This research was intended to explore the way in which the ‘sense of self’ develops over time in humans. The aim was to understand patterns in the kinds of reply given by particular age groups that could help corroborate a sense of developmental progression in children’s sense of self. The study was based on the work of Rosenberg which hypothesized that as we grow from young children into young adults our view of ourselves shifts from the externally-recogniseable facts to reflections on our character and emotional makeup. In this study two interviews were supplied (out of a larger sample) and were analysed with each statement in the interview categorized using Rosenberg’s system. The interviews were conducted by two interviewers; one interviewed Annie (eight years old) and the second interviewer interviewed Adam (sixteen years old). It was a semi structured questionnaire and the questions were based on studies from Harter (1983) and Rosenberg (1979). The data indicated that indeed there is a developmental progression in the view of self, as Rosenberg describes, moving from the ‘outside-in’ to the ‘internal and inside-out’, perhaps reflecting a growing experience of oneself and a larger capacity to analyse abstract issues like one’s own emotions, character etc.
How people see themselves in relation to others and how others see them? How this ‘sense of identity’ get formed and is there a developmental path which is followed? What is identity and how and when it is formed? Rosenberg’s work addressed these questions and sought to uncover the ways in which individuals’ identities develop over time. Understanding the developmental pathway for identity formation, and how identity changes with time, may help us to understand the mechanism of identity formation and how it interacts with the development of more directly measureable cognitive abilities. This study seeks to verify observations already made by Rosenberg that as a child ages his sense of identity seems to shift from externally verifiable facts (e.g. height) to more abstract concepts such as character, aspiration and emotional makeup. The study took the form of a textual analysis of two semi-structured interviews with an 8- and a 16 year old, with the statements in each interview categorized as either Physical (externally-verifiable facts), Character (who they think they are now), Relationships (how they relate to other people) and Inner (why they think they are that way).
A semi-structured interview design was employed with open-ended questions. The structure of the questionnaire was divided in ‘five blocks’ as follow:
1- the participants wrote down their chosen descriptions of themselves as a list of sentences;
2- the interviewer explored further the statements provided by the participants;
3- the interviewer questioned the participants about their ‘self and others’, questioning their feelings of distinctiveness as separate individuals;
4- The ideal self, where the interviewer questioned the type of person the participant would like to become;
5- Questions about the participant locus of self- knowledge
The two participants that took part in this experiment were native English speakers. The age of the participants was eight ‘ Annie (female) and sixteen ‘ Adam (male).
A A4 paper sheet containing a list of ten numbered sentences starting with ‘I’ which answer the question ‘who am I’ was given to be filled by the participants. Standard instructions were used and all participants’ parents completed a consent form (I don’t have a copy).
The participants were pupils of two different schools, one primary (Annie, eight years old) and one secondary (Adam, sixteen years old) in the Milton Keynes area. They were identified by the school as children who would be willing to be recorded talking about themselves. All parents from the participants who agreed to take part in the research signed a consent form.
At the beginning of the interview, the interviewers ‘ Kieron Sheehy (who interviewed the eight years old group) and Peter Barnes ( who interviewed the sixteen year old group) ‘ said to the participants that the recording could be stopped at any time if they found themselves saying something that they didn’t wish to be recorded; none did so. They also did a brief introduction to the objective to the ‘talk’, explaining what would happen in the interview.
The interviews took place at the children’s schools, during the school day. The eight year olds interview happened in a building adjacent to their village school, where they regularly went for PE lessons. These children were accompanied by a classroom assistant. The sixteen year olds interview happened in a room opposite to the school’s secretary office.
The interviews took place in May 2005. Apart from the interviewer and the participant, there were a recordist and a producer in the room.
The data collection was carried out within the BPS ethical code and principles.
The research hypothesis is that there is a developmental progression in the view of self. Younger children develop a good understanding of their basic characteristics and their experiences and actions. Later development consists of acquiring more complex understandings of identity.
Category Percentages for Annie Percentages for Adam
Physical 60% 11%
Character 20% 22%
Relationships 20% 22%
Inner – 44%
As can be seem from the table above, the results of the textual analysis in this case are consistent with Rosenberg’s hypothesis. Indeed there is a shift from the Physical observations of self towards the other categories ‘ notably the reflections on inner emotions and the reasons for being a certain way.
The experimental procedure in this case had a number of shortcomings. The main ones in this case are the lack of a statistically significant number of examples (a whole developmental pathway cannot be reliably inferred from interviews with just two subjects) and the subjective nature of the textual analysis. While the former shortcoming could be overcome by simply increasing the sample size, the latter problem is more subtle and potentially harder to overcome. For example although individual phrases can be categorized, overall sentences may contain multiple phrases with different categorizations, and in this case it is up to the analyst to judge which is the dominant category. Since I was aware of the ages of the children (and indeed it could be inferred from the complexity of their vocabulary) I could inadvertently introduce a bias to confirm or deny the hypothesis. While this potential bias may be soluble for example through the use of multiple analysts and by breaking the text into segments and mixing those from each interviewee prior to analysis, less tractable problems remain. For example it may be that children get better at constructing long and complex sentences as they grow older and that for some reason long sentences are more likely to be categorized as ‘inner’ than shorter ones; in this case the change in categorization would not reflect an evolving sense of self but a growing linguistic ability.
In conclusion this research, despite seeming to support Rozenberg’s developmental model for the sense of identity, lacks statistical significance and may contain biases in the analytical method which make it difficult to infer the underlying reasons for the observed changes in how individuals report their reflections on themselves.