· Q 1. How can I conveniently source data allowing me to map full Scottish postcodes to SIMD quintiles.
A. You could refer to the Scottish Government website The Scottish Index of Multiple Deprivation where you will find up-to-date data in Excel format under the header Postcode to SIMD rank (8MB).
. Q. 2. What, if any, are the main limitations of the depcat score?
A. Strictly speaking, the census variables which make up the Carstairs index from which the depcat score is calculated are mainly measures of the extent to which an indicator of deprivation is present or absent in a geographical area. These variables are therefore interpreted in terms of a spectrum of deprivation (ref. The handbook of inequality and socioeconomic position). A more balanced index would provide for a more extensive study involving an assessment of the association between varying levels of affluence and mortality due to AP. An index based on postal sectors is also open to improvement in terms of differentiating between smaller geographical boundaries. Further, the use of employment status exclusively for males and the criterion of ownership of at least one car (rather than making a distinction between number of cars) in the calculation of the Carstairs index, as is currently the case, is undoubtedly likely to lead to inaccuracies in estimates of deprivation levels, even relative to socioeconomic conditions in 2001. Also, it is anticipated that while the SIMD will be updated every 3 years, the depcat score will only be updated every 10 years.
·Q 3. Can you recommend a resource for converting my postcode data into deprivation (depcat) scores?
A. Yes, pre-calculated Scottish depcat scores from the 2001 census are provided in the document
Carstairs scores for Scottish postcode sectors from the 2001 Census.
This document also provides a wealth of information on the variables used in the derivation of the depcat score.
· Q 4. I see that the Scottish postcodes must be converted to postal sectors in the first instance prior to conversion to depcat scores. This would take unbearably long to do manually as I have so many cases (a very unusual situation for an SSC4 student). Is there an alternative less painstaking method for deriving the postal sectors?
A. Yes, a script is available to be run using SPSS for this purpose.To The script itself is provided in the following file: step1.sps
Before running the script, you should check that it contains YOUR file name and file path. The script can be executed by, for example, selecting the ‘To end’ option under the menu Run once you have opened the file and gained access to the syntax editor. Make sure, however, that you have your spreadsheet open when you run the script file.
If you wish to learn a little more about running SPSS script files, you may consult the reference: http://www.homepages.ed.ac.uk/ercn82/drafts/pdfdocs/usecom.pdf
· Q 5. Again, due to my usually large number of subjects, it would be a painstakingly long process to allocate depcat scores (as provided at the link you recommend) to my calculated postal sectors. Is there a quick alternative?
A. Yes, you can use a file merging faciity available in SPSS. Notice first of all for the Scottish reference file, however, that there are a number of instances where the expresssion ‘(part)’ appears beside the particular sectors and theese postal sectors are allocated a variety of different depcat scores. This is due to the fact that for some sectors different depcat scores are allocated to different council areas. The simplest thing to do here is to delete the rows in the reference file where the expression ‘(part)’ appears and caarry out the matching procedure with the resultant file.
Details of how to carry out the merger are forthcoming from following a similar example provided on pages 43 and 44 at the site Intermediate SPSS.
To deal with missing values (caused by the deletion of the problem rows above), you should use the following file, especially created for your needs: macdougall_lookup.sav
. Q. 6. What is the best approach to preparing my depcat scores for analysis?
A. I recommend that you use the recode facility in SPSS to split your depcat scores into the categories ‘1 – 2’, ‘3 – 5’ and ‘6 – 7’, representing low, moderate and high levels of deprivation, respectively (see Q. 9. at Arranging Data in Spreadsheets for Statistical Analysis).
If you wish to consider the association between deprivation and a categorical variable (such as that representing whether people function best in the morning or later in the day), it would be an excellent idea to then proceed to the New Presentation provided under the solution to Q. 2 at Hypothesis Tests for Categorical Data.
. Q. 7. My dataset spans a period both prior to and including the introduction of the SIMD. How should I proceed in terms of representing socioeconomic status as a factor in my study?
A. Some useful advice on this topic and on categorizing measures of socioeconomic status can be accessed from the guide Deprivation and Urban Rural Measurement in ISD.
Representing Socioeconomic Status by Margaret MacDougall is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
