Survey 

Research: Interviews  
5
and Questionnaires 

Chapter Summary  
Introduction 
Chapter 5 takes a close look at survey research as a process of interviewing. We often think of sur-
veys in the narrower context of self-completed questionnaires or telephone surveys. However,
surveys in research are actually a broader category that refers to structured questioning. It includes
both the questionnaires and the structured interviews. The key task of surveys is to create the list of
questions whose answers can be quantified and easily translated into numbers for coding.

Open or Closed Questions? 
The most significant issue in writing the questionnaire is the decision whether to use open or
closed questions. Open questions leave the response decisions completely to the respondents,
where they can write their response in a free form. By contrast, closed questions present the re-
spondents with a set of answers, and the respondents simply choose an answer category (or cate-
gories) that best represent their answers.

Open Questions
Open questions tend not to be used in quantitative research very often because of difficulties
with converting the answers into numerical data. The main advantages of open questions are the
following:
 Respondents can answer in their own terms.
 They allow for unusual, unanticipated responses.
 They provide no suggestion, so responses may expose knowledge and be more genuine.
 They are good for exploring new or changing areas of research.
 Answers may lead to fixed-choice responses.
The main disadvantages of open questions are the following:
 It is time-consuming to record answers.
 Answers have to be coded, and it’s a tedious and long process
 If used in questionnaires, respondents may balk at the request to write long answers.
 Long verbal answers to questions are more open to recording inaccuracies.
The chief advantage of open questions is giving the respondents an opportunity to answer in
their own terms, while the chief problem with open questions is the difficulty of coding, pro-
cessing, and therefore summarizing these data in the study results.

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
Closed Questions
Closed questions (close-ended, fixed-choice questions) allow the researchers to overcome the
chief disadvantage of the open questions, since closed questions are easier to process, analyze, and
summarize, and respondents’ answers can be compared easily to each other. However, since these
questionnaires offer limited categories from which respondents can choose, closed questions are
more superficial and do not give respondents a chance to elaborate on their answers. The main
advantages of closed questions are the following:
 It is easy to process answers.
 It is possible to compare answers because of standardization.
 Response categories may help the respondent clarify what the question means.
 It is easier and quicker for the respondent to complete.
 Reduces bias in recording answers, because the researcher does not have to interpret the
response.
The main disadvantages of closed questions are the following:
 Loss of spontaneity and authenticity because relevant answers may be excluded from the
choices provided. This issue can be reduced by using open questions to generate the cat-
egories and using a category “other” with an open area to elaborate.
 Categories cannot overlap. A pre-test to establish appropriate and distinct categories may
reduce the likelihood of this occurring.
 Difficult to make forced-choice answers exhaustive. A pre-test to identify appropriate
categories and using a category “other” can be used as a solution.
 Respondents may differ in their interpretation of the wording of fixed responses, e.g., the
meaning of “strongly” in “strongly agree.”
 Respondents may not find a fixed response that they feel applies to them.
 Large numbers of closed questions can reduce rapport in interviews, but may be appreciated
after several open questions.

Types of Questions  
There are several types of questions for structured interviews and questionnaires. Most surveys
will include more than one type of question. The distinction in types should be remembered in
order to clarify what you’re asking, to prevent using a format that is inappropriate for the con-
cept (e.g., a Likert scale for factual question about behaviour), and avoid mixing different types
of questions and thereby reduce measurement validity. Questions can be classified as the follow-
ing types:
 Personal, factual questions: Questions about oneself and recalling one’s activities, e.g., age,
occupation, number times going to a movie, etc.
 Factual questions about others: Questions about others and their activities are problematic,
since we are not particularly informed of other people’s activities or views, and therefore
our knowledge of others may be incorrect. It is therefore advised to avoid asking re-
spondents about the behaviours and views of others. However, respondents can be
asked about their own perceptions of the views and behaviours of others; these questions
are acceptable.
 Factual questions about an entity or event: Respondent is treated as an informant to a phenome-
non or an event as there may be no other reliable source. This may be problematic because
most people are not careful systematic observers.
 Questions about attitudes: These are very common. Quite often a five-point Likert scale is
used to measure attitudes.
 Questions about beliefs: This is the same as questions about attitudes.

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
  Questions about knowledge: These are sometimes used to “test” respondents’ knowledge in
certain areas.

