Age and Method: A Comparison of Semi-structured Interviews Face-to-face and via Telephone with Children
Author: Vogl, Susanne , susannevogl@gmx.de
Department: Sociology
University: University of Eichstaett-Ingolstsadt, Germany
Supervisor: Prof. Dr. Siegfried Lamnek
Year of completion: 2011
Language of dissertation: German
Keywords:
qualitative intervie
, mode effects
, children as responde
, age effect
Areas of Research:
Logic and Methodology
, Childhood
Abstract
Methods used in social science are typically developed with adults in mind. Because the abilities of children differ from those of adults, we cannot use the same methods, at least not in an unquestioned manner. Possible adjustments of research instruments are problematic because we still lack competence in the cognitive, interactive, and verbal capacities of children in interview situations. To tackle the problems of construing adequate research instruments and create encouraging settings for children (of different age groups), we need to take those skills into account.
In this study, I have analysed the applicability and methodological specificities of interviewing children at the ages of 5, 7, 9, and 11 years old. Semi-structured interviews conducted face-to-face versus those conducted over the telephone are compared. The methodological innovation of this study exists on two levels: interviewing children at a very young age and conducting qualitative telephone interviews. 112 semi-structured interviews with 56 German boys and girls at the age of 5, 7, 9, and 11 years old were conducted. Each respondent was interviewed via telephone and face-to-face with a structurally similar question guide. Potential mode and age effects are “measured” by three (interdependent) means: a) verbal, social, and interactive skills; b) data quality; and c) formal ask/question formats.
Main conclusions are as follows: semi-structured interviews are applicable with children at the age of 5 to 11 years but with different potential insights. The results draw a mainly positive picture of children as competent respondents.
In terms of skills, the ability to change perspective turned out to be crucial for the success of an interview. Although this skill could be expected from the age of about 7 years, according to psychological literature, it seems to be present in interviews slightly later than that. Verbal skills seem to be sufficient for interviews even at the age of 5 years. Interactive skills develop over time, with a big leap between the age of 7 and 9 years. Additionally, hypothetical thinking can be expected from about 9 years onward and abstract thinking from 11 the latest. From 11 years onward, required cognitive skills are well comparable to those of adults. Altogether, the analyses of skills draw a surprisingly positive picture of children as informants.
Telephone interviews are most adequate from the age of 11 onward because generational and authoritarian relations between interviewer and interviewee are less obvious. Overall, there were surprisingly few differences in mode. As with age effects, it often is the personality of the respondent that is a stronger determinant of performance than the interview mode. In other words, if one interview mode works well or badly with a child, the other mode is likely to work out similarly.
Concerns about data quality of children's responses could be put into perspective and can be overcome. Additionally, in comparison to adults, those concerns are put into perspective because with the latter, similar difficulties often occur. Nevertheless, interviewing children poses methodological implications and the younger the respondents, the greater the implications.
