Subjective ratings of the concreteness, valence, and crosslingual similarity (with Dutch) of two-word English expressions
收藏资源简介：
These data are described in: S Lindstromberg & J Eyckmans. The retrievability of L2 English multi-word items in a context of strongly form-focused exposure: What matters? Studies in Second Language Acquisition. Accepted 31 May, 2020. The data were collected for an investigation of the effects of a number of pedagogical and lexical variables on the memorability of the forms of two-word English expressions of miscellaneous syntactic structure. The data consist of ratings of Concreteness, Valence, and Similarity (i.e., cognateness and congruence between L1 Dutch and L2 English). The first five columns relate to Concreteness; the next five to Valence; and the last four to Similarity. As detailed in the article mentioned above, it was possible to validate the the ratings of Concreteness and Similarity owing to modest overlapping coverage with existing databases, namely:## Brysbaert, M, Warriner, A., & Kuperman, V. (2014). Concreteness ratings for 40,000 generally known English word lemmas. Behavior Research Methods, 46, 904– 11. https://doi.org/10.3758/s13428-013-0403-5## Warriner A., Kuperman, V., & Brysbaert, M. (2013). Norms of valence, arousal, and dominance for 13,915 English lemmas. Behavior Research Methods, 45, 1191–1207. https://doi.org/10.3758/s13428-012-0314-x For all three sets of ratings indicators of internal reliability are good:Concreteness: Mean split-half r(5000) = .93; Spearman-Brown corrected reliability = .96; SD = 0.11. Valence: split.h r = .95, SB corr reliability = .98; SD = 0.08; Similarity: split.h r = .95, SB corr reliability = .98; SD = 0.10 We used the R function 'splithalf.r' in the package 'multicon' (Sherman & Serfas, 2011) to calculate the mean of 5000 split-half correlations across the ratings of all raters, with the split randomly chosen each time.## Sherman, R., & Serfas, D. (2011). multicon (Multivariate constructs), Version 1.6. http://127.0.0.1:13752/library/multicon/html/multicon-package.html<br>
本数据集的相关研究刊载于：S Lindstromberg与J Eyckmans。《强形式聚焦输入语境中二语英语多词项（multi-word items）的可提取性：关键影响因素探析》，发表于《第二语言习得研究》（Studies in Second Language Acquisition），2020年5月31日获正式录用。 本数据集的采集服务于一项探究，旨在考察多项教学法与词汇变量对句法结构各异的英语双词表达式形式记忆效果的影响。 数据集包含具体性（Concreteness）、效价（Valence）与相似性（Similarity，即第一语言荷兰语（L1 Dutch）与二语英语（L2 English）间的同源性与一致性）三项维度的评分。其中前五列对应具体性评分，中间五列对应效价评分，最后四列对应相似性评分。 如上述文献所述，得益于与现有数据库存在适度的重叠覆盖范围，本数据集的具体性与相似性评分可进行效度验证，相关数据库如下： 1. Brysbaert, M.、Warriner, A.与Kuperman, V.（2014）。《40000个通用英语词元的具体性评分》，《行为研究方法》（Behavior Research Methods），第46卷，第904–911页。https://doi.org/10.3758/s13428-013-0403-5 2. Warriner, A.、Kuperman, V.与Brysbaert, M.（2013）。《13915个英语词元的效价、唤醒度与支配性常模》，《行为研究方法》（Behavior Research Methods），第45卷，第1191–1207页。https://doi.org/10.3758/s13428-012-0314-x 针对三项评分维度，其内部一致性信度指标均表现优异： 具体性（Concreteness）：5000次均分半相关系数r=0.93；经Spearman-Brown校正后的信度为0.96；标准差(SD)=0.11 效价（Valence）：分半相关系数r=0.95，经Spearman-Brown校正后的信度为0.98；标准差(SD)=0.08 相似性（Similarity）：分半相关系数r=0.95，经Spearman-Brown校正后的信度为0.98；标准差(SD)=0.10 本研究使用R语言`multicon`包（Sherman与Serfas，2011）中的`splithalf.r`函数，对所有评分者的评分计算5000次随机分半相关的均值，每次分半均随机选取。 3. Sherman, R.与Serfas, D.（2011）。《multicon（多变量建构工具包）》，版本1.6。http://127.0.0.1:13752/library/multicon/html/multicon-package.html
