STRUCTURED METHODS: INTERVIEWS, QUESTIONNAIRES AND OBSERVATION
11STRUCTURED METHODS:INTERVIEWS, QUESTIONNAIRESAND OBSERVATIONConstantinos N. Phellas, Alice Bloch andClive SealeChapter ContentsInterviews or self-completion questionnaires?182Types of interview183Face-to-face interviews183Telephone interviews183Self-completed questionnaires184Designing studies using structured interviews and questionnaires185Determining the information to be sought185Deciding how to administer the questionnaire or interview186Postal surveys186Internet-based methods188Constructing an interview schedule or questionnaire192Keep it short192Introduction or welcome message192Elements of an effective cover letter193Deciding the order of questions193Include all potential answer choices19311-Seale-4312-CH-11-Part 2.indd 18122/11/2011 4:03:25 PM
Questionnaire layout194Question types195Levels of measurement196Piloting the instrument197Structured observation198The Roter Interaction Analysis System (RIAS)198Flanders Interaction Analysis Categories (FIAC)200Deciding to use structured observation instruments201Conclusion202Learning how to design and use structuredinterviews, questionnaires and observationinstruments is an important skill for researchers. Such survey instruments can be used inmany types of research, from case study, tocross-sectional survey, to experiment. A studyof this sort can involve anything from a shortpaper-and-pencil feedback form, to an intensiveone-to-one interview asking a large number ofquestions, to direct observation of relevantbehaviour. In general, these data collectioninstruments fall into three broad categories:self-completed questionnaires, interviews andobservation schedules. This chapter concerns allof these, explaining how to design and administer structured interview schedules, design anddistribute questionnaires intended for selfcompletion by respondents, and carry out structured observations.Interviews or self-completionquestionnaires?Choosing between an interview and a selfcompleted questionnaire on which the respondent writes their answers is an important decision.Within these there are also choices to be made,each with advantages or disadvantages. Thus,18211-Seale-4312-CH-11-Part 2.indd 182interviews can be done face to face or by telephone. A questionnaire can be sent and returnedby post or email, completed on the Web, orhanded directly to the respondent who completesit on the spot and hands it back. Additionally,some interviews contain pauses for respondentsto complete questionnaire sections, so that theresulting instrument is a combination of things.This can be particularly advantageous if a topicis felt to be socially embarrassing to discuss faceto face and has been used, for example, in surveys of sexual behaviour.Interviews have certain advantages overself-completion questionnaires. The interviewercan explain questions that the respondent hasnot understood and can ask for further elaboration of replies (e.g. ‘Why do you say that?’). Ingeneral, being asked questions by a sympatheticlistener is experienced as more rewarding byrespondents than the chore of filling in a formfor some anonymous researcher, so it is generally found that fewer people refuse to take partand more questions can be asked of each person. However, interviews are more time consuming for the researcher and it may be thecase that interviewer bias, where the interviewer influences the replies by revealing theirown opinions, can be avoided by self-completionquestionnaires.DOING RESEARCH22/11/2011 4:03:25 PM
Self-completion questionnaires have theadvantage of being cheap, but are more suited toissues where there are only a few questions thatare relatively clear and simple in their meaning,and the choice of replies can be limited to fixedcategories. They are especially useful in surveying people who are dispersed over a wide geographical area, where the travelling demands onan interviewer would be excessive.vThe presence of an interviewer allows for complexquestions to be explained, if necessary, to theinterviewee.vInterviews can generally be longer than whenself-completion techniques are used as intervieweesare less likely to be put off by the length or to give uphalfway through.vThere is more scope to ask open questions sincerespondents do not have to write in their answer andthe interviewer can pick up on non-verbal clues thatindicate what is relevant to the interviewees and howthey are responding to different questions.Types of interviewThe interview is a more flexible form than thequestionnaire and, if intelligently used, can generally be used to gather information of greaterdepth and can be more sensitive to contextualvariations in meaning. The classical surveyresearch tradition, geared to producing quantitative data, is generally associated with interviewswhere the wording and order of questions areexactly the same for every respondent. Variationin responses can thus be attributed to respondents and not to variability in the interviewingtechnique. Wording the questions in the sameway for each respondent is sometimes calledstandardising. Asking the questions in the sameorder is called scheduling.Interviews, however, can be non-scheduled,though still partly standardised. This is sometimes called a semi-structured interview. Here,the interviewer works from a list of topics thatneed to be covered with each respondent, butthe order and exact wording of questions is notimportant. Generally, such interviews gatherqualitative data, although this can be coded intocategories to be made amenable to statisticalanalysis.vVisual aids can also be used in the face-to-facesituation.vThe interviewer can control