Synchronizing the asynchronous by anchoring interactions: Examining the effects of anchor media type and learner control of anchor selection on information searches within a hypermedia kiosk

Date of Completion

January 1998


Education, Educational Psychology|Education, Technology of|Education, Curriculum and Instruction




This dissertation incorporates what many describe to be competing metaphors of learning: (1) the acquisition metaphor (AM); and (2) the participation metaphor (PM), (Sfrada, 1998). The AM (Bereiter, 1991) was used to guide the systematic programming efforts and the PM guided the design of navigational and display features (Young, 1994).^ Participants' log-files (Lawless & Kulikowich, 1994) detailing their hypermedia navigation while problem solving were seamlessly collected for analysis (Barab, Bowdish, Young & Owen, 1996; Barab, Bowdish & Lawless, 1997). A theoretical framework grounded in Kadar and Effkin's (1994) ontological extensions to Gibson's (1986) ecological approach to cognitive psychology was used to interpret results.^ Ninety-six undergraduate and graduate educational psychology students took part in two separate experiments (Exp 1 & Exp 2). Their ages ranged between 18 and 37 years. Forty-eight participants (38 female 79.2% & 10 male 20.8%) were randomly assigned to one of four treatment groups (n = 12) in Exp 1. Forty-eight (n = 34 or 70.8 percent female and 29.2 percent n = 14 male) participants were randomly assigned to two groups: (1) program-control and (2) learner-control in Exp 2.^ Both experiments used MANOVA and stepwise DFA design (Tabachnick & Fidell, 1989) to investigate the effects anchor presentation type, anchor problem complexity, and learner-control have with respect to participants' goal-directed information searches. Multiple regression was used to investigate whether proposed aptitude-treatment interactions (ATI) between prior knowledge and performance yielded predictable results (Gray, 1986; Fry, 1972; Jonassen, 1982).^ Results suggest that a multivariate composite of the three DV's were significant predictors of group membership or goal adoption (1-Wilks =.34 & 1-Wilks =.39). Participants' Total Time (F(1,44) = 7.94, p =.007 & F(1,46) = 13.39, p =.001) and Solution Scores (F(1,44) = 7.94, p =.007 & F(1,46) = 19.77, p $<$.001) provided statistical power in each experiment. Whereas, participants' Attitudes Toward the Content (F(1,44) = 1.68, p =.202 & F(1,46) = 0.06, p =.811, respectively) did not. Prior knowledge was not a significant predictor of participants' Solution Scores in either experiment (R$\sp2$ = 0.29 and R$\sp2$ = 0.11 respectively) therefore no ATI was detected. ^