Bad questions any questions that prevent or distort

This preview shows page 2 - 3 out of 3 pages.

the sequence. Bad questions: Any questions that prevent or distort the fundamental communication between the researcher and the respondents. A question is bad when it is: Unanswerable, to complex, Leading, Loaded, Double-barrelled (more than one issue at a time) or exceeds the memory capability of respondent.Step 4: Determine Layout and Evaluate. Introductory section: Gives the respondent an overview of the research. Screening questions: Identify qualified prospective respondents. Prevent unqualified respondents from being included in the study. Research questions section: Second section of the questionnaire that focuses on the research questions. Type of information/questions: Critical information – directly related to research problem. Classification information – demographic/socio-economic information. Identification information – name, address, phone number. Considerations:Introduction should include a general description of the study. Instructions should be clearly expressed. Questions and Scale measurements should follow a logical order (Response order bias - the order of the questions, or of the closed-end responses to a question, influences the answer given). Begin with simple questions. Ask personal questions at end of survey. Place questions that involve personal opinions, attitudes, and beliefs towards end of survey, but before demographics. Avoid asking questions using a different measurement format in the same section of the questionnaire. End with a thank-you statement. Step 5: Obtain Initial Client approval. Copies of the questionnaire should be given to all parties involved in the project. Gives Client’s an opportunity to provide suggestions of topics overlooked or to ask any questions.Step 6: Pre-test, revise and finalise.Pretest the questionnaire on a small sample of respondents to identify potential problems: Content, Wording, Sequence, Form and layout, Question difficulty, Instructions. Respondents should be similar to those who will be in the actual survey. Initially pilot testing should be undertaken via personal interviews irrespective of administration method chosen. Subsequent pilot tests should be carried out using the chosen administration method.Step 7: Implement the Survey. A cover letter can be used to enhance respondents willingness to participate in the questionnaire.Evaluating Measurement Scales:All multi-item scales should be evaluated for reliabilityand validity. Reliability: the extent to which a scale can reproduce the same or similar measurement results in repeated trials. Validity:whether a scales measures what it is supposed to measure. A multi-item scale is normally a construct. A construct is something that can’t be directly measured, that is, it is measured by combining multiple items (questions). Reliability:Cronbach’s Alpha: Can take values between 0 and 1 ▪ 0 = no reliability ▪ 1 = perfect reliability. ▪ Cronbach’s alpha should be at least 0.7before we say the scale is

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture