Conclusion and Discussion. In this report, we have examined the negotiations between the digital labour platform Hilfr and the union 3F as well as the content of the company agreement they concluded in 2018. Our analysis is based on nine qualitative interviews with the negotiating parties, workers and government officials as well as supplementary desk research. It demonstrates that it is possible to use collective agreements as an instrument to improve wage and working conditions in the platform economy as well as the retention of workers on the platform – but not without challenges. Below, we summarize the main findings and discuss their implications for regulating the platform economy via collective agreement within the context of the Danish model. Both Hilfr and 3F wanted to create orderly conditions in the platform economy. They succeeded in reaching a company agreement with comparable wage levels to other parts of the labour market as well as including a number of social benefits, we typically find in sector-level agreements, e.g. pension, sick pay etc. However, when looking closely at the agreement text, it becomes evident that the overall level of working conditions is somewhat lower than in comparable sector-level agreements in private services. This is unsurprising, as most start- up companies are not able to meet the overall level of current sector-level agreements. Furthermore, the agreement is explicitly designed to be a staircase model, where future negotiations (including the ongoing re-negotiations) aim to lift the level of wage and working conditions gradually. A number of contextual factors seem to have played an important role for paving the way for such a novel type of company agreement: • Social responsibility as a clear business strategy: Hilfr had a clear business strategy to differentiate themselves in the market for cleaning platforms as socially responsible. • Cleaning in private households as an unregulated market. Cleaning in private households is less regulated than other parts of the Danish labour market (Mailand and Xxxxxx, 2020), which made it an attractive business opportunity for Hilfr and other digital labour platforms. From a union perspective, it made it easier to justify a staircase agreement where the overall level of wages and working conditions start at a lower level than comparable sector-level agreements. • Tripartite collaboration as a facilitator. The government-led Disruption Council supported the dialogue between the involved bargaining ...
Conclusion and Discussion. The process of secularization in the Netherlands over the last 50 years—in terms of the decline of consistently religious Dutch people and the rise of secular Dutch people—is clear. Our research question has an explanatory nature: To what extent can secularization in the Netherlands between 1966 and 2015, in terms of the decline of institutional Christianity, be explained from a modernization theory perspective, unfolded in terms of period and cohort effects, next to changing individual effects? We have set out to investigate a long-term process, with 50 years of high-quality data, and to distinguish between different kinds of effects. After explicating modernization theory in its partial explanations and derived specific hypotheses, we have tested these in the Dutch case of secularization very extensively. Let us review and interpret the impact of three instances of modernization (of ideologies, of economics, and of social ties) and relate these to the findings of previous research. Our first theoretical perspective explaining secularization is the modernization of ideologies. The impact of the educational level of individuals on their religiosity has been tested previously; however, with mixed results (e.g., Xxxxxx and Xxxxxxxxx 2004; Xxxxxxx et al. 2012; Xxxxxx and Xxx Xxxxxxxx 2009; Xxxx and XxXxxxxx 2012). In this article, we provided evidence on the national level: educational expansion as a period and cohort effect has played a significant role in the Dutch process of secularization. Through educational expansion between 1966 and 2015, high levels of education have become mainstream and dominant in Dutch society. It seems that the accompanying scientific rationalism is increasingly shared nationwide, which has had a negative impact on the religiosity of the Dutch population and, moreover, specifically on younger people in their formative years. Concerning the modernization of economics, Dutch people with high incomes have become increasingly more likely to be secular over the last 50 years. A main focus of scholars is on macro-level social securityinduced secularity. The period effect of the development of social security was not significant in the Netherlands: the general increase of social security over the last 50 years has not increased secularity among the Dutch population. This is opposite to other studies (Xxxxxx and Xxxxxxxxx 2004; Xxxxxxx et al. 2012; Xxxxxx and Xxx Xxxxxxxx 2009), which show that social security in Western countries as a p...
Conclusion and Discussion. Results of the study showed that the PACER Test had high reliability for testing the maximum volume of oxygen among male and female students aged 13 to 15 years. The findings of this study are similar with research findings by Xxx et al. (1992) who reported multiple reliability for the PACER Test via the test and retest approach, attaining .89 intra-class reliability value for 139 male and female subjects aged between 6 to 16 years (Xxxxx, Xxxxxxx, Gadoury, & Xxxxxxx, 1998). As for the Mile Run/Walk Test, the findings showed that only 57.8% of male students aged 13 years passed and succeeded in completing the run within a time range of 7.00 to 10.00 minutes, 50.9 % male students aged 14 years passed and succeeded in completing the run between 7.00 to 9.30 minutes, and 33.2% of male students aged 15 years passed and succeeded in completing the run within a time range of 7.00 to 9.00 minutes, the times stipulated by the FITNESSGRAM Standard. By contrast, the percentage of passes among female students aged 13 years who succeeded in completing the run within a time range of 9.00 to 11.30 minutes was 65.1%, 55.2% female students aged 14 years succeeded in completing the run within a time range of 8.30 to 11.00 minutes, 52.9% female students aged 15 years succeeded in completing the run within a time range of 8.00 to 10.30 minutes, the time determined by the FITNESSGRAM Standard. The results of the study showed that 15 year-old male students’ cardiovascular resistance was very weak as 66.8% of they failed to achieve the passing level determined by the FITNESSGRAM Standard. As for the Pacer Test, the results of the study showed that only 34.2% male students aged 13 years passed and succeeded in completing the run between 41 to 72 repetitions. 39.8% of male students aged 14 years passed and succeeded in completing the run between 41 to 83 repetitions, and 21.4% of male students aged 15 years passed and succeeded in completing the run between 51 to 94 repetitions based on the repeated runs stipulated by the FITNESSGRAM Standard. Meanwhile, all female students aged 13, 14, and 15 years passed and succeeded in completing the run between 23 to 51 repetitions. The result of the study showed that more than 60% of male students aged13, 14, and 15 years failed to pass the level stipulated by the FITNESSGRAM Standard. As for the One-and-a-Half Mile Run or Walk Test, the results of the study showed that only 69.9% of male students aged 13 years, 74.2% male students aged 14 ...
Conclusion and Discussion. We investigated the extent to which three postgraduate teacher education institutes in the Netherlands pay attention to and aim at stimulating the development of community competence. This question was approached through three curriculum representations. It appears that in the intended curriculum community competence is found important, but in the implemented and especially the attained curriculum, the development of community competence receives less attention. Looking at the intended curricula of the teacher education institutes, we found that the development of community competence was considered to be an important topic in the programmes. The study guides revealed that all institutes in some way or other stated the importance of developing community competence by their student teachers. This is in line with the descriptions provided by the teacher educators and heads of department, which show that almost all deemed it important for student teachers to develop community competence. At the same time, community competence was weakly conceptualised within the study guides. This weak conceptualisation was also apparent in the implemented curriculum, where the importance denoted by teacher educators in the intended curriculum was not systematically reflected in their own descriptions of their actions. Teacher educators reported that they paid attention to community competence in the sense that they organised different collaboration activities. At the same time, only a few teacher educators said they stimulated reflection on the development of community competence. Most teacher educators believed that community competence was adequately developed by taking part in collaborative activities. Additionally, most teacher educators stated that community competence was not given explicit attention within the assessment procedure, and for two teacher educators a certain minimum level of community competence was not necessarily a requirement for receiving the teacher's certificate. This lack of systematic assessment of the development of community competence is probably related to the fact that community competence was weakly conceptualised in the study guides. From observations and examination of the electronic learning environment we found that there were many differences between the teacher educators in how they implemented community competence both face-to-face and in the electronic learning environment. In the meetings, some teacher educators played a very active...
