Anonymous wrote:

Anonymous wrote:Okay, this is now the third thread I am posting the exact same comment:

I'm a data scientist and the XLS spreadsheet of scores is a mess to parse. I have little confidence in the conclusions I'm seeing being drawn on this messageboard (not just this thread). Perhaps everyone else is better at this than me, but it is damn near impossible to figure out which columns sum to one another. Unless you can conclusively understand the data architecture (and summation is the first step), you should hesitate to draw conclusions.

I am interested in drawing lessons regarding Hardy and Deal from the PARCC data. I've failed so far. Give me some more time (hours or until tomorrow) to make some headway. But, in general, unless you honestly believe the poster completely understands the data structure (and my professional opinion is to be highly skeptical for this data set), you should cast aside the conclusions.

Can you link to the spreadsheet?

http://osse.dc.gov/parcc/2015results

Anonymous wrote:Okay, this is now the third thread I am posting the exact same comment:

I'm a data scientist and the XLS spreadsheet of scores is a mess to parse. I have little confidence in the conclusions I'm seeing being drawn on this messageboard (not just this thread). Perhaps everyone else is better at this than me, but it is damn near impossible to figure out which columns sum to one another. Unless you can conclusively understand the data architecture (and summation is the first step), you should hesitate to draw conclusions.

I am interested in drawing lessons regarding Hardy and Deal from the PARCC data. I've failed so far. Give me some more time (hours or until tomorrow) to make some headway. But, in general, unless you honestly believe the poster completely understands the data structure (and my professional opinion is to be highly skeptical for this data set), you should cast aside the conclusions.

Can you link to the spreadsheet?

Anonymous wrote:

Anonymous wrote:Okay, this is now the third thread I am posting the exact same comment:

I'm a data scientist and the XLS spreadsheet of scores is a mess to parse. I have little confidence in the conclusions I'm seeing being drawn on this messageboard (not just this thread). Perhaps everyone else is better at this than me, but it is damn near impossible to figure out which columns sum to one another. Unless you can conclusively understand the data architecture (and summation is the first step), you should hesitate to draw conclusions.

I am interested in drawing lessons regarding Hardy and Deal from the PARCC data. I've failed so far. Give me some more time (hours or until tomorrow) to make some headway. But, in general, unless you honestly believe the poster completely understands the data structure (and my professional opinion is to be highly skeptical for this data set), you should cast aside the conclusions.

No, they are not better at it than you. They are just so bad at statistics that they don't realize how bad their "conclusions" are. It's emotional responses and what people want to see that they then "support with evidence from the numbers".

Agreed. Subjective conclusions using obviously insufficient data vs. objective conclusions which currently cannot be made.

Anonymous wrote:Okay, this is now the third thread I am posting the exact same comment:

I'm a data scientist and the XLS spreadsheet of scores is a mess to parse. I have little confidence in the conclusions I'm seeing being drawn on this messageboard (not just this thread). Perhaps everyone else is better at this than me, but it is damn near impossible to figure out which columns sum to one another. Unless you can conclusively understand the data architecture (and summation is the first step), you should hesitate to draw conclusions.

I am interested in drawing lessons regarding Hardy and Deal from the PARCC data. I've failed so far. Give me some more time (hours or until tomorrow) to make some headway. But, in general, unless you honestly believe the poster completely understands the data structure (and my professional opinion is to be highly skeptical for this data set), you should cast aside the conclusions.

No, they are not better at it than you. They are just so bad at statistics that they don't realize how bad their "conclusions" are. It's emotional responses and what people want to see that they then "support with evidence from the numbers".

Anonymous wrote:Okay, this is now the third thread I am posting the exact same comment:

I'm a data scientist and the XLS spreadsheet of scores is a mess to parse. I have little confidence in the conclusions I'm seeing being drawn on this messageboard (not just this thread). Perhaps everyone else is better at this than me, but it is damn near impossible to figure out which columns sum to one another. Unless you can conclusively understand the data architecture (and summation is the first step), you should hesitate to draw conclusions.

