3 times when longer ≠ better

3 times when longer ≠ better

David Card and Stefano DellaVigna have written a paper on the increasing length of economics papers (here's  the paper and a related vox article).

Of course the trend might reflect increasing quality.  But there are at least three times when length is simply an indicator of sloppy writing.  It's like Mark Twain said: "I would have written a shorter letter, but I did not have the time."



1. When disorganized sections lead to a bloated structure


2. Background
   2.1 History of Textiles
   2.2 Existing Evidence
   2.3 Exogeneity
   2.4 Labor Reform
   2.5 State Reorganization

3. Empirical Strategy

Pages: 8.5
2. A Natural Experiment: Reforms 1750-1850
   2.1 Natural Experiment
       Local Labor Reform
       State Reorganization
   2.2 Exogeneity

Pages: 6




2. When the main point is missing.


Sidestepping, for now, the question of why this happens, let's look at examples that it happens.   

Jingle and Pickwick (2011)  study times to approval and show that the  effects of reform are stronger for time-sensitive goods. Weller (2013) investigates sub-components of standard cost measures, and finds that only rapid response times have significant effects.  Snodgrass et al. (2014) link time in holding to different measures of  the composition of cost. Tupman  (2014) use Romanian transaction data to link allocation to channel A to delays, and delays to slower rates of Y. They employ a similar IV framework to the one we employ in this paper.


That's a collection of sentences that may or may not be connected; it's 89 words with no clear point. 


 Here's another one.


Table 4 shows the difference between A and B.  For regions in the northwest, east, and northeast, A is higher than B by .02. This difference is quite small. We find no evidence that C has an impact. In metropolises of the same regions, D has an impact of 40 percent, which is quite sizable.  In the south, the difference between A and B is greater (.3) but it is statistically insignificant. In the rural midwest, the effect of D rises to 60 percent but this may be due to the reform bill having been passed.  


That's 96 words with no clear point.  



3. When sentences are unnecessarily convoluted. 


"To the extent that social activities may impede health reduction...."


“Building on findings that self-uncertainty motivates attempts to restore certainty about the self, particularly in ways that highlight one’s distinctiveness from others, we show that self-uncertainty, relative to uncertainty in general, increases creative generation.”





Of course these problems are not about the cosmetic part of writing; rather they live at the intersection of writing and analysis.  The good news is they can be fixed.















Post by Varanya Chaubey









Flip it!

Flip it!

Presentations: Setting Expectations

Presentations: Setting Expectations