National Records of Scotland

Preserving the past, Recording the present, Informing the future

Reports and Publications

Reports and Publications

The National Records of Scotland Beyond 2011 programme produced a range of publications and reports.

Files are available to download in Adobe Acrobat Portable Document Format (PDF).
Maximum file size 3.5 Mb

PDF files require Acrobat Reader. Download Acrobat Reader free Get Acrobat Reader

Working Papers

Assessing Administrative Data

Comparison of Population Counts from Aggregated Administrative Data and the Mid-Year Population Estimates (13 July 2012)

Technical Reports and Publications

The following papers consist of technical/mathematical notation, information and data that is not always possible to explain in plain English.

National Records of Scotland Reports

Matching the Census 2011 to the NHS Central Register using the Ord Wood method
 

This paper reports a project run in 2012/2013 to match the results of the 2011 Census in Scotland to the NHS Central Register (NHSCR) extract as it stood at the end of May 2011

The Ord Wood Project: A
method of calculating match probabilities from record linkage output

 

This is a technical paper presenting a method for calculating match probabilities from record linkage output

Population Estimates Comparison Project: Data Sources Review

This report documents and reviews the data sources that have contributed to the work of the National Records of Scotland’s (NRS) Population Estimates Comparison Project

Other Organisation Reports

Review of Methods for Estimating Populations with Administrative Data

The aims of this report are to review current research on combining administrative data and to set out a proposal for investigating the use of statistical modelling techniques for population estimation using administrative data.
Report by Southampton Statistical Sciences Research Institute
 

Population estimates for Data and Intermediate Zones Using Administrative Data

A report considering the use of statistical modelling techniques to produce small area population estimates using administrative datasets
Report by School of Geography, Queen Mary University of London
 

Population Estimates for Data and Intermediate Zones Using Administrative Data -
Part 2

A follow-up report to that produced by Queen Mary University of London. This report uses the same statistical techniques but applied to raw administrative data and compared to final Census 2011 data. Report by National Records of Scotland
 

Capture-Recapture Models for Population Estimates

This report describes a feasibility study for estimating population sizes using capture-recapture methods.
Report by University of St Andrews
 

The Use of Administrative Data Sets in Compiling Population Estimates Part 1

This report investigates the degree of consistency between population counts at data zone level taken from the 2011 Census and corresponding counts taken from administrative data sets as they stood at the time the census was taken
 
The Use of Administrative Data Sets in Compiling Population Estimates Part 2 This report investigates the degree of consistency between population counts by gender, five year age band and intermediate zone taken from the 2011 Census and corresponding counts taken from administrative data sets as they stood at the time the census was taken
 

Simulating capture-recapture data

This report investigates how the degree of stochastic dependency between inclusion in each of two surveys can be modelled by assuming that the propensity to participate in surveys follows a beta distribution
 
Inferences on the binomial parameter of a finite population This report proposes that the beta-binomial distribution can be used as the basis for a method to calculate the 95% confidence interval of the estimate of the proportion of a finite population which possesses a dichotomous attribute
 
Confidence intervals for capture-recapture data (A, B, C known) This report proposes a Bayesian argument for estimating the distribution of the number of persons not included in either of two enumeration samples such as a census and a census coverage survey. The distribution is modelled as a function of the degree of stochastic dependency between inclusion in the two samples