What Is a Questionnaire and How Is It Used in Research?
A questionnaire is a set of questions on variables under study set by the researcher and to be answered by sampled respondents. For a study that requires the use of a questionnaire to collect data, those targeted subjects expected to provide responses to the set questions are referred to as respondents.
The questionnaire is administered by the researcher to the respondents, and if there are multitudes of questionnaires to be administered or considering the sensitivity of the researcher's presence while the responses are being provided, the researcher might decide to leave the questionnaires to be self-administered by the respondents themselves.
Questionnaires are always the same, i.e., they must contain the same set of questions in a similar order so that the same relative data are obtained from all members of the sampled population. Sometimes, investigators decide to frame questions in the form of items (indicators) that measure specific concepts or variables under study.
Forms of Questionnaires
In discussing forms in which questions can be structured in a questionnaire, there are open-ended questions and closed or fixed-choice questions, depending on the decision of the researcher.
1. An Open-Ended Question
This type of question allows the respondent to compose and provide a mind-generated answer rather than choosing from a set of predisposed alternatives stated in a questionnaire. This style is reliable to some extent in acquiring valid data from respondents inasmuch as they can write what they mean in their own words.
An example of an open-ended question is given thus: "How did you settle disagreements between friends in the same workplace?" However, questions of this form appearing in a questionnaire suggest that the responses will be classified and recoded in quantitative form in order to be able to subject the answers to the same quantitative analysis that the rest of the questions in the questionnaire will undergo statistically.
Answers must be interpreted carefully before the researcher is able to say, for example, that a certain percentage of respondents settled their disagreements internally or amicably with mutual understanding without the intervention of management or any other third party.
The task of embarking on reclassifying, categorizing, and quantifying open-ended questions is sometimes being circumvented by lazy researchers because of the extra time and logic inherent.
So, some researchers prefer closed or fixed-choice questions, especially student researchers. But open-ended questions are appropriate and cannot be avoided in some situations where the researcher does not find adequate literature or previous standardized question items to measure a variable, studies a new topical area, and embarks on examining a study further to find out more about the set of dimensions of a variable.
2. Closed or Fixed-Choice Question
This is a type of question that requires respondents to select a choice between or among a given set of answers. In this case, the stated answers may assume the opinion of respondents on the questions posed.
For example, the following question can be asked:
What prompted you to go for Ifa therapy?
A. modern health care failed
B. modern health care is too costly
C. lack of trust in the efficiency of modern health care
D. Ifa priests are easily accessible.
From the above, respondents (in this case, the Ifa therapy patients) are expected to pick their choice of answer from the options stated either by ticking or underlining as appropriate, subject to the instruction given, to indicate their preferred response.
However, questions may be posed in a different way that requires the respondents to agree or disagree, or in any other similar manner that suggests the use of a Likert scale. Thus, the question or items will inform the subject to select the rated responses relevant to the itemized statement. For example, I have the opportunity to develop close friendships at my workplace.
1. Strongly agreed 2. Disagreed 3. Undecided 4. Disagreed 5. Strongly disagreed
At times, researchers adopt a score range starting from 1 to 10 to 20 or above while measuring a variable and therefore ask respondents to indicate their score level. For instance, in measuring optimism, questions can be asked where the individual respondent will be rated based on their level of optimism against a pre-set score highlighted by the researcher. On the contrary, closed questions do not allow original answers emanating from the respondents' own ideas, unlike open-ended styled posers. Therefore, it can be cumbersome for researchers to know exactly what they are measuring.
Regarding the closed question exemplified above, the respondent may have an answer that is significant but is not included in the options. In fact, such options may not, in some cases, cover what the respondents intend absolutely, therefore leaving the data paralyzed.
Administering Questionnaire
There are different means through which questionnaires can be administered, or rather, ways by which researchers get questionnaires across to the sampled population, depending on the spread or distance of the prospective respondents. It can be administered to respondents by the interviewer, especially if subjects include elderly people and other groups of people who, by any chance, are not in a better position to fill out questionnaires in the desired manner.
A shortcoming of this is referred to as bias. This means that the responses given are influenced by the presence of the researcher. In addition, it can make research more costly due to the recruitment of trained interviewers in the study field. However, the most common means is the self-administered method, where prospective respondents are given questionnaires to fill out themselves at their convenience.
This self-administered method can be accomplished in various ways while trying to get questionnaires across to the sampled subjects. This can be done through a face-to-face approach, postal service, telephone, email, internet, etc.
1. Face-to-face Approach
In this way, the researcher meets the individual sampled subject in person, introduces the purpose of the meeting, hands over the questionnaire (if there is acceptance by the prospective respondent to give a response), and both parties will agree on when the filled-out questionnaire will be returned. In fact, where the would-be respondents are clustered in a particular location, it makes it easy for the researcher to distribute questionnaires to them as a group once and for all.