the context and theenvironment in which the interview takes place. Forinstance, the interviewer can make sure that thequestions are asked and therefore answered in thecorrect order and that the interview takes place in anappropriate setting which is conducive to accurateresponses.There are however, some problems with face-toface approaches:vThe cost associated with face-to-face interviews canlimit the size and geographical coverage of thesurvey.vInterviewers can introduce bias, which will affect thereliability of responses. Such bias might emerge fromthe way in which questions are asked, or in thepersonal characteristics of the interviewer, or inrespondents’ wish to give socially desirable responses.For instance, there tends to be an over-reporting ofvoting activity and of participation in voluntary activities in data gathered through interviews.Telephone interviewsFace-to-face interviewsUsing face-to-face interviews as a means of datacollection has a number of advantages and disadvantages. The main benefits are:Telephone interviews using interview schedulesare becoming increasingly efficient with developments in computer technology. Computerassisted telephone interviewing (CATI) systems are available and these provide clearSTRUCTURED METHODS11-Seale-4312-CH-11-Part 2.indd 18318322/11/2011 4:03:26 PM
instructions for the interviewer, display theinterview schedule and allow electronic recording of responses as they are given. This cuts outthe data entry part of survey research (i.e.transferring the responses from the interviewschedule to the computer) because responsesare recorded directly onto the computer. Thismakes CATI quick and cheap to use. There areother advantages associated with telephoneinterviews:vBecause the researcher does not have to travel,interviews can take place over a wider geographicalarea.vThere are fewer interviewer effects – that is, thepersonal characteristics of the researcher will be lessobvious than in face-to face situations and is therefore less intrusive.vThe physical safety of the interviewer is not an issue.vTelephone interviews are subject to greater levels ofmonitoring because supervisors can unobtrusivelylisten in to interviews to ensure that they are carriedout correctly.decide whether to use postal, mailed, web-basedor email questionnaires. First though, the goodand bad points of such questionnaires can besummarised. With surveys delivered by thesemeans, questions need to be simple and easy tounderstand and the questionnaire has to beclear and easy to complete because no interviewer is available to assist the respondent.Such surveys can be especially useful whenrespondents need time to gather information orconsider their answers. For example, a survey ofpay levels among university employees by gender would require complex information, so aself-completion survey would provide respondents with time to check their records beforeanswering.Surveys using self-completion questionnaireshave some distinct advantages over face-to-faceinterviews:vThey are cheap to administer. The only costs arethose associated with printing or designing thequestionnaires, their postage or electronic distribution.vThey allow for a greater geographical coverage thanface-to-face interviews without incurring theadditional costs of time and travel. Thus they areparticularly useful when carrying out research withgeographically dispersed populations.vUsing self-completion questionnaires reduces biasingerror caused by the characteristics of the interviewerand the variability in interviewers’ skills.vThe absence of an interviewer provides greateranonymity for the respondent. When the topic of theresearch is sensitive or personal it can increase thereliability of responses.But telephone interviewing has disadvantages too:vvQuestions have to be simple and interviews need tobe kept short because they tend to have higherbreak-off rates (where people refuse to continue)than face-to-face interviews.It can be difficult to ask sensitive questions on thetelephone.vThere is no opportunity to use visual aids or topick up so easily on the non-verbal responses ofinterviewees.vThere are some groups that are underrepresented intelephone surveys. These include people withoutphones (often due to poverty), older people andpeople who are disabled or sick.The main disadvantages of self-completion surveys are:vSelf-completed questionnairesThere are different types of self-completedquestionnaire, and this chapter will help you18411-Seale-4312-CH-11-Part 2.indd 184Questionnaires have to be short and the questionsmust be simple as there is no opportunity to probe orclarify misunderstandings.vThere is no control over who fills out the questionnaire, and the researcher can never be sure that theright person has completed the questionnaire.DOING RESEARCH22/11/2011 4:03:26 PM
