OK, this one has me scratching my head a bit. I have some Ohio school district data that my sister sent me. Schools are basically rated based upon student performance on standardized tests and such. I thought it would be interesting to calculate the Mean (Average) score for all schools, the standard deviation of their scores and then calculate each schools's number of standard deviation's from that mean. Then, I could have a slicer to break the schools down by category (rural, urban, etc.). The thought here was that I could find outliers in the data so that my sister could focus on schools that overperformed for their category and perhaps that would lead to finding out what those schools are doing differently.
So, everything went well up until I put a slicer on the school's category and then I got some really weird results.
So, here is some sample data:
DISTRICT
| DISTRICT IRN | DISTRICT NAME | OVERALL SCORE |
| 442 | Manchester Local | 2.75 |
| 43489 | Akron City | 1.31 |
| 43497 | Alliance City | 1.63 |
| 43505 | Ashland City | 2.63 |
| 43513 | Ashtabula Area City | 1.94 |
| 43521 | Athens City | 2.69 |
| 43539 | Barberton City | 1.69 |
| 43547 | Bay Village City | 4.19 |
| 43554 | Beachwood City | 3.63 |
| 43562 | Bedford City | 1.69 |
| 43570 | Bellaire Local | 1.69 |
| 43588 | Bellefontaine City | 3.25 |
| 43596 | Bellevue City | 2.94 |
| 43604 | Belpre City | 2.19 |
| 43612 | Berea City | 1.94 |
| 43620 | Bexley City | 2.88 |
| 43638 | Bowling Green City | 3.25 |
| 43646 | Brecksville-Broadview Heights City | 4.19 |
| 43653 | Brooklyn City | 2.69 |
| 43661 | Brunswick City | 3.5 |
| 43679 | Bryan City | 2.94 |
| 43687 | Bucyrus City | 1.81 |
| 43695 | Cambridge City | 2.75 |
| 43703 | Campbell City | 1.75 |
| 43711 | Canton City | 1.56 |
TYPOLOGY BY DISTRICT
| IRN | District Name | 2013 Typology |
| 442 | Manchester Local (Adams) | 1 |
| 43489 | Akron City | 8 |
| 43497 | Alliance City | 7 |
| 43505 | Ashland City | 4 |
| 43513 | Ashtabula Area City | 4 |
| 43521 | Athens City | 5 |
| 43539 | Barberton City | 7 |
| 43547 | Bay Village City | 6 |
| 43554 | Beachwood City | 6 |
| 43562 | Bedford City | 7 |
| 43570 | Bellaire Local | 4 |
| 43588 | Bellefontaine City | 4 |
| 43596 | Bellevue City | 1 |
| 43604 | Belpre City | 4 |
| 43612 | Berea City | 7 |
| 43620 | Bexley City | 6 |
| 43638 | Bowling Green City | 5 |
| 43646 | Brecksville-Broadview Heights City | 6 |
| 43653 | Brooklyn City | 7 |
| 43661 | Brunswick City | 5 |
| 43679 | Bryan City | 4 |
| 43687 | Bucyrus City | 4 |
| 43695 | Cambridge City | 4 |
| 43703 | Campbell City | 7 |
| 43711 | Canton City | 8 |
Topologies
| 2013 Typology Code | Major Grouping |
| 0 | Island |
| 1 | Rural |
| 2 | Rural |
| 3 | Small Town |
| 4 | Small Town |
| 5 | Suburban |
| 6 | Suburban |
| 7 | Urban |
| 8 | Urban |
I built these measures (in DISTRICT):
Overall Score District Mean = AVERAGE(DISTRICT[OVERALL SCORE]) Overall Score District StdDev = STDEV.P(DISTRICT[OVERALL SCORE])
And then this custom column in DISTRICT:
OVERALL SCORE STDDEV FROM MEAN = (DISTRICT[OVERALL SCORE] - CALCULATE([Overall Score District Mean],ALL(DISTRICT))) / CALCULATE([Overall Score District StdDev],ALL(DISTRICT))
Here are the relationships
And here are some visuals, you can see I have a slicer and some tables:
Now for the weirdness. I click on Island and both of my two bottom tables are filtered correctly, but the table in the upper right is not!
Now when I click on Rural in the slicer, I get more, different weirdness...
I feel like I am missing something obvious but I can't seem to find it. I think it must have something to do with my measures I have to imagine. Maybe the ALL is doing something bizarre. But, in any case, this is certainly not what I would have expected... Anyone have any thoughts?