which will have to be measured.
For instance, if we have some place of work, various aspects of home computer use domestic accounts, word hypotheses about political participation, then how shall we operationalize processing, computer games , the way people make decisions about major this behaviour, and what variables shall we need to measure: party domestic expenditures?
Later, a re-conceptualization might add to this list, for membership, fund-raising activities, going to political rallies, watching example educational use if they have children ; 'keeping up with the JonesesJ political events on TV, displaying a window poster or bumper sticker?
Designing, or adapting, the necessary research insfrumenfs and techniques gender-related ; leisure pursuits; health-related fears; and so on. The research took place in a major tive methods, check lists or rating scales. Doing the necessary pilof work see Chapter 4 to try out the instruments, migrants from distant rural areas. The researchers found that many of the more making revisions where necessary and trying them out again.
Piloting other recent arrivals in the city were suffering from psychosomatic and stress-related aspects of the research such as how to gain access to respondents. They concluded that this was due to the new migrants being Besigning the sample s1. Will the sample need to be representative that is, a adversely affected by the problems of urban living. However, the researchers did probability sample, see Chapter 3 and, if so, of whom?
Please check your email for instructions on resetting your password. If you do not receive an email within 10 minutes, your email address may not be registered, and you may need to create a new Wiley Online Library account. If the address matches an existing account you will receive an email with instructions to retrieve your username.
First published: December Tools Request permission Export citation Add to favorites Track citation. Share Give access Share full text access.
Share full text access.