1. Why would you need several questions instead of one?
The idea of having several questions allows the researcher to gain more information from the respondents, especially with different contexts and different demographics. One question might have the possibility of being split up into two questions to get a more specific response from the respondent. Surveys are about getting as specific as possible with obtaining information about a respondent’s background. There are several other reasons why one question cannot be enough; for example, in capturing the attitude of a respondent, a question might need to be split up in order to gain more understanding of the background of the individual respondent. Questions with the word “and” usually call for a question to be broken up into at least two questions. The answers to the survey questions will pinpoint more information.
2. Give an example of a question that is too general, and one that is too specific.
Survey question: What kinds of fruits do you like? –Too broad
Survey question: Of the two top candidates running in the Republican nomination race, which one would you select based on the largest amount of money fundraised during their campaign?
3. What’s a “loaded” question?
A loaded question is the type of question that shows a sense of bias toward one response, mainly the opinion of the creator of the survey question. For example, my second example question that was considered too specific deals with choosing a candidate of the Republican presidency campaign based off the amount of money fundraised. This shows that the only response is one answer; it does not give people the choice of choosing another answer. The idea of a loaded question is to not give the respondent a choice on the question at hand. Information would be collected in the wrong way because the numbers would be false; the results of the survey question will not prove anything because it does not show the true attitude of the respondent; instead, it forces the respondent to choose against their true attitude.
4. From what you can tell, what is the difference between open and closed questions?
The difference between open questions and closed questions is the idea of the respondents giving their full attitude and/or beliefs on the question at hand. Open questions allow for a deeper response from the respondent in terms of attitude and what they really feel about the question. A closed question allows respondents to select specific answers that may or may not be the real specifics of their attitudes. Open questions are used in survey where time is not a constraint on getting the results to analyze. Open questions are harder to sort because of the multitudes of respondents’ attitudes going into the answer of the question. Closed questions are easier to analyze and put into numbers to present to a business or organization. Closed questions are not as free to express the true attitudes of respondents compared to the open questions.
5. Provide your own bad examples, that violate the following “do’s”.
a. be concise
a. How do you feel about the city of Cedar Rapids redeveloping the flooded areas on the west side of the river into a green space?
b. use simple, clear language
b. How many times do you eat vegetables on an average day?
c. use mutually exclusive and exhaustive categories
c. Are you an undergraduate or graduate at Mount Mercy?
6. What is a “double barreled question”?
A double barreled question is a question that can be broken up into multiple questions, creating more specific responses from respondents. Double barreled questions can also be identified with the use of “and” separating two ideas. Double barreled questions can be easy to develop, but researchers no better than to put two ideas together, the idea of a survey and survey questions is to receive the most specific response and statistics in order to analyze the different demographics behind each of the respondents.
7. Differences between nominal, ordinal, and interval
Nominal questions have responses listed with numbers to the left of each response. The number has no relation to the response, it just shows that there is a possibility of different choices. For example, if there is the numbers 1,2 and 3 listed next to three responses, the respondent only has three answers to choose from in the question. Ordinal questions have responses with a spot to place a number to rank a certain number of responses in a particular order. For example, a respondent might be asked to rank different movie genres in a certain order to determine their most favorite movie genre, their least favorite movie genre, and other movie genres that fall in between. Interval questions allow respondents to put a rating on different subjects in a question. There are three types of interval questions: the Likert response scale (rate on a scale 1-5), a semantic differential scale (uses a point scale), and a cumulative or Guttman scale (respondents can check or select with the answers they most relate with or agree).
8. Filter/contingency question
A filter/contingency question involves the potential for respondents to answer secondary questions based off of their answer for a primary question. Usually a filter/contingency question is a yes or no response, the yes leading to a secondary question that has the potential to bring in more information about the respondent to the survey. Filter/contingency questions have the potential to get very complicated in trying to steer a respondent to answer the secondary questions. For example, “E-Poll” surveys ask if a person has heard about a certain show. If the respondent says “no”, it moves on to another show; but if the respondent says “yes”, a respondent is lead to more specific answers about the show.
9. What are some of the ways that sequence of questions can affect the results of the survey?
The sequence of questions is important, especially in the effort of the respondent to take the time and answer the survey questions. Certain levels of questions have to be presented at the right time in the survey to keep the respondent through the completion of the survey. A survey should not begin or end with elaborate questions because at the beginning, a respondent might not be ready to go into specific detail about a question, and at the end, a respondent might not have enough focus and energy to answer elaborate questions. These questions can cause surveys to come back to the researchers with incomplete results. Without these answers, researchers cannot move forward in accomplishing their task of compiling statistics. Therefore, different types of questions that can be found in surveys must be put in a particular order that is satisfying to the respondent filling out the survey.
10. What should the opening questions be like?
Opening questions should be easy, descriptive questions that allow the respondent to get into the flow of the survey. “Yes” or “No” questions can be considered opening questions because a respondent only has to only choose between two responses. Demographic questions, such as questions of gender and age, should be opening questions in a survey. Opening questions help build a respondents’ confidence to finish the other questions that will come in the later part of the survey. Questions that require long responses or a large amount of thought at the beginning will draw the respondent away.
11. What should the opening questions be like?
Sensitive questions should be placed in the later parts of the survey; this allows trust to build between the respondent and the creators of the survey who are asking for this type of information. Sensitive questions must have a smooth transition that assures the respondent that the questions are coming up in the survey. For example, there may be a sentence telling the respondent that the next set of questions deals with a personal subject, and they do not have to answer the questions if they feel uncomfortable about giving responses to these questions. Sensitive questions can either make or break a survey, depending on the transition of the survey into these sensitive questions.
12. Explain the “golden rule” of surveys.
The “golden rule” of surveys is : “Do unto your respondents as you would have them do unto you!” In other words, give the respondents the same type of survey experience that you would want in taking a survey. Researchers need to thank the respondents for the time they have taken out of their day to complete the survey. Researchers must also consider the feelings of the respondents, especially when dealing with sensitive questions. Surveys should try to be as short as possible, enough questions to receive the basic statistics of a survey. Researchers must care about the respondents, and in return, the respondents will care more about the survey.