3 ms·
There is something weird going on with the bar chart titled "An unequal big pie". The bar for "financial inequality is a bigger problem for a prosperous economy
by lazaroclapp 11y ago
There is something weird going on with the bar chart titled "An unequal big pie". The bar for "financial inequality is a bigger problem for a prosperous economy" is extremely small. The numeric value is 42%. As in, close to half of the sample of (very rich) tech founders agree that it is a huge problem (versus 59% of the general public sample, which incidentally still looks like less than half the bar). Yet the graph makes it seem as if few founders thought so.
Note that the alternative in the same graph is "mediocre growth is a bigger problem for a prosperous economy" and there the value is 48%. So only a few more tech founders[1] see slow growth as a big issue compared to those that mention income inequality (for the economy, not ethically, mind you).
There is a lot to be said about a future that arrives faster and is quite a bit more equally distributed, and which of the two matters the most. But that bar seems to me as misleading to strengthen the posts view on S.V. (wealthy founder) politics.
[1] 2 actually, given the sample size is 33
- bsbechtel 11y agoI noticed several irregularities in how the data was presented that gave me the uneasy feeling the author was trying to push a Democratic agenda using an appeal to the authority of SV founders. The quotes from Mark Zuckerberg and Travis Kalanick, while technically true, are anecdotal, highly contextual, and also don't align with what I've read about their views/actions from other sources. In addition, the author used the same graph twice in the article, which also seems very odd.
- spacecowboy_lon 11y agoNormally the minimum sample size for political poll is 1k ideally 3k.
- lazaroclapp 11y agoIn all fairness, this is a poll of a very particular sub-population (Crunchbase founders and investors). The entire population is 8499 people, so a sampling size of 3k would be more than a third of the population. The sample size for most questions is 129 (but not random, since those were only the ones who responded). I just found that one graph weird, but admittedly I haven't looked carefully at everything else.