4 ms·

88% of Groom's fellow students currently have job offers on the table, with approximately 60% of those headed to start-ups and the remaining 40% mostly employed

88% of Groom's fellow students currently have job offers on the table, with approximately 60% of those headed to start-ups and the remaining 40% mostly employed at consultancies. If the total group size is 20, why not use actual numbers rather than percentages? I know it sounds like more using percentages, but it feels a little misleading to me. edit: formatting.

lachyg 14y ago
Because 3 of the 20 were not looking for jobs. Thus it would not have made sense to use them in assessing the success. I'm Lachy Groom, btw.
- robgough 14y ago
  I think that's a good enough reason to not include them in the stats, but I find the use of percentages confusing over simply the raw numbers? 20 People. 3 Not Looking, so removed. It mentions your fellow students, so minus 1 more and we're down to sample size of 16. 88% of 16 is 14.08 - 14.08 people is a weird number. Maybe the maths was done including yourself, so that'd be 17... which gives us 14.96 people - so that doesn't make sense either?!? Surely something like "Out of his fellow students 16 were looking for, and have job offers on the table. 10 of whom are headed to startups, with the other 6 mostly employed at consultancies." My numbers aren't accurate, but I think the sentence is much clearer. This is an aside though, congrats on being brave enough to go off to another country to do this - never an easy choice, though sometimes here on HN it comes across as not-a-big-deal.
  - lachyg 14y ago
    We provided the author of this story, and all other stories all the numbers. I guess the percentages just are easier to write about! Thanks for the compliment! It was a crazy ride.
dpritchett 14y ago
Leading with the percentage is an editorial choice that implies that the model might scale; anyone who lands a spot in one of these groups has a 90% chance of getting a decent paying job the day after the program ends. If you use the absolute headcounts then you waste headline space and you admit the statistically insignificant sample size.

lachyg 14y ago

Because 3 of the 20 were not looking for jobs. Thus it would not have made sense to use them in assessing the success. I'm Lachy Groom, btw.

robgough 14y ago
I think that's a good enough reason to not include them in the stats, but I find the use of percentages confusing over simply the raw numbers? 20 People. 3 Not Looking, so removed. It mentions your fellow students, so minus 1 more and we're down to sample size of 16. 88% of 16 is 14.08 - 14.08 people is a weird number. Maybe the maths was done including yourself, so that'd be 17... which gives us 14.96 people - so that doesn't make sense either?!? Surely something like "Out of his fellow students 16 were looking for, and have job offers on the table. 10 of whom are headed to startups, with the other 6 mostly employed at consultancies." My numbers aren't accurate, but I think the sentence is much clearer. This is an aside though, congrats on being brave enough to go off to another country to do this - never an easy choice, though sometimes here on HN it comes across as not-a-big-deal.
- lachyg 14y ago
  We provided the author of this story, and all other stories all the numbers. I guess the percentages just are easier to write about! Thanks for the compliment! It was a crazy ride.

dpritchett 14y ago

Leading with the percentage is an editorial choice that implies that the model might scale; anyone who lands a spot in one of these groups has a 90% chance of getting a decent paying job the day after the program ends. If you use the absolute headcounts then you waste headline space and you admit the statistically insignificant sample size.