Data visualisation Challenges and opportunities in children's social - - PowerPoint PPT Presentation

Data visualisation

James Geddes | 10 December 2018

1

Challenges and opportunities in children's social care

The start of a journey

3

Note: values suppressed by LAs for confidentiality purposes were replaced with 3. The model used to produce the grey shade is a binomial distribution with rate set to the average returning home rate in the data.

Local authority

4

If visualisation is the answer, what is the question?

Image: Bruno Muria, deviantart

5

Where should we start?

Describe

6

Yogi Berra (d. 2015)

You've got to be very careful if you don't know where you are going, because you might not get there.

Where should we start?

Describe

7

Karl Marx (d. 1883)

The philosophers have

• nly interpreted the world,


in various ways. The point, however, is to change it.

Where should we start?

Describe Act

8

9

White White Mixed Mixed Asian or Asian British Asian or Asian British Black or black British Black or black British Other ethnic group Other ethnic group Not stated Not stated Not recorded Not recorded 6%

30/06/2017 30/03/2017

Early Help in the last 6 months

204 Early Help Assessments, Targeted Interventions, and CAFs

47% 11%

Organisation completing assessment Contact source Ethnic backgrounds Children with more than 1 contact in period Children with more than 1 EH record in period Ethnic backgrounds

0% 0% See page 20 for comparisons 82% 5% 5% 7% 0%

Children's services Analysis Tool (ChAT) Page 4

Referral comparison Contact source Age and gender Age and gender Early Help cases that also appear on the Referrals list Contacts that also appear

• n the Referrals list

to from

Contacts in the last 3 months 185 contacts

from 30/12/2016 to 30/06/2017

85% See page 20 for comparisons 1% 7% 0% 1% 0% 19 6 2 contacts 3 contacts 4 or more 12% 26% 0.0% 20% 4.9% 7% 15% 4% 7% 4.3% 0.0% 0.0% Individual Schools Education services Health services Housing LA services Police Other legal agency Other Anonymous Unknown Not recorded 1 2 EHA/CAFs 3 EHA/CAFs 4 or more Yes No 8% 28% 0.0% 0% 10.9% 17% 14% 8% 6% 7.8% 0.0% 0.0% Individual Schools Education services Health services Housing LA services Police Other legal agency Other Anonymous Unknown Not recorded Yes No 58% 20% 7.8% 7% 3.9% 2% 0% 0% 0% 0.0% 5A - LA Services (Internal) 6 - Police 3D - Health Services (Primary) 2A - Schools 3A - Health Services (GP) 3E - Health Services (A&E) Other

15 10 5 5 10 15 Age 0 5 10 15 Aged 20+ 75 Males (41%) 6 Unknown (3%) 104 Females (56%)

10 5 5 10 Age 0 5 10 15 Aged 20+ 109 Males (53%) 1 Unknown (0%) 94 Females (46%)

10

To make a decision we must know the outcome

Describe Act Predict

The Alan Turing Institute 11

Example of "prediction" chart. Redacted.

12

Niels Bohr (d. 1962) (attr.)

Prediction is very difficult,
 especially about the future.

To make a decision we must know the outcome

Describe Predict Act

13

Note: chart shows the percentage of children who left care that were returned to parents or relatives for years 2011-2017 inclusive. Values suppressed for confidentiality purposes were replaced by 3. The model used to produce the grey shaded regions is a binomial likelihood function combined with a gaussian process, and is not final. Local authorities have been anonymised.

14

Act

To predict we must have a model—in other words, a theory

All models are wrong—
 but some are useful.

George Box (d. 2013)

Predict Describe Explain

15

Image by DkEgy - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=61466910

16

Example of "small multiples" plot. Redacted.

17

Act Predict Explain Describe

The steps of data science

? ?

Focussed Exploratory

This chart raises questions ...

18

Note: values suppressed by LAs for confidentiality purposes were replaced with 3. The model used to produce the grey shade is a binomial distribution with rate set to the average returning home rate in the data.

19

Note: Chart shows t-SNE dimensionality reduction of LAs based on the 22 demographic values and associated weights provided in the “Statistical neighbours benchmarking tool”. Plot made using scikit-learn with a particular choice of parameters.

Local authority

... which we hoped this chart might answer

20

Act Predict Explain Describe

We ended up at the beginning

It is a capital mistake to theorise before one has data.

Placeholder for Gapminder

21

22

Image: shipmap.org

23

Early Help Universal Support NFA/ IAG EH Other NFA/ cancel Other NFA/ cancel CIN CP NFA/ cancel CIN Other C P f r

• m

O L A L A C O t h e r NFA/ cancel EH CIN Other C&F Ax Strategy Discussion Section 47 ICPC Initial Contact Referral Closure Universal Targeted Key Front Door Team District Teams Early Help Universal Support

Beginning again

Thank you

24