This study examined associations between student- (N = 1414), classroom-, and teacher (N = 91) growth orientation and mathematics engagement and achievement using a multilevel model. Results ...demonstrated positive associations between student growth orientation and mathematics outcomes and positive associations between classroom and teacher growth orientation and mathematics achievement. There was also a marginally significant interaction effect between classroom and teacher growth orientation on classroom engagement. Findings hold relevance for educational interventions surrounding growth constructs and teachers’ personal growth orientation.
•Previous research has examined students’ underlying growth orientation (GO), encompassing growth mindset and growth goals.•This is the first examination of GO in students and classrooms (with teacher information).•Findings demonstrate there were associations between GO and mathematics outcomes at the student and classroom-levels.•At the classroom-level, findings demonstrate both adaptive and potentially maladaptive components of teachers’ GO.
•Changes of key ecosystem services in Zhangye were analyzed from 1990–2015.•Multilevel modeling was applied to examine driving factors of ecosystem services.•Scale dependence of human and natural ...factors at different levels were concerned.•Effects of natural and human factors on different ecosystem services varied.
Clarifying driving mechanisms of ecosystem services changes is key to sustainable ecosystem management. This study evaluated three key ecosystem services, including carbon sequestration (proxied by net primary production, NPP), water yield, and soil conservation in Zhanyge during 1990-2015, and applied multilevel modeling to examine the impacts of sub-watershed level natural factors and county level human factors on ecosystem services changes. The results showed the ecological vulnerable Qilian mountain area possessed high values of these ecosystem services. During 1990-2015, NPP and water yield increased, and soil conservation decreased. Results of the multilevel models indicated that natural and human factors drove ecosystem service changes in various degrees and directions and it was necessary to consider the scale dependence of the driving factors at different levels. Particularly, forest land, grassland and cultivated land expansion led to increase of NPP, water yield and soil conservation, grain productivity and agricultural machinery level had negative effects on water yield, while had positive impacts on soil conservation. Overall, it is necessary to strengthen ecological protection in the Qilian mountain area and strictly control cultivated land expansion balance to maintain sustainable development of ecosystem with comprehensively considering the effects of driving factors and local conditions.
This article proposes multilevel modelling to account for individuals’ contexts when predicting police confidence. It uses the case study of Brazil considering 107 cities from 25 states to assess ...four evidence and literature-based predictors impacting police confidence measures in the country. The article found that being a crime victim, experiencing police corruption and having low interpersonal trust were negatively and significantly associated with confidence in the police. Levels of variance between individuals grouped by cities and states were a considerable explanatory feature of police confidence. The implications of these results are discussed in relation to police governance.
In this paper, a science literacy and technology course has been designed and implemented to strengthen the national initiative empowering scientifically literate Indonesian society. This paper is ...intended to evaluate to what degree non-science undergraduate students can perform this course. The diverse background of non-science students who participated in this study led to the challenge to evaluate their performance more comprehensively contemplating the nested structure of students’ department and faculty setting. In light of the hierarchical nature of the student data, multilevel modelling was used to conduct the analysis. The first level of analysis involved students’ performance and affective attributes measured using demonstrated science literacy assessment (SLA-D) and motivational beliefs (SLA-MB) respectively. Then, the subsequent level of analysis comprised demographic factors gathered from the institutional record. Findings demonstrated that the impact of demographic factors on the students’ performance of science literacy was not substantial. Different settings of students’ department and faculty level drove the association between affective factors and the learning process toward science literacy courses substantially. The multilevel approach controlled the equitable student assessment within the nature of students’ data structure. This paper suggests an implication of advancement regarding educational data analysis and examines the effectiveness of science literacy courses for higher institutions specifically for non-science majors.
In this study, the effects of some student (e.g., gender, bullying, etc.) and school variables (e.g., emphasis on academic achievement, clarity of teaching, etc.) on the TIMSS 2015 science and ...mathematics achievement of eighth grade students in Türkiye were examined by controlling for the socioeconomic status of the students at the student and school level. The analyses were performed using the multilevel modelling method and the HLM8 package program. The findings show that school variables account for 34% of the variability in the TIMSS 2015 science achievement of eight grade students, while student variables account for 66%. Similar to this, school variables account for 35% of the variability in these students' mathematics achievement and student variables for 65% of it. The socioeconomic status of the school at the school level and students' confidence in learning the lesson at the student level are the two variables that have the strongest effects on students' achievement in science and mathematics. According to the results, other variables that have a significant effect on students' achievement in both science and mathematics at the school level are the clarity of teaching, the emphasis on academic achievement, and the school bullying level. Furthermore, school discipline problems have an effect on students' mathematics achievement. However, school resources and teacher qualifications do not have a significant effect on student achievement. Home educational resources and bullying among students are two important variables that effect how well students do in science and mathematics. The effect of gender and value learning the lesson on science achievement was significant, whereas the effect on mathematics achievement was not. The effect of like learning lesson on student achievement is significant only for mathematics.