Conclusion and Discussion. Our key contribution has been to examine the choice between horizontal merger and RJV as alternative vehicles for exploiting R&D complementar- ities, whereas most existing contributions examine just one of those forms of horizontal agreement in isolation. An innovation in our modelling struc- ture is that we have allowed for the possibility that RJV contracts might be unenforceable. Our key positive findings are as follows: • For the two insiders, RJV always dominates no-agreement, so some form of horizontal agreement (merger or RJV) always arises in equi- librium. • Merger/RJV tradeoff: While merger offers certainty that R&D com- plementarities will be successfully exploited, it leads to an aggressive and profit-reducing reaction by outsiders on the product market. Rises in both brand similarity and contract quality favour RJV over merger. • A rise in the degree of R&D complementarity makes merger “more likely” relative to RJV. However, RJV may continue to be chosen even with perfect R&D complementarity. • Surprisingly, the two insiders may optimally choose to merge even when RJV contracts are always enforceable (this requires a low degree of brand similarity); and they may optimally choose to form an RJV even when the probability of contract enforceability is negligible (this requires a high degree of brand similarity). 38In particular, one might add a social indifference locus to a diagram like Figure 2. How do our results relate to the BP/ARCO case study that motivated our analysis? If we assume that crude oil is a relatively homogeneous product and that the degree of R&D complementarity is high (so that both β and θ are close to 1), then we can see from Figure 3 that the insiders will, as BP and ARCO did, prefer merger to RJV when contract quality is low. Moreover, the welfare analysis in Section 5 is also relevant. There, we showed that society prefers merger to RJV when R&D complementarities are substantial but contract quality is low.39,40 Therefore, our results suggest that a merger with the characteristics of BP/ARCO might benefit both the insiders and society.41 To close, we briefly consider three possible generalisations of our anal- ysis. The first is to allow for Xxxxxxxx (price) competition on the product market. We conjecture that our qualitative results will not survive this gen- eralisation. This is because, under Xxxxxxxx competition, horizontal mergers are generally profitable for market-power reasons alone.42 Therefore, given that merg...
Conclusion and Discussion. This study evaluated whether earlier detection of relapse by Surveillance Imaging had a positive effect on survival outcomes in patients with FL achieving remission after receiving first-line therapy in two cohorts. By first comparing the estimated survival probabilities in clinical detection group versus radiographic detection group, we found that patients who had their disease relapse detected by SI did not show significant improvements in PFS, OS from diagnosis and OS from relapse in both of the two cohorts. This result was reasonable for PFS since detection methods definitely will not have effect on disease relapse. However, the longer median PFS time in SI detection group in MER cohort raised our doubt about the ability of SI for earlier relapse detection. In addition, although in the plot of OS from diagnosis in Figure 1, the radiographic detection group showed an improved survival probability compared to clinical detection group, this improvement did not appear until about 10 years of survival. However, this time is relatively long for patients with FL relapse and a 10-year survival probability of about 0.8 is considerable as mentioned in introduction part, which means SI still did not bring obvious benefit in OS from diagnosis. In single-variable and multivariable analyses based on Xxx models built for each survival outcome, although variables included in the models and corresponding values of hazard ratio were different in two cohorts, the similar thing was that no statistically significant association was observed between detection methods and survival outcomes, which also weakened the usefulness of SI. In a similar study conducted by Xxxxxxx et al, they found that for patients with aggressive NHL (DLBCL), there was no significant difference in median overall 5-year survival between those whose relapses were detected by routine imaging and those by abnormal exam results or reported symptoms.16 Our results were consistent with their findings of no significant associations between detection method and survival outcomes although they also mentioned that routine surveillance scanning helped to identify patients who might have a better outcome based on the age-adjusted international prognostic index determined at the time of relapse (sAAIPI). Moreover, our findings refined the results in Xxxxxx’x study, which also indicated no difference in survival outcomes between the two groups for both aggressive and indolent NHL, by focusing on a specific type of...