Rules for Designing Questions 
Rules for designing good survey questions include general and specific rules. The three very im-
portant general rules make sure that the questionnaire is actually designed to address its purpose—
that is, collecting the data on a specific research question and actually answering that research
question as the result of the study.

General Rules of Thumb

 Keep the research questions in mind. What are you researching? Avoid asking irrelevant
questions or leaving questions out.
 Focus on exactly what you want to know. Be specific. E.g., If you want to know about a
respondent’s standard of living, you would ask “How many children are living at your
home?” versus “How many children do you have?”
 Put yourself in the position of the respondent. You can expose ambiguous questions by
asking yourself, “How would I answer this question?”

Specific Rules When Designing Questions

Avoid ambiguous terms
“Often” and “regularly” as measures of frequency are too vague. They may be interpreted in very
different ways by different respondents. Fix the answer choice with a provided numerical count
or ask the respondent to estimate the number. Be sure that words used have only one meaning.

Avoid long questions

Respondents can lose track.

Avoid double-barrelled questions

Double-barrelled questions are questions that ask more than one question, but provide only
one set of answers, such as “Did you enjoy the film and the lecture yesterday?” The answer of
the respondent may be positive on one issue, but negative on the other. If you pose a double-
barrelled question, you are not sure which issue the respondent actually answered.

Avoid very general questions when you need specificity

For example, rather than asking about the satisfaction with one’s job, you might want to ask
whether the respondent is satisfied with pay, conditions of work, benefits, and so on. Variance in
responses to a very general question is larger, because the question is open to interpretation. This
leads to data errors and needs to be corrected by reformulating the question.

Avoid leading or loaded questions

Leading questions are the questions subtly prompting the respondent to answer in a particular
way, and so the question “leads” to the answer. Loaded questions do the same by including a
controversial or unjustified assumption in the wording. For example, the question “Don’t you
agree that marijuana should be legalized?” pushes the respondent to answer in the affirmative
and is therefore leading. Pushing respondents in a particular direction skews your data.

Avoid questions that include more than one question

E.g., “Which party did you vote for in the 2011 election?” In this case, you actually asked two
questions: “Did you vote?” and then “For which party?”

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
Avoid negative language
The negative may be missed by the respondent, leading to a false or contrary answer.

Minimize technical terms

Use plain language rather than sociological terminology.

Ensure respondents have the requisite knowledge

There is no point asking somebody about issues about which one has no knowledge, unless the
purpose of the survey or the question is to specifically determine what level of knowledge exists.

Ensure symmetry between a closed question and the language of its answer choices
E.g., avoid: “How satisfied are you with your life in general? Choose one: Very good, good, mid-
dling, poor, very poor.” The choices should be “very satisfied, somewhat satisfied, etc.” The an-
swer choices must be appropriate to the question asked.

Ensure the answers for a closed question are balanced between positive and negative choices.

Choose a reasonable time frame

People may not be able to remember the details of what they have done.

Avoid the “don’t know” option

It is often aimed at preventing the respondent from have to make a choice, but it actually serves
to offer an easy way out. It is best dealt with by using a filter question. E.g., Ask the question,
“Do you have an opinion on this issue?” of every respondent and then ask the second question
only of those who say that they do have an opinion.

Pay close attention to question order

 All respondents should receive questions in the same order (unless testing for order effects).
 Asking a particular question may affect the responses given to subsequent questions.
 Early questions should be directly related to the announced research topic. Thus, questions
about age, social status, etc. (i.e., demographic questions) should not be asked at the be-
ginning of an interview.
 Questions likely to be of interest to respondents should be asked early in the interview.
 “Sensitive” questions (e.g., “Have you ever been a victim of sexual assault?”) should be
asked well into the schedule, but not at the very end.
 Questions should be grouped logically and into related sections.
 Within a group of questions, general questions should precede specific ones. General questions
about an issue preceded by specific questions on that issue are likely to be discounted as
“already answered.”
 Opinion and attitude questions should precede behaviour and knowledge questions.
 Even if a respondent provides an answer to a question before it is asked, the question
should be repeated at the appropriate time to reduce question order effects.