2. Postal service
The postal questionnaire is mailed to respondents with a stamped envelope for return to the researcher. It provides an inexpensive way of gathering data, especially if the respondents are dispersed over a wide geographical area. The return rate, though, does not often exceed 50% of the sampled population and is sometimes below 25 percent (Haralambos and Hulborn 2008). The fallout of this method affects data, making them one-sided or biased in terms of responses, especially if the set of respondents that returned the questionnaire are mainly a group with similar but not diverse characteristics. For example, most of them may be the working population, females, etc.
3. Telephone
After identifying the individual prospective respondent and the telephone address is intact, the researcher can put a call to the respondent in order to administer the questionnaire. All the questions are asked over the phone, and the responses are recorded. This is often done when doing marketing research. Meanwhile, a prior agreement would have been made by both parties before the exercise was done. It is important to state that this method is rarely used by sociologists, as it barely satisfies the purpose of the study.
Aside from this, this cannot be appropriate for a questionnaire that contains a large number of questions. In a similar way, it is not time-prudent because, in a case where fixed questions are asked, the researcher might have to repeat options as respondents often forget some options earlier stated. Generally, the response may be low.
4. E-maill
There are some sensitive questions that are best answered without the knowledge of anyone close to the respondents. For instance, a question that requires a respondent to declare his or her HIV/AIDS status needs absolute privacy in order to get an adequate reply. In short, very sensitive questions can best be e-mailed to prospective respondents without considering the task of getting the e-mail addresses of all the samples needed for the study.
Another challenge of this method is that only those who have access to the internet will be accessible; therefore, it is unrepresentative, and there is a possibility of a low rate of return because some individuals can decide to ignore or abeyance the e-mailed questionnaire out of an indifferent or lackluster attitude.
5. Internet
This seems to be the newest and supposedly easiest way of reaching out to prospective respondents, irrespective of their distance from the researcher. In this case, the questionnaire is designed on a particular website, where respondents are asked to visit and fill out the information. In fact, there are some famous vendors who render this service to any researcher who wants this method.
Most of the time, the respondents are lured to fill out the questionnaire while being promised gifts like recharge cards. However, this method also cannot pass the litmus test of adequate representativeness because a sampled population without internet access at the time the study is being done is denied participation. This automatically leads to the possible unevenness of the collected data.
In fact, the major problem common to all the methods mentioned above is that a large group of less privileged people are left out of the study, and this can adversely affect the reliability and validity of the data collected on which analysis and subsequent conclusions can be based.
Advantages of a Questionnaire
- The questionnaire avails of the opportunity to collect a large quantity of data from multiple sets of people within a relatively short period of time.
- Data obtained from questionnaires can be analyzed pronto and efficiently with the aid of computer software programs such as Statistical Package for Social Sciences (SPSS), which is commonly used by sociologists in analyzing large quantities of quantitative data.
- Questionnaire data can be analyzed more scientifically and objectively than qualitative data.
- An adequate comparison of quantitative data obtained through a questionnaire can be achieved. This is done when information is compared for reliability testing and to predict certain actions of subjects in a particular case. For example, several results of the electioneering process can be studied to hypothesize the next voting result.
- Statistical data from questionnaires can be analyzed so that new hypotheses and theories can be produced, especially with a meticulously prepared set of questions.
- Large samples are very easy to acquire through the use of questionnaires. This ensures adequate representativeness and reliability of the results obtained from analysis.
Disadvantages of Using a Questionnaire
- The wording of questions or items may prevaricate on the meaning of some words or concepts to respondents in different ways; hence, obstruction to data validity is bound to occur. For example, "I have the opportunity to get to know my coworkers." Where respondents are asked to indicate their level of agreement with this item, 'get to know' in the sentence may convey different meanings to different respondents; thus, answers may not be reliable for that item. This is another reason sociological researchers always avoid "double head" questions or items, e.g., workplace friendship developed because of nearness and shared experience. In this way, nearness and shared experience convey different ideas because some who agree with nearness may want to disagree with shared experience.
- Because respondents cannot provide information that is not on request, there is a limitation to the volume of data that will be available. A spillover effect from this is that new hypotheses and theories that were not stated ab initio are difficult to develop.
- Operationalization of concepts in questionnaires can lead to distortion of such concepts, contrary to the real picture of the concepts in the social world. This is a result of the researcher's imposition, a situation where researchers impose their own assumptions on research content.
- The responses and validity of questionnaire data can also be hindered due to the problem of reactivity, when subjects intentionally give responses that they assume to be the target of the researcher even though that response is not true and original to them.
- The opportunity of reaching out to far-distant subjects with the associated supposed objectivity of study can turn out to be a disadvantage because this rules out the possibility of respondents understanding the meaning and motives of the subjects of research.
- The process of reclassifying and recoding open-ended questions into quantitative data deteriorates the quality of the responses supplied because some important information would have been precluded.