I am interested in drawing lessons regarding Hardy and Deal from the PARCC data. I've failed so far. Give me some more time (hours or until tomorrow) to make some headway. But, in general, unless you honestly believe the poster completely understands the data structure (and my professional opinion is to be highly skeptical for this data set), you should cast aside the conclusions.

Did you ever reach any conclusions? It seems to me that the data for PARCC results is intentionally presented this way so that smaller population groups (less than 25) cannot easily be isolated if not impossible to isolate. This is the case at Hardy. For example, not enough students took Algebra or Geometry at each grade level for us to see their scores and in the case of Geometry, not enough students took this class school-wide to even be able to see Hardy's overall success with Geometry. Two hopeful conclusions can possibly be made: 1) A least for Algebra, the scores appear to be high (not fair to compare to Deal since Deal has a much higher population size) and 2) Geometry is being taught at Hardy which is advanced for 8th graders.

If as others are saying, the feeder schools for Hardy are increasingly sending more students, perhaps soon we will be able to see test results in the more advanced math classes since all the feeder schools to Hardy are high performing schools.

Anonymous wrote:Hardy scores are below the district average. Just terrible. Guess Deal is still the only game in town.

Way to much emphasis placed on the scores. One could also say when you look at the advanced classes Geometry and Algebra that Hardy outperformed Deal. The sample sizes are way to small to draw any conclusions. Hardy has strong leadership and strong offerings. The PTO has very committed parents, feeder school enrollment is growing and more and more families are buying in (50% of last years feeder school 5th grade kids chose Hardy this year) and the future is extremely bright. The good news is that DC will have two great middle schools and hopefully more in the future.

Okay, this is now the third thread I am posting the exact same comment:

I'm a data scientist and the XLS spreadsheet of scores is a mess to parse. I have little confidence in the conclusions I'm seeing being drawn on this messageboard (not just this thread). Perhaps everyone else is better at this than me, but it is damn near impossible to figure out which columns sum to one another. Unless you can conclusively understand the data architecture (and summation is the first step), you should hesitate to draw conclusions.

I am interested in drawing lessons regarding Hardy and Deal from the PARCC data. I've failed so far. Give me some more time (hours or until tomorrow) to make some headway. But, in general, unless you honestly believe the poster completely understands the data structure (and my professional opinion is to be highly skeptical for this data set), you should cast aside the conclusions.

Anonymous wrote:I'm not sure, but if I had to guess, I think 20:40 is calling OP a racist.

Where do you get that?

I'm not sure, but if I had to guess, I think 20:40 is calling OP a racist.

Anonymous wrote:Deal's double, and these scores are in percentages.

What exactly are we talking about here, to tired to look at the other thread

Deal's double, and these scores are in percentages.

Anonymous wrote:

Anonymous wrote:

Anonymous wrote:Hardy scores are below the district average. Just terrible. Guess Deal is still the only game in town.

Ha! Have you not noticed the tide is changing? More IB families are going to Hardy. Maybe you haven't yet had the chance to compare white student performance between the two schools. Why don't you do that and report back.

-IB for Deal but considering lottery for Hardy and Basis

Say what???

Yep. Maybe once you check out the scores, add that to size of Deal.

Anonymous wrote:

Anonymous wrote:Hardy scores are below the district average. Just terrible. Guess Deal is still the only game in town.

Ha! Have you not noticed the tide is changing? More IB families are going to Hardy. Maybe you haven't yet had the chance to compare white student performance between the two schools. Why don't you do that and report back.

-IB for Deal but considering lottery for Hardy and Basis

Say what???

Anonymous wrote:Hardy scores are below the district average. Just terrible. Guess Deal is still the only game in town.

Ha! Have you not noticed the tide is changing? More IB families are going to Hardy. Maybe you haven't yet had the chance to compare white student performance between the two schools. Why don't you do that and report back.

-IB for Deal but considering lottery for Hardy and Basis

Hardy scores are below the district average. Just terrible. Guess Deal is still the only game in town.