Previous research has documented that social interactions are central to adolescents' experience (Hartup & Stevens, 1997; Youniss & Smollar, 1985). For adolescent girls in particular, as they age they perceive their relationships to be more intimate than during previous developmental stages (Buhrmester & Furman, 1987; Laursen, 1996). Beyond the qualitative sense that their relationships are closer as demonstrated through questionnaires, adolescents' emphasis on peer group is also apparent in studies that have incorporated more objective measures. Specifically, by using the Experience Sampling Method (ESM), Csikszentmihalyi and Larson (1984) demonstrated that adolescents spend three times more time with their peers than with their parents.
Although previous studies have attempted to examine adolescent relationships, in general, such work has been limited by the types of methods that have been used to capture the daily experience of adolescents. Traditionally, researchers have relied on self-report methods to ascertain the social experience of adolescents. Since questionnaires provide an easy way to gather information about participants, this leads to their extensive usage in multiple fields. However, researchers can experience problems when they attempt to draw conclusions based solely on this method. Social desirability effects, as one example, have confounded research that has relied on this method. Specifically, participants may be unwilling to report certain information or may report such information inaccurately in an attempt to present themselves in a particular manner. As a result, these social desirability effects may compromise the integrity of conclusions derived primarily from relying on self-reports. For example, past studies have demonstrated that adolescent boys report lower levels of intimacy and closeness in their friendships than do adolescent girls (Buhrmester & Furman, 1987). However, because societal stereotypes characterize intimate friendships as feminine, adolescent boys may underestimate the closeness of their relationships in an attempt to conform to societal expectations.
Since this finding underscores distinct gender differences, it is unclear whether females and males use the same definitions when completing self-reports. For example, when asked about the closeness of their friendships, they may have different definitions of "a close friend" which would greatly influence the use of such self-reports. Although this problem has not been adequately addressed in adolescent research, it has been documented by researchers in the sensory sciences in particular. Previous studies on taste have demonstrated that when participant groups have different definitions for a particular concept, across-group statistical analyses may become invalid. Such comparisons may underestimate or even reverse observed differences (see Bartoshuk, 2000 for a critical review). As a result, developmental researchers need to rethink incorporating across-group statistical analyses in their experimental designs with both genders to avoid such potential problems.
Recent research has addressed some of the inherent problems of self-reports through the use of more qualitative methods. Specifically, research has begun to incorporate narrative analysis as a way of targeting the subjective experience of the adolescent (McLean, 2005). Typically, narrative analysis provides researchers with information that cannot be obtained through questionnaires. Previous research has examined the thematic content, word usage, motivations, and construction of narratives (McAdams & Losoff, 1984; McLean & Thorne, 2003; Pennebaker, Mehl, & Niederhoffer, 2003).
The use of narratives in psychological research has advantages over studies that rely on self-report measures for a number of reasons. First, analyzing the content of the narratives provides rich, descriptive information that would normally be difficult to obtain with a questionnaire. …