Register–based Statistics – Statistical Methods for Administrative Data, 2e

<p>Preface xi</p>
<p>Chapter 1&nbsp; Register Surveys An Introduction 1</p>
<p>1.1 The purpose of the book 1</p>
<p>1.2 The need for a new theory and new methods 3</p>
<p>1.3 Four ways of using administrative registers 5</p>
<p>1.4 Preconditions for register–based statistics 6</p>
<p>1.4.1 Reliable administrative systems 7</p>
<p>1.4.2 Legal base and public approval 8</p>
<p>1.5 Basic concepts and terms 10</p>
<p>1.5.1 What is a statistical survey? 10</p>
<p>1.5.2 What is a register? 11</p>
<p>1.5.3 What is a register survey? 13</p>
<p>1.5.4 The Income and Taxation Register 14</p>
<p>1.5.5 The Quarterly and Annual Pay Registers 16</p>
<p>1.6 Comparing sample surveys and register surveys 20</p>
<p>1.7 Conclusions 23</p>
<p>Chapter 2&nbsp; The Nature of Administrative Data 25</p>
<p>2.1 Different kinds of administrative data 25</p>
<p>2.2 How are data recorded? 26</p>
<p>2.3 Administrative and statistical information systems 27</p>
<p>2.4 Measurement errors in statistical and administrative data 29</p>
<p>2.5 Why use administrative data for statistics? 30</p>
<p>2.6 Comparing sample survey and administrative data 32</p>
<p>2.6.1 A questionnaire to persons compared with register data 32</p>
<p>2.6.2 An enterprise questionnaire compared with register data 34</p>
<p>2.7 Conclusions 36</p>
<p>Chapter 3 Protection of Privacy and Confidentiality 37</p>
<p>3.1 Internal security 38</p>
<p>3.1.1 No text in output databases! 38</p>
<p>3.1.2 Existence of identity numbers 39</p>
<p>3.2 Disclosure risks tables 40</p>
<p>3.2.1 Rules for tables with counts, totals and mean values 41</p>
<p>3.2.2 The threshold rule analyse complete tables! 43</p>
<p>3.2.3 Frequency tables are often misunderstood 44</p>
<p>3.2.4 Combining tables can cause disclosure 45</p>
<p>3.3 Disclosure risks micro data 45</p>
<p>3.4 Conclusions 46</p>
<p>Chapter 4&nbsp; The Register System 47</p>
<p>4.1 A register model based on object types and relations 47</p>
<p>4.1.1 The register system and protection of privacy 53</p>
<p>4.1.2 The register system and data warehousing 53</p>
<p>4.2 Organising the work with the system 54</p>
<p>4.3 The populations in the system 56</p>
<p>4.3.1 How to produce consistent register–based statistics 57</p>
<p>4.3.2 Registers and time 58</p>
<p>4.3.3 Populations, variables and time 59</p>
<p>4.4 The variables in the system 60</p>
<p>4.4.1 Standardised variables in the register system 60</p>
<p>4.4.2 Derived variables 62</p>
<p>4.4.3 Variables with different origins 63</p>
<p>4.4.4 Variables with different functions in the system 64</p>
<p>4.5 Using the system for micro integration 65</p>
<p>4.6 Three kinds of registers with different roles 70</p>
<p>4.7 Register systems and register surveys within enterprises 72</p>
<p>4.8 Conclusions 74</p>
<p>Chapter 5&nbsp; The Base Registers in the System 77</p>
<p>5.1 Characteristics of a base register 77</p>
<p>5.2 Requirements for base registers 78</p>
<p>5.2.1 Defining and deriving statistical units 78</p>
<p>5.2.2 Objects and identities requirements for a base register 80</p>
<p>5.2.3 Coverage and spanning variables in base registers 81</p>
<p>5.3 The Population Register 83</p>
<p>5.4 The Business Register 88</p>
<p>5.5 The Real Estate Register&nbsp; 93</p>
<p>5.6 The Activity Register 94</p>
<p>5.7 Everyone should support the base registers! 98</p>
<p>5.8 Conclusions 101</p>
<p>Chapter 6&nbsp; How to Create a Register Matching and Combining Sources 103</p>
<p>6.1 Preconditions in different countries 103</p>
<p>6.2 Matching methods and problems 105</p>
<p>6.2.1 Deterministic record linkage 105</p>
<p>6.2.2 Probabilistic record linkage 106</p>
<p>6.2.3 Four causes of matching errors 112</p>
<p>6.3 Matching sources with different object types 114</p>
<p>6.4 Conclusions 120</p>
<p>Chapter 7&nbsp; How to Create a Register The Population 121</p>
<p>7.1 How should register surveys be structured? 121</p>
<p>7.2 Register survey design 125</p>
<p>7.2.1 Determining the research objectives 125</p>
<p>7.2.2 Making an inventory of different sources 128</p>
<p>7.2.3 Analysing the usability of administrative sources 128</p>
<p>7.3 Defining a register s object set 131</p>
<p>7.3.1 Defining a population 131</p>
<p>7.3.2 Can you alter data from the National Tax Agency? 134</p>
<p>7.3.3 Defining a population &nbsp; primary registers 135</p>
<p>7.3.4 Defining a population &nbsp; integrated registers 136</p>
<p>7.3.5 Defining a calendar year population 137</p>
<p>7.3.6 Defining a population frame or register population? 138</p>
<p>7.3.7 Base registers should be used when defining populations 141</p>
<p>7.4 Defining the statistical units 142</p>
<p>7.4.1 Units and identities when creating primary registers 143</p>
<p>7.4.2 Using administrative objects instead of statistical units 144</p>
<p>7.5 Creating longitudinal registers the population 145</p>
<p>7.6 Conclusions 146</p>
<p>Chapter 8&nbsp; How to Create a Register The Variables 147</p>
