American Sociological Review · 1955 · 32 citations · 2 references
Family MedicineFamily StructurePsychometricsSame InterviewerFamily FormationSocial SciencesPsychologyInterviewer TraitsInvoluntary ChildlessnessSurvey (Human Research)Family LifeSelf-report StudyPsychological EvaluationFamily RelationshipsJob AnalysisSubtle ScalesBehavioral SciencesFamily StabilityExperimental PsychologyChild DevelopmentSociologyFamily PsychologyDemographyMedicineSurvey Methodology
We have pointed up two different sets of interviewer traits, those connected with conscientiously completing the questionnaire and those derived from understanding of the study. The first set was associated with high reproducibility on scales with obvious patterns, the second with high reproducibility on subtle scales. As the two traits were found to be negatively correlated, it is likely that few interviewers will excel in both abilities. Which ability is preferable, if a choice must be made, will depend on the kind of interviewing job required. In surveys where all questions are preceded, where the interviewer has to follow exactly all instructions, one type will be superior. In qualitative studies, where each interviewer has a relatively small number of interviews and uses primarily open-ended questions, the second type would be preferred. It is important to realize that the same interviewer will not do well in all survey jobs. A determined effort should be made to help the interviewers who are weak in either of the areas. This is especially true of the interviewer who has difficulties in understanding the purpose of the study. The necessity of completing the questionnaire according to instructions is usually stressed sufficiently in training, while the instruction on the background of the study may be neglected. In this study it was noted the interviewers were eager to learn more about the background of the study, although close to two days were devoted to this topic. There was a tendency in the post-training evaluation questionnaire for the interviewers who turned out high on the subtle scales to demand more background information, while the interviewers who were high on the prohibition scale wanted more instruction on the interviewing and completing the questionnaire. That is, both groups asked for more training in the field in which they were already good. This difference between the two groups, although small, emphasizes the necessity of creating interest in those topics to which the trainee would be not drawn by himself. Similarly, during the supervision of the interview in the field, conscientious completion of the questionnaires is likely to receive the principal attention. It is easier to devise checks and to set standards for following the explicit instructions than to keep a continuous check on the understanding of the interviews. It is hoped, therefore, that the type of measures discussed here can be developed as aids in field supervision.
2