RESPONSE EFFECTS IN THE NATIONAL CRIME SURVEY

• NCJ Number: 54224
• Journal: Victimology Volume: 3 Issue: 1-2 Dated: (1978) Pages: 110-124
• Author(s): R G LEHNEN; A J REISS
• Date Published: 1978
• Length: 15 pages
• Corporate Author: Visage Press, Inc
• Sale Source: Indian University School of Public Affairs
• Publication Type: Survey
• Language: English
• Country: United States of America

Annotation

THE EFFECTS OF SURVEY DESIGN FEATURES ON THE RESPONSES OF PARTICIPANTS IN THE NATIONAL CRIME SURVEY (NCS) ARE ANALYZED.

Abstract

THE ROTATION TECHNIQUE USED BY NCS TO ELIMINATE CERTAIN SAMPLING BIASES (E.G., THE TENDENCY OF RESPONDENTS EITHER TO OMIT VICTIMIZATION INCIDENTS OR TO MISPLACE THEM IN TIME) MEANS THAT RESPONDENTS ARE INTERVIEWED REPEATEDLY, ONCE EVERY 6 MONTHS FOR 3 YEARS. THE SUBJECTS RESPOND TO A BASIC SCREENING QUESTIONNAIRE IN EACH INTERVIEW. WHENEVER A RESPONDENT REFERS TO AN INCIDENT OF VICTIMIZATION IN THE SCREENING, A DETAILED CRIME INCIDENT REPORT IS COMPLETED. ALTHOUGH THE MAJORITY OF INTERVIEWS ARE CONDUCTED IN PERSON, SOME ARE CONDUCTED BY TELEPHONE. THE RESPONSES OF SURVEY PARTICIPANTS BETWEEN JULY 1, 1972, AND DECEMBER 31, 1975, ARE ANALYZED TO DETERMINE THE EFFECTS OF REPEATED EXPOSURE TO THE BASIC QUESTIONNAIRE, REPEATED EXPERIENCE WITH CRIME INCIDENT REPORTING, AND MEDIUM OF INTERVIEW ON THE LEVEL OF INCIDENT REPORTING BY RESPONDENTS. ALL THREE VARIABLES HAVE A SUBSTANTIAL IMPACT ON THE LEVEL OF REPORTING. DEPENDING ON THE TYPE OF INTERVIEW MEDIUM, NUMBER OF PRIOR INTERVIEWS, AND PREVIOUS INCIDENT-REPORTING EXPERIENCE, THE PERCENTAGE OF RESPONDENTS REPORTING AT LEAST ONE INCIDENT VARIES FROM 4 TO 36 PERCENT. INCIDENT REPORTING DECLINES WITH REPEATED EXPOSURE TO THE BASIC QUESTIONNAIRE BUT INCREASES WITH REPEATED INCIDENT-REPORTING EXPERIENCE. TELEPHONE INTERVIEWS RESULT IN LESS INCIDENT REPORTING THAN DO INPERSON INTERVIEWS. THE FINDINGS INDICATE THAT 'FATIGUE' AND RESISTANCE INCREASE OVER A RESPONDENT'S 3-YEAR EXPERIENCE WITH NCS BUT THAT RESPONDENTS WHO HAVE REPORTED INCIDENTS IN PAST INTERVIEWS ARE LIKELY TO CONTINUE TO DO SO. ALTHOUGH THE INTEPRETATION OF THESE EFFECTS IS OPEN TO SOME QUESTION, THEIR MAGNITUDE IS SUFFICIENT TO CAUSE CONSIDERABLE ERROR IN ESTIMATES OF VICTIMIZATION BASED ON NCS DATA. THE POSSIBLITY OF REDESIGNING THE SURVEY OR ADJUSTING THE VICTIMIZATION ESTIMATES TO COMPENSATE FOR THIS ERROR IS RAISED. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. (LKM)