Questionnaires

The material on these pages is from the 4th edition of Successful Writing at Work, by Philip C. Kolin, published by D.C. Heath and Company, 1994. This material is no longer available in the current edition of the text and is provided here due to this lack of availability. Be aware that the material is copyrighted and is not to be distributed beyond our class. Students in the English 202D course may print one (1) copy for use within the course only. They may print one additional copy ONLY if the original has been distroyed.

Two Audiences:

  1. The Respondents
  2. The Person(s) who receive(s) the report

Two Basic Types of Questions

Open-Ended Questions (Essay)

    Examples:
        Describe the changes you would make in the registration process at PSU-Hazleton.
        Explain which part of your job is the hardest to perform and why.
        Discusss your reasons for attending PSU-Hazlton.

    Open-ended questions can be valuable for generating responses you did not anticipate. In addition, open ended questions can provide you with an indication of how knowledgable your respondents are about a given subject. The questions can be quite valuable if you have the time to examine the answers thoroughly and if the number of respondents is relatively small, such as fifteen or twenty.

    This type of question requires a great deal of effort from the respondents, who will have to organized, compose, and write their answers. If your question is not direct and sufficiently focused, you could easily get many irrelevant answers from respondents. Moreover, valuable answers might be mixed with these irrelevant comments. Finally, responses to open-ended questions are challenging to summarize (especially when large groups are surveyed), and these responses must be properly coded.

Closed Questions

    Closed questions offer respondents a limited number of choices. Because closed questions are easier to answer, code, and tabulate, they are used more often than open-ended questions. Closed questions fall into five catagories:
    1. Questions that offer only two choices: These are sometimes called dichotomous questions because they present the respondent with a dichotomy, a division of the subject into two mutually exclusive parts. The respondent's choice is limited to one of these two parts. Use this type of question only when the topic can be reasonably understood and explained in either/or terms.
    Examples:
        a. Yes/no questions
            Should women be required to register for the draft?    yes_____________  no_______________
            Is there a history of breast cancer in your family?        yes_____________  no_______________

        b. True/false questions
            People who smoke in public places should be fined    true_____________  false_____________
            Licensed Practical Nurses should, after special
            training, be allowed to start IV's.                                true_____________  false_____________

        c. Two specific objects or types identified
            Which kind of radio station do you prefer?                AM_____________  FM______________
            What kind of coffee do you usually drink?                 regular___________  decaf_____________
 

    2. Multiple-choice questions: These questions usually offer respondents three to five answers from which to choose. One of those answers can be marked "other," "options not given," "do not know," "undecided," or "none."
    Examples:
        What type of domesticated animal would you choose for a pet?
        dog_____  cat_____  bird_____  fish_____  other (please specify)__________
        Which kind of music do you like to listen to most often?
        rock 'n' roll_____  country/western_____  jazz_____  classical_____  blues_____  New Age_____
        How many times a day do you use E-mail?
        1_____  2_____  3_____  4_____ more than 4 (please specify)__________

    3. Rating-scale questions: These questions ask readers to rate (or evaluate) an individual, program, policy, or option according to a carefully graduated scale. The respondent indicates the degree or extent of his or her opinion by marking an appropriate number on the scale. Always make sure that you specify precisely what the numbers on a scale mean. Usually this involves defining the items at either end of the scale. If the scale has an odd number of gradations, the middle position is considered a "neutral" opinion, as the following examples illustrate.
        What is your overall view of the Clinton administration?
        Excellent                                        Poor
           1            2            3            4        5
        PSU-Hazleton should change from a semester to a quarter calendar.
        Strongly agree                                Strongly disagree
            1            2            3            4        5

    4. Ranking questions: With these questions a respondent is asked to assess the relative significance of a series of options and to assign each a vlaue, often by laveling them 1, 2, 3, 4.
        Indicate your order of preference for the kind of nursing you would like to do after graduation.
        acute care_____  industrial_____  home health_____  school_____
        Please rank in order of importance the following reasons for your decision to do your banking at First National.
        Superteller_____  Saturday hours_____  Free checking_____  Location_____  Investment couseling_____

    5. Short-answer questions: These questions require respondents to fill in the blank or write a brief answer.
        Give your date of birth (day, month, year):_________________________
        How long have you lived at your current address?___________________
        What was your chief reason for taking the noon section of English 202D?__________________________

Reliable and Valid Questions

    1. Phrase your questions precisely. Specify exact quatitites, times, or money. Avoid generalizations like "enough" or "good."
    2. Ask manageable questions. Broad questions ask respondents to write ten pages just to begin answering them. As a resulty, respondents will either leave them unanswered or write brief, often vague answers. Your questions should offer direction, but allow the respondent to make up his or her own mind.
    3. Write questions that let respondents decide for themselves. Avoid using loaded words, such as "should we stop requiring business majors to take impractical courses like 202D?" Or "Do you believe Americans pay too much money in taxes?"
    4. Your qestions should not insult or indict the respondent. No matter how a question such as "Have you stopped picking fights with your spouse?" is answered, it accuses the respondent of the deed.
    5. Do not write a question that requires the respondents to do your research. Courtesy requires that you not ask them to go to anymore trouble than it takes to fill out the questionnaire.
            Do NOT ask your respondents questions like the following:
               Ask your immediate superior to supply you with …
                After checking your files and last year's manual, indicate …
    6. Limit your questions to recent events. Do not ask respondents to search their memories to recall opinions that they held years ago or to discuss details of an event that they may not now clearly remember.
    7. Write questions in language appropriate for your audience. A questionnaire directed to specialists may well include a few technical terms, but if respondents are not familiar with your jargon, avoid it.
    8. Each question should cover only one item. Do not confront respondents with a question that may demand an unnecessary, misleading, or contradictory choice. For example, "Do you prefer the Beatles or the Rolling Stones?" asks respondents to choose between the two. But respondents may like both groups and could not register their preference as the question is worded. Make two separate questions to be sure of obtaining a reliable response.
    9. Do not ask the same question twice. A question writer may think that there is a fine, subtle difference between two questions, but respondents may be unable to detect any difference. Check one question against another. Avoid:
           What is your favorite television program?
            Which program do you like to watch most on television?
    These are the same question.
    10. In multiple-choice questions, supply your respondents with clearly differentiated options. Avoid comparing apples to oranges, in other words. Also avoid overlapping catagories.
    11. Include all necessary option in multiple-choice questions and those that ask respondents to rank items. Omissions are especially dangerous when respondents are forced to choose between two extremes and are not given enough options for qualified agreement or disagreement. Give respondents the option of saying that they have no opinion or that they see no change.