A very high percentage of the studies in social science fields and their models are based on data sets from surveys, census, or even economic statistics though dubious data gathering practices as those mentioned in Mujun’s article. This is partially attributed to the complex, subjective, and non-quantitative nature of these social concepts in the first place. Unlike the speed of an object or the weight of a cubic meter of water at the temperature of 23 degree Celsius at air pressure of 760mmPh (14.696 PSI), the social or economic data we encounter can not be easily defined in simple reductive terms. For example, how happy is happy, or how democratic is democratic, how poor is considered as poor, or even how dense an urban area is considered as crowded are not only subjective but also hard to measure or compare between samples. First, the causal relationship is so much entangled. Second, the concepts are too compounded and complex to be meaningful.
I remember when I was a graduate assistant working SAS and SPSS 18 years ago, one of my project was to run multivariate regression on a Chinese census data set to prove my boss’s hypothesis of CCP membership would lead to higher income after excluding education, age, gender, regional, and other social economical factors. My question at that time was – huh, was not CCP supposed to recruit or admit the more elite member from the general population in the first place? Secondly, what if we run the same exercise on Republican or Democratic Party members. Last and not the least, I can prove my hypothesis no matter what the data shows. If the data told me that CCP membership contributes to higher weight in nominal income over time, I could have proved that the social economic system favors the political elite more and more; on the other hand, if the number did not show significant contribution change, I could have theorized that the party membership has degraded over the years, blah, blah. In such economic or social study, it is always “head I win and tail you lose” situation. Well in all fairness it could be that gray income has been hidden from the income statistics or that the causal relationship has been reversed in the first place. Other than getting funding from NSF or political think tanks who needs some bullet points for their agenda, such study really has not much value, and definitely is not scientific.
To me, the only meaningful social and economic measures at per capita level are life expectancy and birth rate. The former identifies survival, the fundamental human need, and the latter identifies the success rate of propagation, the ultimate driver behind the need for social status, money, power, politics, culture and love. If a government or society can sustain progress along those 2 deliverables, that’s much better than delivering on the scale of democracy (distribution of power?) or human right. BTW, what good is human right if the average Joe can not protect himself from war or disease or high medical bill or work stress? What good is democracy if the average person doesn’t have the wherewithal to sustain their DNA into the next generation? That’s natural selection at its worst. A populist leader or organization that cannot aggregate and sustain the public goods for the group survival has no space in the history no matter how righteous or how good-intentioned they are. I don’t mind ceding some of my power and right to those capable institutions either through explicit social contract such as the Constitution or through their manufacturing of public consent such as producing propaganda of social studies to discredit their opponents and to prove their own legitimacy. I guess that’s the societal game to play.