Avoid provoking response sets

These occur when the respondent answers a series of items the same way, suggesting a motive
other than truthfulness for the responses. Inclusion of these data in a research project leads to
measuring errors for concepts. To reduce these problems during face-to-face interviews, inter-
viewers must stay aware of their own verbal and physical cues that might be interpreted as some
kind of judgment, and remain vigilant to avoid them. Respondent data that exhibit characteristics

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
of response sets should be removed, because their answers were more likely to be motivated not
by the desire to answer questions accurately, but by any of the following:
 Acquiescence: Respondents agreeing or disagreeing with a large chunk of questions in a
row, just trying to appear “cooperative” and finish the interview. In order to avoid acqui-
escence, the format of the questions should vary in the questionnaire, so that not too
many questions with the same answer options occur in a row.
 Social desirability: Respondents give answers that make them seem respectable and likeable
rather than just truthful. For example, respondents may hide their illnesses or other char-
acteristics they consider undesirable.
 Laziness/boredom: answers are given just to get interview over with. To avoid this, ensure
the correct answers cannot be answered well with a single response (e.g., no, no, no),
vary the formats, and the answer options for the questions.

Consider using vignette questions when appropriate

These involve presenting people with one or more scenarios and asking them how they would re-
spond. They anchor the choice of answer in a realistic situation, but create distance between ques-
tion and respondent which should hopefully lead to more candid responses. The weakness with
these questions is that people may act quite differently in real life than they say they would. Thus,
other research methods such as experiments, structured observation, historical analysis, and others
may be important to corroborate information obtained in a survey. In spite of these reservations,
the vignette technique can provide useful information, or at least a starting point for further re-
search.

Run a Pilot Study

A pilot study is a test run of the designed survey on a small number of respondents to verify the
performance and efficiency of questions. Generally relies on respondents who will not be part of
the main study, since exposure to the questions and answers in the pilot study would likely influ-
ence the answers in the main study. Where the research population is small, you may have to go
outside of the population being studied to a similar population in order to find suitable substitutes.
A pilot study serves the following purposes:
 May be used to test whether individual items or the instrument as a whole operates well.
Flaws are identified and corrected prior to conducting the main study.
 May be used with open questions to generate the key issues to be measured in subsequent
studies with closed questions.
 Provides interviewers with experience in administering survey instruments; provides
training and builds interviewer confidence.
 Can be used to ensure there is sufficient variability in the answers to provide meaningful data.
 Helps to identify questions that are embarrassing, uninteresting, etc.
 Can identify questions that are difficult to understand.
 Can be used to review the question order and determine whether the main instrument
has an overall flow that is satisfactory.
You may want to use existing questions. If certain questions have worked well for other re-
searchers, they may be appropriate for your own study, but in this case you must cite the original
source. Using existing questions may guide the development of your research ideas and ques-
tions. The questions may be modified to suit your own purposes or used exactly as others have
used them. Re-using the same questions may provide a level of standardization that will allow the
research data to be compared among different samples and studies because the measurement
apparatus is consistent.

   

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
Alternatives to In‐Person Interviews 
Face-to-face interviewing is the preferred method in academic research, while a lot of com-
mercial and government research is done by telephone interviewing. Telephone interviewing
has the following advantages over face-to-face interviewing:
 It is cheaper and quicker to administer, especially where the respondents are geographically
spread out.
 It is easier to supervise and reduce interviewer errors up front.
 Telephone interviewing also reduces bias arising from “interviewer effect,” which results
from the interaction of the interviewer with the interviewee and his or her personal char-
acteristics, which may influence respondents’ answers to certain questions.
At the same time, telephone interviewing has some weaknesses compared to in-person interview-
ing, and they are the following:
 People without telephones (poor), “unlisted numbers,” cell phones, and the hearing impaired
may be excluded from the sample group unless a computer random-digit-dialling program is
used. (Reaching cell phones by this method is prohibited by law in some areas.)
 Telephone interviews are hard to sustain for long periods of time (20–25 minutes).
 Personal interviews tend to be more effective with sensitive issues.
 Telephone interviews cannot gather any additional information about the respondent,
such as his or her social conditions, but this information can be obtained indirectly by in-
terviewer observation during the face-to-face interview, which is usually collected at the
respondent’s place of residence.
 It is difficult to be sure the targeted respondent is the person actually answering the ques-
tionnaire.
 Visual aids cannot be used to assist the interview.
Considering these weaknesses of telephone interviewing, academic research often shows that
data obtained from the face-to-face interviews are far higher quality than the data obtained from
telephone interviews.

Computer‐Assisted Interviewing 
Computer-assisted interviews are the interviews conducted with the help of a computer program.
The use of these interviews is increasing, since they substantially cut down the data processing
time. The two main types of computer-assisted interviewing are computer-assisted telephone
interviewing (CATI) and computer-assisted personal interviewing (CAPI).
 CATI involves the interviewer calling the respondent on the phone and recording his or
her answers directly into the computer.
 CAPI involves face-to-face interviewing where the interviewer immediately inputs the
answers into the personal computer or a device.
Apart from reducing processing time, the other advantage of the computer-assisted interviewing
is that it allows easy filtering of answers. Questions come up on the screen and are filtered based
on the respondent’s previous answer, so that the respondent does not have to answer questions that
do not apply to his or her situation. Computer-assisted interviewing also eases the randomization of
the question order, if this is required by research design.
Overall, computer-assisted interviewing improves standardized asking and recording of ques-
tions, makes it more convenient for respondents to participate, and shortens the time needed for
data processing and obtaining research results.

   

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
Using Online (Email) Personal Interviews 
Online personal interviewing combines the quality of face-to-face interviews with the efficiency
and economy of the Internet. Respondents can be contacted easily for follow-up interviews. Alt-
hough face-to-face may garner more rapport, emailed questionnaires provide the opportunity for
respondents to think about their answers over time and provide more complete responses.
A high drop-out rate occurs with online personal interviews, but that may be overcome by
working to develop mutual trust. Techniques for encouraging participation, and ultimately devel-
oping mutual trust, include soliciting an agreement before sending questions, self-disclosure by
the researcher, and disclosure about the research project.
Methodological questions come into play. Do you send the questions in bulk or do you utilize
a question-followed-by-reply process? Each has strengths and weaknesses. Sending questions in
bulk may lead to answering only the most interesting, the easiest, or the quickest questions. Asking
one question at a time is more reliable, but also more time consuming.

Conducting Interviews 
Interviewers need to be instructed on several issues before the structured interviews, since know-
ing how to behave during the interview will improve the quantity and quality of data they are
able to collect. These issues include the following:
 Know the interview schedule: The interviewer must know what the schedule is for the interview
to prevent being confronted by unforeseen difficulties, frustration, or leaving questions out.
 Introduce the research: Respondents must be given a good reason to participate in the re-
search. The introduction and invitation can be given verbally or in writing.
 Establish rapport: This is key to a productive interview experience for both the interviewee
and the interviewer. Rapport is a sense of comfort on the part of the interviewee, but it
must be restrained somewhat. If the interaction becomes too friendly, it may cause the
interview to get side-tracked, go on too long, or bring an interviewee to tailor his or her
responses toward “pleasing” the interviewer. This is best accomplished with face-to-face
interviews because of the visual communication cues.
 Know the introductory statement: Introductory statement is the start of the process. It should
include the following:
o Who you are
o The sponsoring organization
o Funding sources
o Topic of the research
o How the respondent has been selected
o Confidentiality procedures (assuring respondent that they cannot be identified)
o Participation is voluntary
o Respondents have the opportunity to ask questions about the research
 Questions, answers, order: As addressed above, the question should be asked exactly as stated.
Small changes to wording can make a big difference. Questions should also be asked in the
order that they are given on the interview schedule. Answers must be recorded as exactly as
possible. It is important to contemplate the questions and their order when designing a
questionnaire. If the questionnaire was designed with care, the question order has some
logic to it and it should not be arbitrarily changed during the process of interview.
 Probing: Sometimes probes are used if the respondents need help with their answers.
Probes are problematic because they may affect the response given, which may lead to
reliability issues. If probes have to be used, they should follow a standardized format.
Open-ended questions require a set option for the probe. However, the best probe for a
closed-ended question is to simply repeat the available answers. The interviewer should