By taking into account the functions of socially shared metacognitive regulation (SSMR) (i.e. the role that SSMR plays in the (dis)continuation of ongoing interaction), the present study sheds light ...on the differential effectiveness of SSMR. The study more particularly investigates how individual students' involvement in SSMR that confirms, changes, activates, or stops interaction, predicts their understanding of the learning content on the short and middle-long term, taking into account students' prior knowledge. Sixty university students were involved. Multilevel models were run to investigate the relation between individual students' engagement in the functions of SSMR and their conceptual understanding. Contributing to SSMR that changes and that activates collaborative learning appeared significantly positively related to students' immediate understanding of the learning content, whereas participating in SSMR that confirms or that stops ongoing interaction was not. Contributing to SSMR (regardless of its function) appeared not significant for predicting students’ conceptual understanding on the middle-long term.
•Shared metacognitive regulation promotes students' immediate understanding.•The functions of shared metacognitive regulation determine its effectiveness.•Shared regulation that changes or activates collaborative learning is important.•The effectiveness of shared metacognitive regulation is not sustainable.
Abstract The term ‘left behind’ has come to connote political disaffection, alongside social and spatial inequalities in wealth and opportunity. Yet the term is also widely contested, often ...prioritising a regional and economic perspective at the expense of a more local and nuanced approach. In response, we argue that neighbourhood context is integral to understanding and identifying ‘left behind’ places. Building a neighbourhood classification of ‘left behindness’ for England, we evaluate the extent to which the neighbourhood trajectory contributes to our understanding of a range of multidimensional individual‐level outcomes. Our findings reveal a geography of neighbourhoods that are systematically disadvantaged over time, concentrated in major urban conurbations, and post‐industrial and coastal towns. The magnitude and impact is highlighted through poorer economic, health, social and political outcomes for those living in ‘left behind’ areas.
The role of gender in career mobility is a major practical and scholarly concern. Drawing on boundaryless career literature and social role theory, we examined whether gender influences employees' ...psychological mobility, whether this influence varies depending on the nature of career boundaries (the boundaries of job, organisation and industry) and whether it is contingent on organisational or occupational characteristics. We conducted cross‐classified multilevel analyses on 3,527 Australian employees nested within 725 organisations and 43 occupations. The results suggest that females show higher mobility preferences than males when it comes to crossing industries but not changing organisations or jobs. However, their preference for crossing career boundaries is significantly reduced when their organisation has a strong presence of female leadership. We also find that in female‐dominated occupations, females show a higher cross‐organisational mobility preference than males, while in male‐dominated occupations, females show a lower cross‐organisational mobility preference.
The literature of spatial inequalities is currently “fragmented” across ethnic segregation and built environment domains. Inequalities in these dimensions are often considered in isolation one from ...another. For example, ethnic segregated areas are often more disadvantaged in terms of unemployment, housing conditions and access to services. By using a combination of the most recent Census and a series of openly available datasets related to ambient urban environment characteristics at the Lower Super Output Area (LSOA) for the 12 biggest cities in England, we employed a series of multilevel models to explore the within and between city variations in the relationship across ethnic segregation, and key socioeconomic and built environment features of neighbourhoods. The results showed that ethnic minority segregated areas are associated with distinct and remarkably systematic patterns of socioeconomic characteristics, but do not lack access to good quality schools and public transport compared to the predominant white population communities. Ethnically segregated communities seem to consistently experience higher unemployment, long-term illness, higher average household size, and high exposure to pollution compared to white population groups across English cities. These relationships are remarkably similar across ethnic groups evidencing the extent of systematic persistent disadvantage experienced by minority communities relative to white populations. Moreover, the strength of these relationships varies widely across cities, although the overall systematic pattern prevails.
•Ethnically segregated areas in England are associated with systematic patterns of socioeconomic characteristics.•Access to good quality schools and public transport does not differ between ethnically segregated and predominant white population communities.•Ethnically segregated communities experience higher unemployment, long-term illness rates, average household size, and exposure to pollution.•The strength of these relationships varies widely across cities, although the overall systematic pattern prevails.
Under examination Owen, Gwilym; Harris, Richard; Jones, Kelvyn
Progress in human geography,
06/2016, Volume:
40, Issue:
3
Journal Article
Peer reviewed
Since the 1990s, multilevel models have become popular tools for looking at contextual effects upon health. However, the way that geography is incorporated into these models has received criticism ...due to somewhat arbitrary definitions of what counts as context, the models’ discrete and, arguably, aspatial view of geographical effects, and the lack of any clear theoretical specification of the processes involved. This review draws together and extends these criticisms, arguing that while currently there are problems with how geography is conceived within multilevel models, there are ways of addressing them, and indeed that it is important to do so.
