A procedure simulating Likert scale item responses

Data collected during a survey by means of a questionnaires are, in general, expressed with references to a Likert type scale, giving rise to non-metric data (ordinal categorical). However most of the statistocal procedures used to analyze survey data (for example FA or SEM) required at list interval scale measures, that may be obtained by using proper scaling procedures. In order to compare by simulation, the quantification thecniques most commonly used in the literature we consider necessary to achieve in advance an appropriate algorithm that best reproduces the discretization process followed by the respondents to the Likert questionnaire items. Accordingly, we propose a discretization procedures that, starting from a continuous random variables, describing all possible individual responses to a given stimulus, generates the corresponding categories, chosen among finite sets of integer values.

通过问卷调研收集的数据，通常以李克特量表（Likert scale）为参照进行编码，由此产生的数据集属于非度量数据（即有序分类数据）。然而，绝大多数用于分析调研数据的统计方法（如因子分析（FA）、结构方程模型（SEM）），至少需要等距尺度的测量数据，而这类数据可通过恰当的量表转换方法获取。为通过仿真对比当前文献中最常用的各类量化方法，我们认为有必要预先构建一套适配的算法，以精准复现受访者针对李克特问卷题项作答时的离散化过程。据此，我们提出一种离散化方法：该方法以连续随机变量为基础，用以表征针对某一特定刺激的所有个体可能作答结果，并从有限整数集合中选取对应类别，最终生成符合要求的离散数据。