DETERMINANTS OF PERCEIVED FLOW: FACTOR ANALYSIS

FACTOR ANALYSIS

Factor analysis using varimax rotation was performed to measure construct of perceived flow which combined three dimensions: perceived enjoyment, perceived control and attention focus. Previous research has found significant correlations among these dimensions (Huang, 2006; Koufaris, 2002). This shows that they are interchangeable and will covary with each other. They are driven by the same underlying construct as it has the same determinants and consequences. In fact, previous research has also used these dimensions as reflective indicators of flow (Siekpe, 2005; Wang, Baker, Wagner, & Wakefield, 2007). Thus, it is appropriate to integrate these dimensions into a reflective factor known as perceived flow. Table 5 exemplifies that item loadings ranging from 0.848 to 0.685 with item ‘I felt that using this mobile SNS is interesting’ has a relatively highest loading on perceived flow with Kaiser-Meyer-Olkin Measure of Sampling Adequacy value of 0.896. Results have suppressed small coefficients of absolute value below 0.50 where item ‘when using this mobile SNS, I felt confused’ was deleted as it does not load heavily to the factor.

Table 5 – Factor analysis
table5Determinants of Perceived Flow-5

MULTIPLE REGRESSION ANALYSIS

To further testing the proposed hypotheses, multiple regression analysis was performed. The level of significance (a) was set at 0.05. As evident in Table 6, perceived information quality (Pi = 0.223, p<0.05) and perceived system quality (p3 = 0.536, p<0.05) are determinant of perceived users trust, supporting all the hypotheses.

Table 6. Influence of perceived information quality and perceived system quality on perceived user trust
table6Determinants of Perceived Flow-6
Notes: *p<0.05; R2 = 0.518; b = unstandardised beta; SEb = standard error beta;
P = standardised beta

Table 7 depicts that perceived flow is a determinant of perceived information quality and perceived system quality of mobile SNS (p2 = 0.170, p4 = 0.593, p<0.05). Thus, the hypotheses are supported and 53.3 percent of variance in perceived flow is explained by the perceived information quality and perceived system quality.

Table 7. Influence of perceived information quality and perceived system quality on perceived flow
table7Determinants of Perceived Flow-7

Results in Table 8 indicate that there is a correlation between perceived user trust and perceived flow, (p5 = 0.725, p<0.05). Thus, the hypothesis is supported and 52.5 percent of variance in perceived flow is explained by the perceived user trust.

Table 8. Influence of perceived user trust on perceived flow
table8Determinants of Perceived Flow-8

Users’ loyalty is a determinant of both perceived trust (p6 = 0.198, p<0.05) and perceived flow (p7 = 0.641, p<0.05). As a result, both hypotheses are thus supported and 63.3 percent of the variance in loyalty is explained by these antecedents (Table 9).

Table 9. Influence of perceived user trust and perceived flow on loyalty
table9Determinants of Perceived Flow-9

Figure 1 illustrates a visual representation of the full model and the causal relationships tested. It depicts that all proposed hypotheses are supported and significant at the 0.05 level.
Fui\Fig1Determinants of Perceived Flow-10
Note: ““ Denotes relationship is significant at the 0.01 level Figure 1. Theoretical framework with causal relationships