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
 not make any suggestion toward an answer. Show cards are sometimes used in face-to-
face interviews to display the possible answers. This is done when there is a long list of
options (e.g., magazines respondents read); if there is a range of the same answer options
for several questions in the interview; when asking a sensitive question, and allowing the
respondent to number his or her response rather than verbalizing it. It avoids the tedium
and wasted time.
 Prompting: The interviewer suggests a specific answer to an interviewee. This should be
used only as a last resort if a respondent absolutely cannot come up with his or her own
reply. Prompting may tell more about the interviewer than the interviewee. The inter-
viewer has to do everything to make the respondent formulate his or her own response.
 Leaving the interview: The interviewer must remain professional and have a set response to
rebuff respondents who wish to engage the interviewer after the interview.
 Training: Good interviewing is a skill to be developed. It usually requires considerable
training and supervision to become good at it. Usually a minimum of one day of training
is required. Verification of an acceptable skill level is often completed through checking
response rates, tape-recording a sample of interviews, reviewing completed schedules for
accuracy, and making call-backs to a sample of respondents (usually 10 per cent).

Questionnaires  
Questionnaires are essentially structured interviews without an interviewer. They involve filling
out a form which is then returned to the researcher, often by mail, dropbox, or in person. Be-
cause the respondent has to fill in the questionnaire without the aid of an interviewer, it has to be
very clear and easy to follow. Typically, the self-completed questionnaires have fewer open-
ended questions, have easy-to-follow designs, and are shorter than interviews to avoid “respond-
ent fatigue.” 

Advantages of Questionnaires over the Structured Interview

 Questionnaires are cheaper and more convenient to administer than interviews, although
telephone interviews are completed immediately while questionnaires may be delayed by
mail.
 There is no interviewer effect with questionnaires. Research indicates that the impact of
social desirability bias is less with questionnaires than with an interviewer.

Disadvantages of Questionnaires versus Structured Interviews

 The researcher cannot explain the question.
 There is greater risk of missing data because of a lack of probing or supervision.
 It is difficult to ask a lot of questions.
 It is difficult to ask open-ended questions and questions requiring a filter.
 The researcher cannot be sure that the questionnaire is answered in a correct (originally
planned) order, since the respondent may read and answer the questions in the order
they want. This way, order effects may occur.
 They are not appropriate for certain respondents, e.g., those with limited literacy.
 The designated respondent may not have completed questionnaire, but the questionnaire
may be completed by another member of the household or a different person. There is
no way to control this in self-completion questionnaires.

The most common types of (self-completed) questionnaires are (1) paper questionnaires, (2)
online surveys, and (3) diaries.

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
Online Social Surveys  
Online surveys are another type of self-completion questionnaire and are done in two ways:
email surveys and web surveys.
 Email surveys tend to be aimed at smaller, more homogenous groups whereas web surveys
tend to be used more to study large online groups. The email survey may be sent with the
questionnaire embedded in the email or attached as a separate document. The embedded
format tends to appear more dull and featureless because of the reduced ability to for-
mat, but it is easier to complete. The attachment style allows for more professional for-
matting, but it requires more computer skill and often a particular computer program.
Research shows that the embedded format tends to generate more responses than the at-
tachment style.
 Web surveys allow for much greater formatting. They look more professional. They also
provide the opportunity to control the questions (filtering) and response options. The an-
swers can be downloaded directly into a database. Finally, web surveys give the researcher
access to very large populations and sample groups that might otherwise be inaccessible.
It is important to stress, however, that email, web, and mail surveys all carry the disadvantage
that one cannot be sure who the respondent is. The following are advantages and disadvantages
of online surveys compared to mailed questionnaires.
Advantages:
 Low cost
 Faster response and processing
 Fewer unanswered questions
 Better response to open questions
Disadvantages:
 Low response rate
 Restricted to online populations
 Requires motivation
 Confidentiality and anonymity problems with email surveys
 Multiple replies

Designing the Questionnaire 
There are also certain guidelines for the overall design of the questionnaire. These are the most
important:
 Clear presentation: The layout should be easy to read and understand. Various print styles
may be used, but they must be used consistently.
 Vertical or horizontal closed answers: Vertical presentation tends to be preferred because it
reduces the risk of confusion, makes better distinction between questions and answers
and makes coding easier, especially if pre-codes are inserted right into the survey instru-
ment. The horizontal format is often used for Likert scales that deal with a continuum of
options (e.g., a range from strongly agree to strongly disagree) rather than a group of
specific fixed choices (e.g., a list of political parties).
 Identifying the response sets in a Likert scale: To identify a response set, answers are first pre-
coded to allow immediate scoring of the items. Then the researcher may apply reverse
coding to some questions to identify response sets where the respondent answers all the
questions with the same answer. In addition, the researcher may intentionally create a
scale in which agreement with some items goes in one direction of the concept (e.g., high
job satisfaction), but agreement with other items indicates another direction (low job sat-

