Multilevel Multivariate Regression: An Illustration Concerning School Teachers’ Perceptions of their Pupils

Survey data on teachers’ perceptions of the ability and of the economic circumstances of their pupils are used to illustrate multilevel multivariate linear regression. The main substantive conclusions are that (i) on average, teachers in a school share beliefs about the pupils, and that these beliefs form a fairly coherent view in which economic disadvantage is seen to go along with low ability, but that (ii) individual teachers within the school vary substantially around this average. These conclusions have a bearing on the debate about school ‘ethos’. The main methodological conclusion is that investigating such topics can be made more subtle and more able to respect the complexities of educational structures by adopting both a multilevel and a multivariate approach, and that computer software now exists to achieve this. Résumé Les données d’une enquête sur les perceptions des enseignants relativement aux capacités de leurs élèves et à leur appartenance socio-économique ont été utilisées ici pour illustrer l’utilisation de la méthode de régression lin?aire multiple sur plusieurs niveaux. Les conclusions principales sont les suivantes: (i) en moyenne, dans une école, les enseignants échangent sur leurs croyances par rapport à leurs élèves, et l’on peut remarquer une grande cohérence dans le fait de croire qu’être issu d’un milieu économique désavantagé va de pair avec des capacités moins grandes; (ii) mais on peut constater que le point de vue des enseignants — pris individuellement — s’écarte considérablement de cette moyenne. Ces conclusions alimentent le débat sur le “génie” (“ethos”) à l’école. Sur le plan méthodologique, on peut principalement conclure au bien-fond? du choix d'une approche … la fois multiple et sur plusieurs niveaux dans le cas d'investigations, telles celle-ci, portant sur des thématiques nécessitant beaucoup de finesse et de respect envers la complexit? des structures de l’éducation. Relevons encore que le software est actuellement disponible pour effectuer des démarches de ce type. (Traduction: Catherine Cusin, Collaboratrice scientifique du CSRE, Aarau, Suisse.)     

Social class,gender and secondary education in Scotland in the 1950s

Newly accessible data from the Scottish Mental Survey 1947 are used to investigate the legacy in the 1950s of reforms to Scottish secondary schooling in the first four decades of the 20th century. These reforms had sought to extend opportunity for post‐primary education beyond the children of the professional middle class who had formed the largest part of the clientele of older secondary schools. The reforms also provided unprecedented opportunities to girls. The findings of the analysis are that the widening of opportunity was maintained in the stable, selective system of the 1950s. A broadly homogeneous system of academic secondary education had been created for about 45% of the age group, by means of selection at age 12. Social class continued to influence pupils’ progress in this secondary sector, but not to any different extent in the older or the newer schools. Progress in secondary school was not related to gender except in science.     

转移方法和推迟分析对酶活性的影响

目的本研究的第一个目的是模拟在血液收集和分析场所之间针对样品转移可选择的方法,第二个目的是对比同一样本在不同的天数时所得的分析结果.方法26名健康男性(年龄20～50岁)志愿提供血液样本.血液离心后得到的血清,被分成7部分.每一部分都被单独处理以模拟不同的转移方法和存储条件.随后分析每一部分中的乳酸脱氢酶(LDH),乳酸脱氢酶同工酶1(LDH-1),肌酸激酶(CK)和肌酸激酶心型同工酶(CK-MB)的活性.结果CK和CK-MB在基础值和其他样本中所得结果没有显著性差异(P＞0.05);LDH和LDH-1也有相似的结果,只是在模拟不带冰转移时,其结果显著高于基础值(P＜0.05).结论采用带冰的容器转移血液样本,然后冷冻.这样,即使推迟5天测量,其结果也不会受到影响.     

Objectivity as an Educational Imperative

This paper examines the effects of the broader social, political and economic context on participation in state‐funded adult literacy programmes. Using the state as the unit of analysis it employs both a cross‐sectional and longitudinal research design to determine the impact of state characteristics on aggregate adult education participation rates.