vThose with low levels of literacy or poor access toemail or the Internet are unlikely to complete aquestionnaire, meaning that they are excluded fromthe study.vResponse rates tend to be low and it is difficult toknow the characteristics of those who have not filledin the survey and how their non-response will affectthe findings.Response rates in self-completion surveys tendto be maximised when respondents have aninterest in the subject of the research and aretherefore motivated to complete the questionnaire. In addition, response rates can be increasedby sending out reminder letters and emails andfollow-up postings of the questionnaire, thoughthis does mean that the fieldwork element ofsuch surveys can be lengthy.Ways of encouraging a good response rate arealso discussed later in this chapter (and werementioned in Chapter 9 too). In addition, theappearance and layout of questionnaires areimportant, and this chapter will cover this, aswell as discussing different question types andthe pre-testing of questionnaires.Designing studies using structuredinterviews and questionnairesThe most important goal of a study using such aninstrument is to learn about the ideas, knowledge,feelings, opinions/attitudes and self-reportedbehaviours of a defined population. To carry outa survey the researcher must:1determine the information to be sought2define the population to be studied3construct the interview schedule or questionnaireand decide how it is to be administered4draw a representative sample5administer the instrument6analyse and interpret the data7communicate the results.These procedures are overlapping and eachdemands careful work. We will focus in thischapter on steps 1 and 3 in particular. Othersteps are more fully discussed in other parts ofthis book.Determining the information to be soughtSocial research begins with an idea that sometimes might be quite vague and unclear. As aresearcher you must systematically develop andrefine your initial ideas, usually starting with agood understanding of the related literature (seeChapter 6). There will eventually be a need forconcepts in the literature – if they are to beinvestigated in the study you are going to do – tobe operationalised as questionnaire items, so thatclear concept–indicator links are established.Therefore, you must make clear what you wantto find out about. The research questions of theproject determine who you will survey and whatyou will ask them. If your research questions areunclear, the results will probably be unclear. Themore precise you can make these, the easier itwill be to get usable answers.Let us imagine that we are about to carry outa survey in order to answer the followingresearch questions:1Does the possession of a university degree enhancethe job prospects to a different extent in differentethnic groups?2Are people without degrees more likely to have jobsin which they experience alienation?3How do women and men graduates compare inbalancing the demands of home and work?If you examine these three questions you willsee that they contain a number of concepts.These are possession of a degree qualification,ethnic group, having a job, alienation, gender andthe demands of home and work. In designingquestions, a researcher should ensure that theconcepts contained within the aims of the studyare comprehensively covered. If one forgot to askSTRUCTURED METHODS11-Seale-4312-CH-11-Part 2.indd 18518522/11/2011 4:03:26 PM
a question about whether people had a degreequalification, for example, it would not be possible to fulfil the aims of the study.The questions chosen for inclusion in an interview schedule or a self-completed questionnairecan be understood as indicating the conceptscontained in the research questions. Ensuringgood links between concepts and their indicatorslies at the heart of good question design. Someconcepts are easier to indicate than others. Theconcept of sex or gender, for example, is in mostcases not controversial and might, in an interview, be indicated by the interviewer recordingtheir impression rather than asking a questionabout it. The concept of having a degree qualification might also be indicated fairly easily, byasking a person to list their educational qualifications. Whether a person has a job, however,might pose more problems. What does one doabout part-time workers, for example? Do wecount housework as a ‘job’? Decisions abouthow to categorise people into ethnic groups areoften controversial.Additionally, many of the more interestingconcepts in social research are multidimensionalconcepts, which is to say that they are made upfrom several different things. Alienation is anexample. Finding questions to indicate the extentof a person’s alienation requires some furtherconceptual work, and perhaps some reading tosee how different authors have used the term. Aresearcher interested in finding indicators for thisconcept would need to subdivide it into severalcomponents. Alienation involves, amongst otherthings, a sense of powerlessness, of normlessness(being outside normal society), isolation andself-estrangement (seeing a part of oneself as if itwere a stranger). It is easy, for example, to seehow one could be powerless without being isolated, so in order to count as ‘truly’ alienated aperson would need to indicate that they experienced all of its components, requiring questionsindicating each of the dimensions of alienation.The chapter will return to how questionsin survey instruments can be designed so thatthey reflect good concept–indicator links. First,18611-Seale-4312-CH-11-Part 2.indd 186though, we will consider the decision as to howto administer a questionnaire or an interview.Deciding how to administer thequestionnaire or interviewWe saw earlier that there were several ways toadminister a self-completed questionnaire, thesebeing to send and return the questionnaire bypost, and internet-based methods (web oremailed questionnaires). The advantages anddisadvantages of each were summarised. Lessoften used, but nevertheless distinct from theseapproaches, are the group administered surveyand the household drop-off survey. In addition,we saw that there were two main ways to carryout interviews: face to face or by telephone.Either of these might involve computer assistance at the data collection stage, with the interviewer entering responses and being promptedto ask questions as the interview proceeds,though this is more commonly used in telephoneinterviewing. The best approach will always bebased upon a combination of factors such astime, the complexity of the data collectioninstrument, the sample profile and budget.Postal surveysPostal surveys (sometimes called mail-out surveys) usually involve mailing self-completedquestionnaires to a target group of people. Themain advantages of postal surveys are that largenumbers of questionnaires can be sent out atfairly low cost. Questions that are difficult to askon the telephone or in face-to-face interviews canbe asked in a postal questionnaire. For example,personally sensitive information (about income,sexual orientation, drinking behaviour) are bestasked about in a way that saves the respondentthe embarrassment of facing a stranger and reporting something they may feel awkward about.Box 11.1 gives an example of a study that askedabout illegal behaviour in this way.DOING RESEARCH22/11/2011 4:03:26 PM
