Interviewer and coder bias are removed because the interviewer is simply checking a box, circling a category, recording a number, or punching a key. Reading response alternatives may jog a person’s memory and provide a more realistic response. Also, because the option of expounding on a topic is not given to a respondent, there is no bias toward the articulate. Finally, the coding and data acquisition process is greatly simplified.
There is a difference between a pre-coded open-ended question and a close-ended question. An open-ended question allows the respondent to answer in a freewheeling format. The interviewer simply checks the points on the pre-recorded answers as they are given. Probing is used, but a list is never read. If an answer is given that is not pre-recorded, it is written verbatim in the ‘other’ column. In contrast, the closed-ended question requires alternatives to be read or shown to the respondent.
Traditionally, the data acquisition process has separated the two-item response option from the many-item type. A two-choice question is called dichotomous and the many-item type is often called multiple-choice or multi-chotomous. With the dichotomous closed-ended question, the response categories are sometimes implicit. For instance, how would you respond to the following question: “Did you buy gasoline for your automobile in the last week?” Obviously, the implicit options are “Yes” or “No”, regardless of the fact that a respondent may say, “I rented a car last week and they filled it up for me. Does that count?” the questions would still be classified as dichotomous closed-ended.
The simplest form of data acquisition is a closed-ended question or dichotomous choice. They are easy to administer and usually evoke a rapid response. For example, limiting the responses to a simple “Yes” or “No”, “Agree” or “Disagree” or “Greater than” or “Less than”. Many times a neutral or ” no opinion/don’t know” option is added to dichotomous questions to take care of those situations. Sometimes the interviewer will jot down “DK” for “Don’t know” or “NR” for “No response” if the neutral option is omitted from the data acquisition questionnaire.
Dichotomous questions are prone to a large amount of measurement error. Because alternatives are polarized, the wide range of possible choices between the poles is omitted. Thus, question wording is very critical to obtain accurate responses. Questions phrased in a positive form may well result in answers opposite from those expressed in a negative format. In questions requiring a “Greater than” or “Less than” response, response may vary. These problems can be overcome using a split ballot technique. One-half of the questionnaires are worded with greater than listed first and the other half with less than first. This procedure will aid in reducing potential bias.
Each type of closed-ended question has unique disadvantages. For the dichotomous data acquisition form, frequently the responses fail to communicate any intensity of feeling from the respondent. In some cases, the matter of intensity does not apply, as for the previous example on gasoline purchasing, but instances do arise in which the respondent feels very strongly about an issue but the intensity is lost in the dichotomous data acquisition form. The multiple response closed-ended question has two additional disadvantages. First, the researcher must spend time gathering the list of possible responses, secondly the range of possible answers. If the list is too long, respondents may become confused or disinterested.
Leave a Reply