<p>8.1 The variables in the register 147</p>
<p>8.1.1 Variable definitions 148</p>
<p>8.1.2 Variables in statistical science 149</p>
<p>8.1.3 Variables in informatics 150</p>
<p>8.1.4 Creating register variables check list 151</p>
<p>8.2 Forming derived variables using models 151</p>
<p>8.2.1 Exact calculation of values using a rule 152</p>
<p>8.2.2 Estimating values with a rule 153</p>
<p>8.2.3 Estimating values with a causal model 154</p>
<p>8.2.4 Derived variables and imputed variable values 157</p>
<p>8.2.5 Creating variables by coding 158</p>
<p>8.3 Activity data 159</p>
<p>8.3.1 Activity statistics 160</p>
<p>8.3.2 Activity data aggregated for enterprises and organisations 161</p>
<p>8.3.3 Activity data aggregated for persons multi–valued variables 161</p>
<p>8.4 Creating longitudinal registers the variables 165</p>
<p>8.5 Conclusions 169</p>
<p>Chapter 9&nbsp; How to Create a Register Editing 171</p>
<p>9.1 Editing register data 171</p>
<p>9.1.1 Editing one administrative register 173</p>
<p>9.1.2 Consistency editing is the population correct? 175</p>
<p>9.1.3 Consistency editing are the units correct? 178</p>
<p>9.1.4 Consistency editing are the variables correct? 180</p>
<p>9.2 Case studies editing register data 181</p>
<p>9.2.1 Editing work within the Income and Taxation Register 181</p>
<p>9.2.2 Editing work with the Income Statement Register 183</p>
<p>9.2.3 What more can be learned from these examples? 184</p>
<p>9.3 Editing, quality assurance and survey design 185</p>
<p>9.3.1 Survey design in a register–based production system 185</p>
<p>9.3.2 Quality assessment in a register–based production system 186</p>
<p>9.3.3 Total survey error in a register–based production system 191</p>
<p>9.4 Conclusions 192</p>
<p>Chapter 10 Metadata 193</p>
<p>10.1 Primary registers the need for metadata 193</p>
<p>10.1.1 Documentation of administrative sources 194</p>
<p>10.1.2 Documentation of sources within the system 195</p>
<p>10.1.3 Documentation of a new register 195</p>
<p>10.2 Changes over time the need for metadata 195</p>
<p>10.3 Integrated registers the need for metadata 196</p>
<p>10.4 Classification and definitions database 197</p>
<p>10.5 The need for metadata for registers 198</p>
<p>10.6 Conclusions 200</p>
<p>Chapter 11 Estimation Methods Introduction 201</p>
<p>11.1 Estimation in sample surveys and register surveys 202</p>
<p>11.2 Estimation methods for register surveys that use weights 203</p>
<p>11.3 Calibration of weights in register surveys 204</p>
<p>11.4 Using weights for estimation 207</p>
<p>11.5 Conclusions 208</p>
<p>Chapter 12 Estimation Methods Missing Values 209</p>
<p>12.1 Make no adjustments, publish value unknown 210</p>
<p>12.2 Adjustment for missing values using weights 214</p>
<p>12.3 Adjustment for missing values by imputation 215</p>
<p>12.4 Missing values in a system of registers 218</p>
<p>12.5 Conclusions 220</p>
<p>Chapter 13 Estimation Methods Coverage Problems 221</p>
<p>13.1 Reducing overcoverage and undercoverage 221</p>
<p>13.1.1 Coverage problems in the Population Register 221</p>
<p>13.1.2 Coverage problems in the Business Register 222</p>
<p>13.2 Estimation methods to correct for overcoverage 224</p>
<p>13.3 Undercoverage in the administrative system 226</p>
<p>13.4 Conclusions 228</p>
<p>Chapter 14 Estimation Methods Multi–valued Variables 229</p>
<p>14.1 Multi–valued variables 229</p>
<p>14.2 Estimation methods 232</p>
<p>14.2.1 Occupation in the Activity and Occupation Registers 232</p>
<p>14.2.2 Industrial classification in the Business Register 236</p>
<p>14.2.3 Importing many multi–valued variables 238</p>
<p>14.2.4 Consistency between estimates from different registers 242</p>
<p>14.2.5 Multi–valued variables what is done in practice? 245</p>
<p>14.2.6 Additional estimation methods 247</p>
<p>14.3 Application of the method 251</p>
<p>14.4 Linking of time series using combination objects 254</p>
<p>14.4.1 Linking time series 254</p>
<p>14.4.2 Changed industrial classification in the Business Register 256</p>
<p>14.5 Conclusions 258</p>
<p>Chapter 15 Theory and Quality of Register–based Statistics 259</p>
<p>15.1 Is there a theory for register surveys? 259</p>
<p>15.1.1 Statistical inference at a national statistical office 260</p>
<p>15.1.2 Theory–based methods or ad hoc methods 262</p>
<p>15.1.3 The survey approach and the systems approach 263</p>
<p>15.2 Measuring quality why and how? 267</p>
<p>15.3 Analysing administrative sources input data quality 271</p>
<p>15.4 Output data quality 278</p>
<p>15.5 The integration process integration errors 279</p>
<p>15.5.1 Creating register populations coverage errors 280</p>
<p>15.5.2 Creating statistical units errors in units 282</p>
<p>15.5.3 Creating statistical variables errors in&nbsp; variables 283</p>
<p>15.6 Random variation in register data 288</p>
<p>15.7 The register system and data warehousing 291</p>
<p>15.8 Conclusions 295</p>
<p>Chapter 16 Conclusions 297</p>
<p>References 301</p>
<p>Index 305</p>