Archives   Calendar    Coordinating Center    Multisite Activities    
Project Description  COSP References  Study Sites    Search Our Site   
Table of Contents    Upcoming Meeting Agendas

______________________________________________

II. The Survey Process

    In a typical survey procedure there are a number of important steps that must be completed, including the construction of the survey instrument or questionnaire, the testing and re-testing of the instrument, the collection of the data, the coding of the collected data, the analysis of the data, and preparation of the final report.  In the COSP study we are focusing on all of these steps.  As an interviewer you will be involved in collecting data from: (site specific information) consumers, consumer operated program staff, service providers and agency records. The data are the answers given by the respondents to the questions on the survey instrument, and we must interview each respondent to obtain the answers.  Interviewing skills are therefore a very important part of the process.  The interviewer must understand the purpose of the survey, know how to create a comfortable interview environment, how to ask the questions, how to record the answers, how to communicate with the field coordinator, and how to keep track of the process from beginning to end.

  The answers obtained from all of the respondents will be grouped together for analysis so that the information reported in the results is about the group of answers, not about an individual.  Because it is the group of answers that will be analyzed, it is important that the questions be asked of each person in the same way.  This allows for the standardization of the instrument, and thus assures that each respondent is asked the same question.  For example, if you were to ask a sample of people how long they had been attending a specific peer-support program, you would have to ask each person the same question.  If you suddenly begin asking people how long they had been attending peer-support programs in general, you would no longer be able to compare the answers.

  We are concerned with maintaining the validity and reliability of the instrument.  Validity is whether the questions being asked are really measures of what the researchers want to know.  Reliability refers to the consistency of the responses received, or whether the answers would be the same if the question were asked at a future time.  If you change the meaning or the wording of the questions, the validity and reliability of the responses would also be affected thus making it very difficult to compare the responses.  It is also important that the survey is completed as much as possible.  Finally, you must maintain a neutral position while being able to clarify questions for a respondent and probe for a response when information is unclear or incomplete.  As an interviewer, your role is one of the most critical in the survey process.  The quality of the data depends on you.  You are a source of error when:  

1. You do not read questions as worded
   
2. You probe directively
   
3. You bias answers by the way you relate to respondents
   
4. You record answers inaccurately
   
5. You assume an answer rather than asking the respondent directly