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
 isfaction). This helps to identify the response sets if they occur and the data can be ex-
cluded if necessary.
 Clear instructions: Tell the respondents if more than one response is allowed, or if the an-
swer should be circled or underlined so they are not left to guess.
 Keep question and its answer together: Do not allow answers to trail over onto the next page. The
respondent may miss the possible answer or just skip the question altogether.

Researcher‐Driven Diaries as a Form of Questionnaire 
Researcher-driven diaries can be considered a form of questionnaire. They can be used for
quantitative or qualitative research. Participants record their feelings, perceptions, actions, etc.
on a form shortly after they occur. The diaries may be “structured” with closed questions like a
regular questionnaire or “free text” where the experiences are written down without fixed re-
sponses (like answers to “open” questions). The “free text” method has the same coding and
time concerns that open-ended interview do. To be effective, participants should be given ex-
plicit instructions on how to complete the diary, the time periods for recording responses, the
types of experiences to be recorded and given a sample of a completed diary. There are several
advantages of diaries:
 Fairly accurate data about the frequency or time spent on behaviours of interest, and
about sequencing of behaviours.
 Good for getting data on sensitive matters, e.g., sexual practices.
There are also several disadvantages:
 More expensive than personal interviews.
 People may get tired of using them (attrition).
 Details may not be recorded fully enough, which may lead to errors and omissions.

Secondary Analysis of Survey Data 
Secondary analysis examines the large amounts of quantitative data that already exist. Using sec-
ondary data rather than collecting your own has the additional advantage of sparing an already
over-surveyed public yet another round of questions.
Advantages of Secondary Analysis
 Cost and time.
 High-quality data.
 Opportunity for longitudinal analysis.
 Subgroup analysis.
 Opportunity for cross-cultural (international) analysis.
 More time for data analysis.
 Reanalysis can offer new interpretations.
 It fulfills the wider obligations of the social researcher.
Limitations of Secondary Analysis
 Lack of familiarity with the data.
 Complexity of the data.
 The ecological fallacy.
 No control over data quality.
 Absence of key variables.

   

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
The Feminist Critique 
Some feminist theorists have argued that structured interviews and questionnaires are exploitive
because they are designed and utilized for a one-way relationship. In their view, the researcher
establishes rapport with the respondent in the survey in order to extract the information, but
guards against the respondent becoming too close, and does not do anything to reciprocate the
favour of granting the interview. However, feminist critique of structured interviews has declined
recently because surveys nowadays express more concern for the respondents’ rights to privacy
and decision making. Interviewers receive better training on dealing with sensitive topics and a
growing number of research findings consistent with feminist ideals now come from research
based on structured interviews.

Learning Objectives 
In this chapter, you should learn to do the following:
 Understand and explain the main advantages and disadvantages of structured interviews,
including reducing error, increasing accuracy, and dealing with interviewer effects
 Differentiate between self-completion structured interviews (questionnaires) and interviewer-
led structured interviews (survey research)
 Differentiate and describe the main types of interviewer-led structured interviews: face-
to-face interviews, telephone interviews, and computer-assisted interviews; list major
advantages and disadvantages of these types compared to each other
 Apply to the main guidelines for conducting interviewer-led structured interviews, in-
cluding establishing rapport, introducing the research, correctly posing questions, and re-
cording answers
 Differentiate and describe the main types of self-completion structured interviews: ques-
tionnaires, online social surveys, and diaries; list major advantages and disadvantages of
self-completion structured interviews
 Describe and suggest the ways to remedy the respondent problems in structured inter-
views, such as response sets, acquiescence, social desirability, laziness, or boredom
 Formulate good survey questions and propose an overall design for a survey questionnaire
 Differentiate between two main types of questions in an interview schedule—closed and
open-ended questions—and describe the general advantages and disadvantages of closed
questions as opposed to the open questions
 Apply general and specific rules for designing an interview schedule
 Understand the significance of the pilot study as a test run for a larger study to be con-
ducted later

Media Resources 
1790 US Census Findings http://www.censusfinder.com/1790-census.htm
 Compare the census questions for the 1790 census and the 1930 census (links are at the
bottom of the page).
 What are the differences between the two?
 What social conditions underlie each census?
 Are there any obvious biases that underlie each census?

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
50+ in Europe: The Survey of Health, Ageing and Retirement in Europe http://share-
dev.mpisoc.mpg.de/new_sites/Fragebogen/Vignettes/Vignettes%20Generic%20Typ%
20A.pdf
 Review the vignette questions in this questionnaire.
 How do they address their key issue?

Hopkins, D.J., and King, G. (2010). Improving anchoring vignettes: Designing surveys to
correct interpersonal incomparability. Public Opinion Quarterly 74(2): 201–222.
http://gking.harvard.edu/files/gking/files/implement.pdf?m=1360040626
 Read the article on the strategies of anchoring vignettes by Hopkins and King.
 Are these strategies used to anchor the vignettes in the survey referenced above?

Clinical Vignettes
http://www.go2itech.org/HTML/CM08/toolkit/tools/vignettes.html
 Try the vignette questions for the doctor and for the nurse.
 Compare to the answers provided.
 What is the difference between your questions and those posted?
 Are these standard questions we have been asked by a doctor/nurse?
 Do you need specialized training to administer the questionnaires?

Open versus Closed Questions Exercise

http://www.skillsconverged.com/FreeTrainingMaterials/tabid/258/articleType/Articl
eView/articleId/629/Open-versus-Closed-Questions-Exercise.aspx
 Try this exercise with a group of five students.
 What is the impact of being able to use only verbal language to describe a visual object?
 How important is it to take care in writing survey questions?
 How does the difficulty in writing questions differ between open and closed questions?

Morality Play http://www.philosophyexperiments.com/moralityplay/Default.aspx

 Try the morality test.
 Have the authors tried to deal with acquiescence? If so, how?
 What biases exist within the survey? What bias exists in the findings?
 How would your answers be affected by doing this survey face-to-face with an interviewer?
How would your answers be affected by doing this survey face-to-face with a friend?

The Alternative Fix, Frontline (Chapter 2)

http://www.pbs.org/wgbh/pages/frontline/shows/altmed/view/2_hi.html

Eisenberg, D.M., Kessler, R.C., Foster, C., Norlock, F.E., Calkins, D.R., and Delbanco,
T.L. (1993). Unconventional medicine in the United States: Prevalence, costs and pat-
terns of use. New England Journal of Medicine, 328(4): 246–252.
http://www.chiro.org/alt_med_abstracts/ABSTRACTS/Unconventional_Medicine.shtml
 What are the implications of these surveys conducted more than 20 years ago?
 What might lead patients to hold back their use of alternate medicine from their doctors?
 What might lead them to disclose to researchers?
 What errors might exist due to the methodology?
 What is final conclusion from follow-up research?

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019 
The Alternative Fix, Frontline (Chapter 3)
http://www.pbs.org/wgbh/pages/frontline/shows/altmed/view/3_hi.html
 What is the difference between scientific research and survey research?
 What sort of survey was conducted for the research into St. John’s Wort?
 What are the shortcomings to survey research regarding the effectiveness of medical
treatments?

Beres, M.A., Crow, B., and Gotell, L. (2009) The perils of institutionalization in neoliber-
al times: Results of a national survey of Canadian sexual assault and rape crisis centres.
Canadian Journal of Sociology, 34(1): 135–163.
http://ejournals.library.ualberta.ca/index.php/CJS/article/view/1613/5215
 What methods did the researchers use in this study?
 Would online personal interviews be effective in this study? Why or why not?
 What other methods could have been included to strengthen the research?
 How do the results of the survey correlate with the statistical data reported by the Canadian
government?
 Why is there variation between the two?

A Profile of Criminal Victimization: Results of the 1999 General Social Survey. Statistics
Canada. http://www.statcan.gc.ca/pub/85-553-x/85-553-x1999001-eng.pdf

Measuring Violence against Women: Statistical Trends 2006. Statistics Canada.

http://www.statcan.gc.ca/pub/85-570-x/85-570-x2006001-eng.pdf

Social Research Methods, Fifth Canadian Edition 
© Oxford University Press Canada, 2019