بهبود مدل دموکراسی: نمونه ای از اثرات عقب مانده توسعه اقتصادی آموزش و پرورش و برابری جنسیتی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|35882||2015||15 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Social Science Research, Volume 46, July 2014, Pages 169–183
The author examines how time delayed effects of economic development, education, and gender equality influence political democracy. Literature review shows inadequate understanding of lagged effects, which raises methodological and theoretical issues with the current quantitative studies of democracy. Using country-years as a unit of analysis, the author estimates a series of OLS PCSE models for each predictor with a systematic analysis of the distributions of the lagged effects. The second set of multiple OLS PCSE regressions are estimated including all three independent variables. The results show that economic development, education, and gender have three unique trajectories of the time-delayed effects: Economic development has long-term effects, education produces continuous effects regardless of the timing, and gender equality has the most prominent immediate and short term effects. The results call for the reassessment of model specifications and theoretical setups in the quantitative studies of democracy.
It is a long-standing tradition in quantitative research on democratization, as well as other quantitative studies in general, to include a broad range of control variables in the baseline model. One reason this tradition exists is to eliminate or reduce omitted variable bias and to avoid spurious correlations. The second reason is theoretical – to account for alternative explanations and theories. Yet, little attention has been paid to the operationalization of the control variables in quantitative models of democracy. Specifically, there is often inadequate explanation or no discussion of the reasons why certain control variables are employed with immediate effects on the dependent variable, while other control variables are incorporated in a lagged form, that is, when delayed effects are supposed to take place. Consequently, this leads to the current state of research with insufficient investigations of the lags of the main independent and control variables in statistical models of democracy. The limited discussion of using immediate versus lagged effects raises larger theoretical issues regarding the extent to which social processes and phenomena can instantaneously influence each other. Empirically, inadequate analysis of lags leads to poor model specifications where the effects of the independent variables, such as GDP, are often contradictory in different studies. Table 1 summarizes a sample of the studies that analyzed political democracy and used economic development as one of the independent variables. Surprisingly, no studies in this table explicitly discuss the operationalization of the concepts in terms of lagged effects. Most papers simply state that the authors chose a certain lag of GDP without an explanation, while some papers never even mention that the author/s chose a certain lag. Using a systematic analysis of lags can potentially reduce this level of disagreement about the effects of economic development and other independent variables on democracy.