IMD 2004 to IMD 2010

[AMJ-coder]

Question / Problem statement

In all of the data tables we have IMD 2004. If I want to ensure that this is updated to IMD 2010 (to most closely match the particular cohort of children I am working with) how would you suggest I approach this please?

Compliance checklist

• I confirm I have not entered any information from within the ONS SRS (or only information that has been cleared for release by the ONS SRS output checking team).
• I confirm I have asked only one question in the textbox above.
• I confirm I have checked for similar questions amongst existing Q&As.

[zcabh35]

Hi, a "potential" solution to this is to see how correlated they are, and if it's highly correlated, then you could potentially argue that IMD 2004 and IMD 2010 are estimates of each other and just use those values.

You could do this by finding a set of sequential censi , one containing IMD 2004 and one containing IMD 2010. Given that most PMR's which exist in one census appears in the next, you could then cross tabulate IMD 2004 with IMD 2010 to see the correlation matrix.

Unfortunately, without raw postcode data, we cannot merge in (through time) the IMD values. In my own causal work, we've argued that IMD decile is relative to the time at which the pupil is assigned treatment, and thus backfilling etc might not be appropriate.

[matthewjay15]

Hi

The name of IMD04 in the HES data tables is misleading. You will see in the field description that it is not actually just the 2004 version that is used but different versions depending on the year of HES activity. This is a scheme employed by NHS England in preparation of HES and so would be what you get whether you use ECHILD or you got your own extract of HES.

As for NPD, by default NPD only comes with IDACI (the income deprivation affecting children index, one of the indicators of the income domain). When we set up ECHILD, we provided DfE with a file of all LSOAs and the IMD and all its domains plus the IDACI and IDAOPI across all versions and asked them to apply the same scheme as NHS England. You will therefore see separate indices of deprivation variables in the NPD data.

Thanks
Matt