BOX 11.1EXAMPLE OF HOW TO ASK PERSONALLY SENSITIVEINFORMATIONA postal survey of UK doctors reported by Seale (2009a) asked them to report on whether they had takenvarious decisions about the end-of-life care of their last patient who had died. These decisions includedthings like withdrawing or withholding treatment, considered to unnecessarily extend life when a patientwas already suffering a great deal. Doctors were also asked if they had prescribed or administered a drugwith the sole intention of ending a patient’s life (known as ‘assisted dying’, ‘euthanasia’ or ‘physician-assistedsuicide’). This last type of decision was not legal in the UK at the time of the survey. Doctors were sent apostcard when they were sent the questionnaire, and they were told that they could return the postcardseparately to say that they had replied to the survey, so that they would not receive follow-up remindersto reply. The questionnaire itself contained no information that could link the particular questionnaire tothe identity of any one of the 10,000 doctors who received it. This reassured respondents who reportedillegal action that they could not be identified.A serious problem with postal surveys is thatresponse rates are usually lower than interviewsurveys. This is largely because people find talking to someone more pleasant than filling in aform on their own.Factors that affect response rates are the questionnaire’s length, the way it is laid out (e.g. is iteasy to answer?), whether the issue it enquiresabout is important to the respondent, andwhether incentives are offered. In addition, inpopulations of lower educational and literacylevels, response rates are lower. This makes it difficult, for example, to use postal surveys withgroups that may be particularly important tounderstand, such as immigrant populations, orsocially deprived people.A low response rate is a problem becauseresponders may not be representative of theentire population if they are systematically different on some dimension from non-responders.With self-completed questionnaires, as with anysurvey, you need to look at the characteristics ofth
182 DOING RESEARCH Learning how to design and use structured interviews, questionnaires and observation instruments is an important skill for research-ers. Such survey instruments can be used in
Structured interviews, also known as behavioral interviews, situational interviews, rated interviews, ranked interviews, and/or competency interviews, offer a clearly superior method of hiring. These interviews are twice as reliable as their unstructured, instinct-based counterparts. This is because structured interviews can control for
In the beginning, electronic questionnaires should be relatively concise (not more than 50 questions, preferably closer to 30 for a 3. rd. or 4. th. degree charge). All questionnaires must include standard questions about juror qualifications. Questionnaires must not include questions that seek a juror's individual demographic information.
viewer uses to determine the progression of the interview according to their pre-existing ideas; the research questions; and the resulting questions in the interview schedule. Interviews are usually differentiated as structured, semi-structured or unstructured. Structured interviews will hav
Structured interviews 101 4 So, in structured interviews, hiring managers ask a set of predetermined questions in a specific order. These questions are carefully created and linked to job-relevant criteria based on job analysis.
structured interviews, using data extracts from her interviews to illustrate this. Consideration 1: Is thematic analysis useful for me? Thematic analysis is a hugely popular analytic method. Its popularity partly reflects its independence from any particular theoretical approa
Candidate Experience The Wave questionnaires use work-relevant language to ask candidates what they enjoy doing (motive) and what they think they are good at (talent). The questionnaires use a unique 'rate and rank' format eliciting more detailed information about a candidate's workplace preferences. This avoids central tendency in
naires. When it comes to designing Web-based questionnaires, however, sim-plicity is usually best. Design Principles for Web-Based Questionnaires Dillman (2000) identified a set of design principles that can be applied to Web-based questionnaires. These principles were an extension of
Astrology: The alignment of the planets and stars was very important, looking at when the patient was born and fell ill to decide what was wrong with them! This became more popular after the Black Death (1348) Astrology is a SUPERNATURAL explanation for disease. Apothecaries mixed ingredients to make ointments and medicines for the physicians. They learned from other apothecaries. They also .
