Academia.eduAcademia.edu

Figure 14 – uploaded by Alexander J A M Van Deursen

See full PDFdownloadDownload figure
Factor structure CFA factor analysis

Table 14 Factor structure CFA factor analysis

Related Figures (16)

Table 1. Demographic profile (UK: N = 324; NL: N = 306).
Table 1. Demographic profile (UK: N = 324; NL: N = 306).
“The Information Navigation skill was reversed since it contained negatively worded items. Table 2. Scale characteristics (Overall N = 622, UK N= 317, NL N= 305).
“The Information Navigation skill was reversed since it contained negatively worded items. Table 2. Scale characteristics (Overall N = 622, UK N= 317, NL N= 305).
Table 4. Factorial invariance tests (Operational, Information Navigation, Social, and Creative scales). Note: All y* are significant at p <.001. This is not surprising since the factorial model is quite complex. The next step was to test whether the factor structures proposed in the previous section fit similarly in the UK and the Netherlands. We conducted CFA using AMOS with tests for factorial invariance. We tested for configural, metric, scalar, and uniqueness invariance (Table 4). Configural invariance indicates the same factor structure, Metric invariance indicates the same factor loadings, Scalar invariance indicates the same item intercepts, Uniqueness indicates the same unique error terms. For the purposes of scale construction we were interested mostly in configural and metric invariance because we needed, at the very least, the same factors to be identifiable within the Netherlands and the UK and for the items to load similarly on these different constructs. The full model including all factor structures (see Appendix 1 for coefficients and 2 for covariances and correlations) has a moderate to good fit for complex model indicators on the merged database (Kline,
Table 4. Factorial invariance tests (Operational, Information Navigation, Social, and Creative scales). Note: All y* are significant at p <.001. This is not surprising since the factorial model is quite complex. The next step was to test whether the factor structures proposed in the previous section fit similarly in the UK and the Netherlands. We conducted CFA using AMOS with tests for factorial invariance. We tested for configural, metric, scalar, and uniqueness invariance (Table 4). Configural invariance indicates the same factor structure, Metric invariance indicates the same factor loadings, Scalar invariance indicates the same item intercepts, Uniqueness indicates the same unique error terms. For the purposes of scale construction we were interested mostly in configural and metric invariance because we needed, at the very least, the same factors to be identifiable within the Netherlands and the UK and for the items to load similarly on these different constructs. The full model including all factor structures (see Appendix 1 for coefficients and 2 for covariances and correlations) has a moderate to good fit for complex model indicators on the merged database (Kline,
*y* difference significant at p <.01. **y? difference significant at p< .001. Table 3. x’ differences (df = 1) for paired construct test.
*y* difference significant at p <.01. **y? difference significant at p< .001. Table 3. x’ differences (df = 1) for paired construct test.
Table 5. Model fit on CFA for the individual factors. 2005): y* (510) = 1667.93, y*/df = 3.27; CFI = 0.93; RMSEA = 0.06 (ci. 0.06-0.06); AIC = 1977.93.
Table 5. Model fit on CFA for the individual factors. 2005): y* (510) = 1667.93, y*/df = 3.27; CFI = 0.93; RMSEA = 0.06 (ci. 0.06-0.06); AIC = 1977.93.
Table 8. CFA fit (N = 1337; weighted full population).
Table 8. CFA fit (N = 1337; weighted full population).
Table 7. Scale characteristics in Dutch Internet user population.
Table 7. Scale characteristics in Dutch Internet user population.
Table 6. Demographic profile Dutch Internet user sample (N= 1107, weighted N = 1337). GTOZ equEdss 17 Z7'10 & [BEM | TeysPawp ] Aq pepeow~moqd 4.2 Confirmatory factor analysis
Table 6. Demographic profile Dutch Internet user sample (N= 1107, weighted N = 1337). GTOZ equEdss 17 Z7'10 & [BEM | TeysPawp ] Aq pepeow~moqd 4.2 Confirmatory factor analysis
Note: Base. Dutch Internet Users (N = 1337). Table 9. Reliability (a) in different groups. socio-demographic groups, (2) testing for convergent and discriminant validity of the scales, and (3) testing whether the scale characteristics were consistent through random resamples of the population using bootstrapping techniques and whether they relate simi- larly for different socio-demographic groups.
Note: Base. Dutch Internet Users (N = 1337). Table 9. Reliability (a) in different groups. socio-demographic groups, (2) testing for convergent and discriminant validity of the scales, and (3) testing whether the scale characteristics were consistent through random resamples of the population using bootstrapping techniques and whether they relate simi- larly for different socio-demographic groups.
Table 11. Convergent and discriminant validity indicators skills scales.
Table 11. Convergent and discriminant validity indicators skills scales.
Table 12. Correlations between short scales in population survey of Dutch Internet users (N = 1337).
Table 12. Correlations between short scales in population survey of Dutch Internet users (N = 1337).
Table 10. Factor correlation and AVE? (on diagonal).
Table 10. Factor correlation and AVE? (on diagonal).
Note |: Italic items are added to create longer scales; depending on the aim of the research project these can be used to replace those on the short scale. Note Il: There is also a set of Critical (literacy) skills that are not included because they were shown to be individual context dependent and not easy to measure in general population survey research. Note III: The Information Navigation items are all negatively formulated. We recommend that future research use positively formulated items measuring the same skills. Table 13. Proposed items and factors to measure Internet skills.
Note |: Italic items are added to create longer scales; depending on the aim of the research project these can be used to replace those on the short scale. Note Il: There is also a set of Critical (literacy) skills that are not included because they were shown to be individual context dependent and not easy to measure in general population survey research. Note III: The Information Navigation items are all negatively formulated. We recommend that future research use positively formulated items measuring the same skills. Table 13. Proposed items and factors to measure Internet skills.
Construct and error term covariates and correlations CFA
Construct and error term covariates and correlations CFA
Appendix 2 Continued.
Appendix 2 Continued.
Related topics:
Digital DivideInformation LiteracyMedia LiteracyDigital InequalityInternet SkillsICT and Internet Literacy Skills
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved