AUTNES Content Analysis of Party Manifestos 2006

The AUTNES dataset on party manifestos covers the manifestos of the relevant parties that competed in the Austrian national election 2006. All natural sentences within the manifestos are part of the dataset. The units of analysis are standardized statements derived from these natural sentences by means of a set of unitizing rules based on grammar, not content. The coding procedure applies the AUTNES relational approach of recording subjects, predicates, and objects to party manifestos. The subject actor is usually the party, but it can also be another actor being cited in the manifesto. There are two types of objects: issues and object actors. Issues are recorded by coders selecting from the AUTNES issue coding scheme the one issue that fits the content of the statement best. One object actor can be recorded per statement, each with their name (if an individual is present) and organizational affiliation. The predicate numerically records whether the subject’s position towards the issue or object actor (if present) is one of (conditional) support, (conditional) rejection, or conveys a neutral stance. In addition to the basic subject–predicate–object structure we code character traits for all subject and object actors as well as justifications for issue statements, party record and pledges.

Topics: sentence (the text of the natural sentence); statement derived from the natural sentence; page number of the manifesto where the natural sentence can be found; organizational affiliation of the subject actor (usually the party); name of the subject actor if an individual; attributes of the subject actor; predicate (the relation between subject and object); predicate special (neutral position or a description of objective facts); issue category selected for the statement; statement refers to regulation of the issue on the European level; number of justifications referring to: the economy, protection of the welfare state, protection of the environment, security, education, governance, ethnic or nationalist rationales, religion, universalistic values, and number of justifications in the residual category; reference to a party’s own or a political opponent’s past behavior or achievements at different levels of governance (national level, land level, international level, historical); pledge (subjective pledges: promises whose evaluation requires a value judgement, objective pledges: promises that can be objectively tested); organization of the object actor (usually the party); name of the object actor if individual; characteristics of the object actor (attributes: competence, character, leadership, appearance); year of election for which the manifesto was published; party authoring the manifesto; title of the manifesto.

Additionally coded was: ID number for each observation; sentence-ID.

AUTNES政党宣言数据集涵盖了2006年奥地利全国大选参选政党的全部官方宣言，数据集中包含宣言内所有自然语句。分析单元为基于语法（而非内容）的单元划分规则从上述自然语句中提取得到的标准化陈述。编码流程采用AUTNES关系分析法，对政党宣言中的主语、谓语与宾语进行标注：主语行为体通常为政党，也可以是宣言中引用的其他行为体；宾语分为两类，即议题与宾语行为体。编码人员需从AUTNES议题编码体系中选取最贴合该陈述内容的单一议题完成标注，每条陈述可标注一个宾语行为体，需记录其名称（若为个体）与所属组织。谓语部分以数值形式标注主语针对议题或宾语行为体（若存在）的立场，涵盖（附条件的）支持、（附条件的）反对或中性态度。除基础的主-谓-宾结构外，我们还对所有主语与宾语行为体的性格特征，以及议题陈述、政党过往记录与竞选承诺的佐证理由进行编码。

数据集包含以下核心字段：自然语句文本（sentence）、从自然语句提取的标准化陈述、该自然语句所在宣言的页码、主语行为体的所属组织（通常为政党）、主语行为体（若为个体）的姓名、主语行为体属性、谓语（主语与宾语间的关系）、特殊谓语（中性立场或客观事实描述）、该陈述所属的议题类别、陈述是否涉及欧洲层面的议题监管、各类佐证理由的数量：包括经济、福利国家保护、环境保护、安全、教育、治理、族裔或民族主义诉求、宗教、普世价值，以及剩余类别的佐证理由数量；提及政党自身或政治对手在不同治理层级（国家级、地方级、国际级、历史层面）的过往行为或政绩；竞选承诺（主观承诺：需通过价值判断评估的承诺；客观承诺：可通过客观标准检验的承诺）、宾语行为体的所属组织（通常为政党）、宾语行为体（若为个体）的姓名、宾语行为体的特征（属性：能力、性格、领导力、形象）、该宣言所属选举的年份、发布该宣言的政党、宣言标题。

此外还额外编码了以下内容：每条观测的ID编号、语句ID（sentence-ID）。