3 Replies Latest reply on May 16, 2016 10:12 AM by Yuriy Fal

# Help developing a Viz with standard scores that are range-based and change with age cohort

I am working with data from a product called Teaching Strategies Gold.  It has standards scores that are normed by age cohort (i.e. Birth-1 year; 1-2 years; 2-3 years; Pre-school; Pre-K) and expected functioning (Widely Held Expectations) falls within a range of scores by dimension.  For example in the social-emotional dimension, for the birth-1 year cohort, the expected scores are 272-387; for 1-2 years the scores are 387-394; for 2-3 years they are 4585-546 and so on.  We test our students three times per year.  That said, because a child's age cohort may change between assessment intervals, I'm at a loss as to how to display changes in scores over time.  I had the thought of creating a calculated field for each so that different age cohorts could be compared across the dimension (see below), but I'm not exactly sure if this is the correct path and if so, what a Viz would look like.  I've attached some sample data in Excel to better illustrate.  I'm basically looking to see if children improve over time.  Thanks in advance for the thoughts and suggestions.

IF [Age Class Grade 1]="Birth-1 Year" AND [Social-Emotional 1] <272 THEN "Does Not Meet WHE"

ELSEIF [Age Class Grade 1]="Birth-1 Year" AND [Social-Emotional 1] >=272 AND [Social-Emotional 1] <=387 THEN "Meets WHE"

ELSEIF [Age Class Grade 1]="Birth-1 Year" AND [Social-Emotional 1] > 387 THEN "Exceeds WHE"

ELSEIF [Age Class Grade 1]="1-2 Years" AND [Social-Emotional 1] <387 THEN "Does Not Meet WHE"

ELSEIF [Age Class Grade 1]="1-2 Years" AND [Social-Emotional 1] >=378 AND [Social-Emotional 1] <=494 THEN "Meets WHE"

ELSEIF [Age Class Grade 1]="1-2 Years" AND [Social-Emotional 1] > 494 THEN "Exceeds WHE"

ELSEIF [Age Class Grade 1]="2-3 Years" AND [Social-Emotional 1] <458 THEN "Does Not Meet WHE"

ELSEIF [Age Class Grade 1]="2-3 Years" AND [Social-Emotional 1] >=458 AND [Social-Emotional 1] <=546 THEN "Meets WHE"

ELSEIF [Age Class Grade 1]="2-3 Years" AND [Social-Emotional 1] > 546 THEN "Exceeds WHE"

ELSEIF [Age Class Grade 1]="Preschool 3" AND [Social-Emotional 1] <516 THEN "Does Not Meet WHE"

ELSEIF [Age Class Grade 1]="Preschool 3" AND [Social-Emotional 1] >=516 AND [Social-Emotional 1] <=614 THEN "Meets WHE"

ELSEIF [Age Class Grade 1]="Preschool 3" AND [Social-Emotional 1] > 614 THEN "Exceeds WHE"

ELSEIF [Age Class Grade 1]="Pre-K 4" AND [Social-Emotional 1] <589 THEN "Does Not Meet WHE"

ELSEIF [Age Class Grade 1]="Pre-K 4" AND [Social-Emotional 1] >=589 AND [Social-Emotional 1] <=690 THEN "Meets WHE"

ELSEIF [Age Class Grade 1]="Pre-K 4" AND [Social-Emotional 1] > 690 THEN "Exceeds WHE"

END

• ###### 1. Re: Help developing a Viz with standard scores that are range-based and change with age cohort

Hi Jason,

This is my try.

It could be much simpler (Tableau-esque)

to have one data row per each ID and Checkpoint.

So one could analyze all measures and attributes

across Checkpoints (== across Date) for each ID.

I've shaped your data using Tableau Pivot option (Excel / CSV data only)

and collapsed "paired" Dimensions into their single cousins (named like [YF : Dim ID]).

As for Soc_Emot MIN-MAX and Mean values, I outta join the Soc_Emot table twice

(once for each Checkpoint), since a "Pivot-then-Join" is not an option in Tableau (yet?).

Please find the attached wb.

Hope it could help a bit.

Yours,

Yuri

2 of 2 people found this helpful
• ###### 2. Re: Help developing a Viz with standard scores that are range-based and change with age cohort

Yuri - thank you very much for your time and expertise.  I like the look and how you visually integrated the mean scores by time period as reference lines.  I'm going to have to tease through all your calculated fields to get a better understanding for how you produced everything.  It's very thought provoking and the Viz also make we want to further explore how staff are administering the tool since I'm seeing since decreases in scores I a couple classrooms over time.  Many thanks and kindest regards - Jason

• ###### 3. Re: Help developing a Viz with standard scores that are range-based and change with age cohort

Jason, you're welcome.