Conclusion and Discussion. Our analysis shows that during the last decades, Dutch society has strongly changed socio-culturally – and that the membership of the public broadcasting associations has changed with it. In our results, we see a clear trend towards individualization – a trend that other scholars have found as well (e.g., Sociaal en Cultureel Planbureau 1998; Felling 2004). Our analysis also showed, that the socio-cultural changes in the membership of all public broadcasting associations resemble the socio-cultural changes in Dutch society at large. In short, that boils down to the conclusion that in the time span that our study covers, the membership of all broadcasting associations became increasingly individualized. Based on our data from 1979 and 1985, Xxxxxx (1989; for comparable findings based on the 1979 data alone, see Xxxxxxx et al. 1983) argued that within the Dutch public broadcasting system three distinct groups of broadcasting associations could be distinguished: (i) confessional broad- casting associations, (ii) ‘neutral’ broadcasting associations, and (iii) left- wing broadcasting associations. According to Xxxxxx each of these three groups of broadcasting associations had memberships with a distinct value set. In our analysis we can still distinguish these three groups of broad- casting associations in 2005. However, nowadays, the Catholic KRO seems to fit in with the group of neutral broadcasting associations better than with the confessional group. Within the group of neutral broadcasting associations, the ideological positions of the membership of AVRO (liberal conservatives), TROS 607 EUROPEAN SOCIETIES (embracing popular culture), KRO (originally Catholic), as well as the ideological position of non-members, differ only modestly. The original broadcasting association of the liberal conservatives in the 1920s (AVRO) has gotten company of other broadcasting associations with like-minded memberships. Downloaded by [Radboud Universiteit Nijmegen] at 02:43 25 April 2015 In the group of left-wing broadcasting associations, the ideological positions of the membership of BNN (young hedonists), VARA (socialists) and VPRO (nonconformist libertarians) are close together. So the original broadcasting association of the socialists (VARA) has also gotten company of other broadcasting associations with like-minded memberships. The socialist pillar has not disappeared altogether from the ether. The group of confessional broadcasting associations changed strongest over...
Conclusion and Discussion. Overall, the imputation accuracy of SVD, LLS, KNN and BPCA are close, shown by both LRMSE and scatterplots between imputed value and reference value on the log scale. The performance of scImpute varies between observations with different numbers of missingness due to its feature of identifying true zeros in the dataset. The performance of scImpute is better than the other methods on true zeros. However, the accuracy of scImpute to identify the true zeros isn’t ideal, with the false discovery rate 66.3% for the positive data set and 60% for the negative data set.
Conclusion and Discussion. A. How to avoid legal disputes between China and South Asian in future? There must be proper legal corporation between China and South Asian countries harmonizing different application of legal issues in these jurisdictions. Force the governments in the region to introduce Common Directive relevant to finance, trade, trade cooperation, interconnection, non-traditional security, energy security like in EU.
Conclusion and Discussion. To summarize, our research provides five main findings: • Nature of rating tasks in social computing tend to be more subjective than objective. • Most social computing studies do not adequately report the information needed to evaluate the agreement of crowd workers. • The reported agreement scores in social computing stud- ies are not high, averaging at around 0.60 for both Kappa and Alpha metrics. • The more subjective the task, the worse the agreement tends to be, regardless of the used metric. • Different metrics penalize differently compared to ob- served agreement - Fleiss’ kappa seems to penalize relatively much, while Gwet’s gamma penalizes relatively little. Therefore, using Gwet’s gamma, or even the simple percentage agreement might be appropriate. Three key points emerge for discussion. First, even if the inter-rater score is low, the crowd raters may still be reliable, i.e., giving honest responses. A practical challenge then is how to identify between the two, determining when the disagreement is true due to bad guidance or rater quality and when due to the subjectivity of the task. Second, the commonly used metrics, namely Fleiss’ kappa and Krippendorff’s alpha, use chance-adjustment. However, there is no “chance” involved when the rater is giving his or her true opinion on a subjective matter. In fact, using chance- adjusted reliability scores is misleading in this case as they incorrectly bias away from true values. Coincidentally, this leads us to venture further away from perceiving crowd as an anonymous mass of people, as it was originally intended [35], into perceiving crowd workers more as opinionated indi- viduals. Such a step could be useful for computational social scientists, even though it requires developing new sampling strategies and finding out more about the crowd raters. Finally, although most researchers did report the basic numbers, from the reviewed papers, we conclude that crowd labeling is not always properly reported by researchers. This hinders replication and evaluation of the use of crowd annota- tors. Specifically, only a few studies reported category-specific agreements despite encouragements in the prior literature to do so [7]. Future research utilizing crowd raters should (a) report the sources of disagreement, and (b) set the goal for agreement score according to the level of task subjectivity. ACKNOWLEDGMENT The authors thank Xx. Xxxxxxxx Xxxxx from Qatar Com- puting Research Institute for giving access to da...
