2001年，Daron Acemoglu, Simon Johnson and James Robinson 在 American Economic Review 发表了一篇论文： The Colonial Origins of Comparative Development: An Empirical Investigation，主要的发现是，制度对不同非洲前殖民地国家经济发展有着很强的解释力。
##
## Call:
## lm(formula = replies ~ title_width, data = cos_threads)
##
## Residuals:
## Min 1Q Median 3Q Max
## -13.1 -7.8 -5.5 -2.0 423.7
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.193 5.746 1.08 0.28
## title_width 0.066 0.106 0.62 0.53
##
## Residual standard error: 34.1 on 198 degrees of freedom
## Multiple R-squared: 0.00195, Adjusted R-squared: -0.00309
## F-statistic: 0.386 on 1 and 198 DF, p-value: 0.535
##
要装可怜？
##
## Call:
## lm(formula = replies ~ title_width + w1 + w2 + w3 + w4 + w5,
## data = cos_threads)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.1 -8.0 -5.6 -1.5 422.5
##
## Coefficients: (4 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 6.9972 5.8659 1.19 0.23
## title_width 0.0755 0.1072 0.70 0.48
## w1TRUE -3.5683 5.0844 -0.70 0.48
## w2TRUE NA NA NA NA
## w3TRUE NA NA NA NA
## w4TRUE NA NA NA NA
## w5TRUE NA NA NA NA
##
## Residual standard error: 34.1 on 197 degrees of freedom
## Multiple R-squared: 0.00444, Adjusted R-squared: -0.00567
## F-statistic: 0.439 on 2 and 197 DF, p-value: 0.645
##
或许，我应该用第一篇回复等待时间而不是回复的绝对数量?
COS你们是不是自己跟自己玩啊？
##
## Call:
## lm(formula = replies ~ title_width + manager, data = cos_threads)
##
## Residuals:
## Min 1Q Median 3Q Max
## -42.8 -4.2 -2.7 0.5 391.3
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.5325 5.4542 0.65 0.52
## title_width 0.0388 0.1005 0.39 0.70
## managerTRUE 36.3348 7.2895 4.98 1.4e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 32.2 on 197 degrees of freedom
## Multiple R-squared: 0.114, Adjusted R-squared: 0.105
## F-statistic: 12.6 on 2 and 197 DF, p-value: 6.85e-06
##
