siapapun kalian, jadilah orang yang bermanfaat

Nasywa Nisa Insani 

Kelompok : Alfred Adler

Article title "Clustering digital learning pathway preferences from the perspectives of epistemic justification on self-regulated learning, social presence, and resources."

    Based on the above two aims, our last aim is to qualitatively compare the perceptions of the key features towards digital learning by people of different clusters. Almost all previous similar studies were conducted in general context, i.e., using Internet-based epistemic beliefs self-reported instrument to show their associations with other factors such as learning behaviours and self-regulation . However, peo­ple could hold different beliefs in different context of DLPs.

    Convenient sample participants were invited to participate in an online survey during the COVID-19 pandemic in the end of January in 2020. Two authors posted a survey link that was created on an online survey service platform on their personal social media platforms to call for people to participate. The online survey consisted of three components: consent form, demographic information , and DLP choices for a formal learning in a digital context.

    After completing the online structured survey, participants were invited to respond to a list of open-ended questions. The questions were about their digital learning opinions, motivations, experiences and learning perceptions . DLPs considered a person-centred approach that can identify how individuals can be classified according to a set of variables .

    Based on survey interviewees’ responses on the open-ended questions, a qualitative content analysis with conceptual analysis and relational analysis was conducted. Ipants’ experiences, opinions, and perceptions towards digital learning. Second, the contents of opinions, experiences, activities, and behaviours that were repeatedly mentioned were coded . Third, after all the contents were coded, the codes with similar meaning would be merged as a concept or an attribute of a concept based on the coding scheme .

    For each coded segment above, sentiment analysis was conducted to label them into positive perception, neutral perception, and negative perception. Sentiment analysis enables us to identify the general sentiments expressed in text data. In this study, BytesView data analysis tool was applied. BytesView is one of the most effective and easiest ways to get quantitative-oriented sentiment result from unstructured text data. It supports multiple languages, including Chinese.

Latent class analysis based on DLPs beliefs

    As can be seen from Table 4, the statistics of Bayesian Information Criterion , Akaike Information Criterion , entropy, and the Lo-Mendell-Rubin Likelihood Ratio Test were assessed to reflect the model fit.

    Lo-Mendell-Rubin Adjusted LRT Test more were considered to be very strong evidence of model fit corresponding to the odds of 150:1 . Entropy values indicate the degree to which clusters may be considered distinguishable from one another with standardised values ranging from 0 to 1 . An entropy value closer to 1 indicates the presence of clear, distinct clusters and greater power to predict cluster membership . 

The LMR-LRT is a test of statistical significance, where the null hypothesis is the number of classes, c, estimated minus one. Chi-Square test on gender, educational background and family income Chi-Square test for independence had been conducted on the gender, educational background and family income distribution in the four clusters. For gender, it indicated significant association between gender and the clusters, 2 = 10.66, p = .01, Cramer’s V = 0.23. The number of female digital learning skeptics was significantly smaller than expected , while the number of male digital learning skeptics was significantly larger than expected.