Self Report In Research and Using And Preparing Structured Self-Report Instruments (VII)

Types of Questions In Research Questionnaire and Composite Scales

Self Report In Research and Using And Preparing Structured Self-Report Instruments (VII)Composite Scales.

    Assembling a high-quality structured self-report instrument is a challenging task. This section discusses components of such instruments and offers some guidance in constructing them. Specific Types of Closed-Ended Questions It is especially challenging to create good-quality closed-ended questions. 

    Researchers must pay careful attention to the wording of questions and to the content, wording, and formatting of response options. Nevertheless, the analytic advantages of closed-ended questions are compelling. 

Various types of closed-ended questions, many of which are illustrated in Table 15-1, are discussed here.

    1:Dichotomous questions require respondents to make a choice between two response alternatives, such as yes/no or male/female. Dichotomous questions are considered most appropriate for gathering factual information.

    2:Multiple-choice questions offer more than two response alternatives. Dichotomous items are often considered too restrictive by respondents, who may resent being forced to see an issue as either “yes” or “no.” Graded alternatives are preferable for opinion or attitude questions because they give researchers more information (intensity as well as direction of opinion) and because they give respondents a chance to express a range of views. Multiple-choice questions most commonly offer three to seven alternatives.

    3:Cafeteria questions are a special type of multiplechoice question that asks respondents to select a response that most closely corresponds to their view. The response options are usually full expressions of a position on the topic.

    4:Rank-order questions ask respondents to rank target concepts along a continuum, such as most to least important. Respondents are asked to assign a 1 to the concept that is most important, a 2 to the concept that is second in importance, and so on. Rank-order questions can be useful but need to be handled carefully because respondents sometimes misunderstand them. Rank-order questions should involve 10 or fewer rankings.

    5:Forced-choice questions require respondents to choose between two statements that represent polar positions or characteristics. Several personality tests use a forced-choice format.

    6:Rating questions ask respondents to evaluate something along an ordered dimension. Rating questions are typically bipolar, with the end points specifying opposite extremes on a continuum. The end points and sometimes intermediary points along the scale are verbally labeled. 

    The number of gradations or points along the scale can vary but should always be an odd number, such as 7, 9, or 11, to allow for a neutral midpoint. (In the example in Table 15-1, the rating question has 11 points, numbered 0 to 10.)

    7:Checklists encompass several questions that have the same response format. A checklist is a twodimensional arrangement in which a series of questions is listed along one dimension (usually vertically) and response alternatives are listed along the other. 

    This two-dimensional character of checklists has led some people to call these matrix questions. Checklists are relatively efficient and easy for respondents to understand, but because they are difficult to read orally, they are used more frequently in SAQs than in interviews. Figure 15-1 presents an example of a checklist.

    8:Calendar questions are used to obtain retrospective information about the chronology of different events and activities in people's lives. Questions about start dates and stop dates of events are asked and recorded on a calendar grid.. Respondents can often better reconstruct the dates of events when several events are recorded in juxtaposition.

    9:Visual analogue scales (VAS) are used to measure subjective experiences, such as pain, fatigue, nausea, and dyspnea. The VAS is a straight line, the end anchors of which are labeled as the extreme limits of the sensation or feeling being measured. Subjects are asked to mark a point on the line corresponding to the amount of sensation experienced. 

    Traditionally, the VAS line is 100 mm in length, which facilitates the derivation of a score from 0 to 100 through simple measurement of the distance from one end of the scale to the subject's mark on the line. 

Composite Scales

    A scale provides a numeric score to place respondents on a continuum with respect to an attribute being measured, like a scale for measuring people's weight. Many studies that collect data through self-report use a psychosocial scale, which is used to discriminate quantitatively among people with different attitudes, fears, motives, perceptions, personality traits, and needs. 

    Scales are usually created by combining several closed-ended items (such as those described in the previous section) into a single composite score. Many sophisticated scaling techniques have been developed, only two of which are discussed here.