中国科学软件网-首页
HLM 7——分层线性模型分析软件

在社会研究等领域,研究数据往往具有层次结构。也就是说,单独研究的课题可能会被分类或重新划分到具有不同特性的组中。在这种情况下,个体可以被看成是研究的第一层(level-1)单元,而那些区分开他们的组也就是第二层(level-2)单元。

这可以被进一步的延伸,第二层(level-2)的单元也可以被划分到第三层单元中,第三层(level-3)单元的也可以划分到第四层单元中。这方面的例子比比皆是,比如教育,学生在第一层,教师在第二层,学院在第三层,学院部门在第四层,和社会学,个人在第一层,社区在第二层。显然,对这些数据的分析需要专门的软件。分层线性和非线性模型(即所谓的多层次模型)已经被用来研究单个分析中的任意层次间的关系,而不忽略与层次结构的每一级相关的可变性。


HLM软件将模型拟合到生成变量的线性模型中,并利用每一级指定的变量生成解释变量。HLM不仅估计每个层级的模型系数,还能预测每个取样单元的随机效应。由于在教育领域的数据层次结构普遍存在,所以在教育领域运用最为普遍,但实际上它适合于任何层次的结构研究领域的数据使用。包括纵向分析,其中个体的重复测量可以嵌入在被研究的个体内。另外,虽然上面的例子表明,这种层次结构的任何级别的成员只嵌套在较高级别的成员中,HLM还可以提供一种情况,成员关系不仅可以是“嵌套的”,也可以是“交叉的”,就像学生在学习期间可能是不同教室的成员一样。


HLM程序允许连续计数、序数和名义变量和假设结果的期望和一组解释变量的线性组合的函数关系。这种关系是由一个合适的链接函数定义的,例如,身份链接(连续结果)或logit链接(二进制结果)。


HLM大大的扩展了可以被评估的分层模型的范围.它同样提供了比先前版本更大的便利.下面是有关关键新特征和选项的综述。


HLM最新版本在建模多层次和纵向数据方面提供了前所未有的灵活性。HLM功能的亮点包括三个新的进程,处理二进制、计数、序数和多项式(名义)的响应变量,以及连续响应变量的一般理论层次线性模型。


四级嵌套模型
四级嵌套模型的横截面数据,如学校教室内学生的项目反应模式。
纵向数据的四级模型,如在社区内人与人之间在时间点内的项目。


四路交叉分类和嵌套混合
对学生在学校内随时间推移,教师的重复措施,或在移民中嵌套项目反应,根据原籍国和目的地进行交叉分类。
对同时居住在某一特定地区并参加某一学校的人的重复措施。


相依随机效应的递阶模型
空间相关邻域效应
社交网络互动


HLM还提供了估计分层广义线性模型的自适应Gauss-Hermite Quadrature (AGH)和
高阶拉普拉斯Laplace近似最大似然法。AGH的方法已经被证明是有效的,尤其当集群规模小,方差分量大的时候。高阶Laplace方法需要较大的集群大学,但允许任意数量的随机效应(当集群较大时非常重要)。

HLM新版本的输出,为统计模型提供优雅的符号,包括视觉上有吸引力的表格、用户可以剪切和复制输出内容到结果中。


HLM统计应用模型选项


适用平台
HLM只适用于Windows系统。

In social research and other fields, research data often have a hierarchical structure. That is, the individual subjects of study may be classified or arranged in groups which themselves have qualities that influence the study. In this case, the individuals can be seen as level-1 units of study, and the groups into which they are arranged are level-2 units. This may be extended further, with level-2 units organized into yet another set of units at a third level and with level-3 units organized into another set of units at a fourth level. Examples of this abound in areas such as education (students at level 1, teachers at level 2, schools at level 3, and school districts at level 4) and sociology (individuals at level 1, neighborhoods at level 2). It is clear that the analysis of such data requires specialized software. Hierarchical linear and nonlinear models (also called multilevel models) have been developed to allow for the study of relationships at any level in a single analysis, while not ignoring the variability associated with each level of the hierarchy.

The HLM program can fit models to outcome variables that generate a linear model with explanatory variables that account for variations at each level, utilizing variables specified at each level. HLM not only estimates model coefficients at each level, but it also predicts the random effects associated with each sampling unit at every level. While commonly used in education research due to the prevalence of hierarchical structures in data from this field, it is suitable for use with data from any research field that have a hierarchical structure. This includes longitudinal analysis, in which an individuals repeated measurements can be nested within the individuals being studied. In addition, although the examples above implies that members of this hierarchy at any of the levels are nested exclusively within a member at a higher level, HLM can also provide for a situation where membership is not necessarily "nested", but "crossed", as is the case when a student may have been a member of various classrooms during the duration of a study period.

The HLM program allows for continuous, count, ordinal, and nominal outcome variables and assumes a functional relationship between the expectation of the outcome and a linear combination of a set of explanatory variables. This relationship is defined by a suitable link function, for example, the identity link (continuous outcomes) or logit link (binary outcomes).

HLM 7 is Compatible with Windows 8. It has been tested on Windows 8 and no problems were reported.

HLM 7 is Compatible with Windows 7. It has successfully passed Microsoft designed tests for compatibility and reliability on Windows 7. It can be used on both the 32-bit and 64-bit editions. Compatible with Windows 7 products install without worry and run reliably with Windows 7.

Microsoft has awarded SSIs HLM software its prestigious Certified for Windows Vista logo. Only applications that pass rigorous testing procedures for compatibility, functionality, and reliability on Windows Vista-based personal computers are granted this logo.

New in HLM 7
HLM 7 offers unprecedented flexibility in modeling multilevel and longitudinal data. With the same full array of graphical procedures and residual files along with the speed of computation, robustness of convergence, and user-friendly interface of HLM 6, HLM 7 highlights include three new procedures that handle binary, count, ordinal and multinomial (nominal) response variables as well as continuous response variables for normal-theory hierarchical linear models:

Four-level nested models:
Four-level nested models for cross-sectional data (for example, models for item response within students within classrooms within schools).
Four-level models for longitudinal data (for example items within time points within persons within neighborhoods).

Four-way cross-classified and nested mixtures:
Repeated measures on students who are moving across teachers within schools over time, or item responses nested within immigrants who are cross-classified by country of origin and country of destination.
Repeated measures on persons who are simultaneously living in a given neighborhood and attending a given school.

Hierarchical models with dependent random effects:
Spatially dependent neighborhood effects.
Social network interactions.

HLM 7 also offers new flexibility in estimating hierarchical generalized linear models through the use of Adaptive Gauss-Hermite Quadrature (AGH) and high-order Laplace approximations to maximum likelihood. The AGH approach has been shown to work very well when cluster sizes are small and variance components are large. the high-order Laplace approach requires somewhat larger cluster sizes but allows an arbitrarily large number of random effects (important when cluster sizes are large)

New HTML output that supplies elegant notation for statistical models including visually attractive tables is also now available, allowing the user to cut and paste output of interest into manuscripts.

HLM 7 manual
A hard copy of the HLM 7 manual is not available.
PDF copies of the HLM 7 manual are available via the HLM 7 Manual option on the Help menu of the full, rental, trial, and student editions of HLM 7 for Windows.


视频课程

  • 中国区典型用户

    • 中国人民大学
    • 中国社会科学院
    • 哈尔滨工业大学
    • 中国水稻研究所
    • 浙江大学
    • 北京师范大学
    • 天津师范大学
    • 北京科技大学
    • 成都理工大学
    • 华东师范大学
    • 广州市红十字会医院
    • 中山大学
    • 北京航空航天大学
    • 江苏大学
    • 南开大学
    • 新乡医学院
    • 华北理工大学
    • 西南财经大学
    • 江苏科技大学
    • 洛阳师范学院
    • 长治学院