Infographics and statistics are great. Incredibly powerful and easy to understand
Also incredibly dangerous because you can make it say whatever you want
Also incredibly dangerous because you can make it say whatever you want
#1 Duh, of course overwhelming majority of HK people are against HK Independence. We’ve been saying that since the beginning. Even Joshy admits to that
#2 The tweet authors then try to pull a fast one and equate protesters/rioters to all of HK people, which again, we’ve been saying they’re not. Hence the term #SilentMajority
Let us look at how the raw numbers get manipulated to paint the picture we want
From pori.hk/s/sp_rpt_speci…, 11.5% polled responded Yes to “Do you agree to Hong Kong being independent?”
From pori.hk/s/sp_rpt_speci…, 11.5% polled responded Yes to “Do you agree to Hong Kong being independent?”
It is a weighted result based on sample age. The weighting is derived from “Mid-year population for 2018” provided by Census & Statistics Dept. So far so good
Why weighted? Well, the age distribution of the poll turned out to be different from the actual HK demographics, so the weighting helps us understand how the poll results would look like based on known data about the population
In this case, the poll was over-representing the under 39 group and under-representing the 60+ group.
If you look at the raw poll results, the 11.5% becomes 12.8%. A little different, but meh, we already know overwhelming majority of HK people are not for HK Independence
If you look at the raw poll results, the 11.5% becomes 12.8%. A little different, but meh, we already know overwhelming majority of HK people are not for HK Independence
Then, there’s the breakdown by age represented by the 100% stacked chart. This subtly implies that each age group are represented the same, which of course, is not true. Remember, it’s weighted :)
Infographics also use a color scheme that draws attention to what they want to say. In this case, the 83% No you should be looking at, in red
If we look at the raw data, and move the Don’t Knows to the middle, then all of a sudden, it doesn’t look so lop-sided, especially the 18-29 bracket 
So why 60+ has so many responses? Well, what actually happened was that HKPOP aggregated the 60-69 group with the 70+ group to create a new 60+ group. They then further combined it with the 50-59 group to create a 50+ group and then combined 30-39 & 40-49, to create a 30-49 group
Why? To bring down the yes involving the 30-39 group by mixing it with the 40-49. A more reasonable grouping would be 18-39, 40-59, and 60+
@threadreaderapp unroll
Only 71% of HKers under 40 agree with HK being part of China, plummets to 65% for under 30
Most people prefer a rising left to right, we can highlight the increase by reversing the age group order
