Differences in Employee Attitudes and Behaviors Based on Rotter's (1966) Internal-External Locus of Control: Are They All Valid?
研究发现，员工内外控差异的研究结果因分类策略、统计方法和研究设计而异，在极端分组、横截面设计、自我报告和专业人员样本中更易得到支持，提示部分已有结论可能不稳健。
Rotter's (1966) Internal-External Locus of Control (LOC) has been widely used to explain differences in many employee attitudes and behaviors. A review of organization studies employing Rotter's LOC scale reveals, however, that researchers used different strategies for classifying employees as “internals” and “externals,” and different statistical techniques and research designs to test hypothesized internal-external (I-E) differences. The present study examined the influence these variations in research method and design have on expected (I-E) differences in a blue-collar and a professional sample. Results showed that findings regarding expected 1-E differences varied by research method and design and were more likely to be supported: (1) when they were tested with I-E subgroups formed with only employees scoring at the extremes on Rotter's LOC scale, (2) when they were examined in a cross-sectional rather than a longitudinal research design, (3) when they were based on self-report rather than behavioral outcomes, and (4) when they were examined within the professional rather than the blue-collar sample. The present study's results suggest, therefore, that reported I-E differences on some employee attitudes and behaviors may not be as valid as suggested by past studies. Moreover, the results suggest a need for a meta-analysis of the organizationally-relevant LOC literature to determine which differences in employee attitudes and behaviors are due to true differences on the LOC construct and which are due to statistical artifacts. Further implications of the present study's results and recommendations for future LOC research are discussed.