Liljegren Mats, Nordlund Anders, Ekberg Kerstin
National Centre for Work and Rehabilitation, Department of Health and Society, University of Linköping, Sweden.
Scand J Psychol. 2008 Apr;49(2):169-77. doi: 10.1111/j.1467-9450.2007.00620.x.
The aim of the present study was to evaluate and further validate a modified Exit, Voice, Loyalty and Neglect (EVLN) instrument (Hagedoorn, Van Yperen, Van de Vliert & Buunk, 1999), in a Swedish sample (n= 792). To test the underlying scaling assumptions, the convergent and divergent validity, a multitrait/multi-item analysis was conducted and factor analyses were used to evaluate the factor structure. The concurrent validity was tested by using the modified EVLN instrument as predictor and three different forms of justice as criteria in the analysis. The criterion-related validity was tested and an association between exit behavioral response and actual exit behavior was found (predictive validity). The results showed that the instrument may be considered to be a valid measure with the exception of the aggressive voice scale.
本研究的目的是在一个瑞典样本(n = 792)中评估并进一步验证一种经过修改的退出、表达、忠诚和忽视(EVLN)工具(哈赫多恩、范·伊佩伦、范·德·维利特和布恩克,1999年)。为了检验潜在的量表假设、收敛效度和区分效度,进行了多特质/多项目分析,并使用因子分析来评估因子结构。通过在分析中使用经过修改的EVLN工具作为预测变量,以及三种不同形式的公平作为标准,来检验同时效度。检验了与标准相关的效度,并发现退出行为反应与实际退出行为之间存在关联(预测效度)。结果表明,除了攻击性表达量表外,该工具可被视为一种有效的测量方法。