1 IMPROVING THE QUALITY OF ADMINISTRATIVE REGISTERS BETTER SOCIAL WELFARE STATISTICS? Vibeke Sky ([email protected])[email protected] Unni Beate Grebstad ([email protected])[email protected].

1

1

IMPROVING THE QUALITY OF ADMINISTRATIVE REGISTERS

BETTER SOCIAL WELFARE STATISTICS?

Vibeke Sky ([email protected])

Unni Beate Grebstad ([email protected])

Division for Social Welfare Statistics

2

Reasons for adjustments

• Changes in the Social Welfare Services

• To provide the Public Administration with relevant, complete as well as accurate statistics

• Political decision-making in regard to social policy; dependent on good statistics

• Adjustments are usually necessary every 5th year

3

The modification processes of the two statistics

Changes in these areas follows a three-years run;

2012:Planning

2013: Registration

2014:• Data Collection • Processing • Publishing

Family counselling services and Child Welfare Services

4

Family counselling services and Mediation for parents

The purpose of the alteration:

• Relevance

• Quality

• Prepare automatically extraction of man-years from registers

• Cost effectiveness

• Non-excessive reporting burden

5

Child Welfare Services statistics

Why alteration?

• New categories of measures the last 5 years

• Need for good time-series must be held against the need for change

• Coherence and comparability

• Longitudinal data for researchers

• Easing the task burden

6

Statistics Norway and the readjusting processes

• File description/electronic forms

• Checking program

• Internal editing system ( e.g. ISEE)

• Figures

• Reporting guidelines/instructions

7

Challenges; Family counselling

• Relevant legislation

• Users needs (Directorate for Children, Youth and Family Affairs)

• Time series vs. change

• Extracting man-years from registers

Results: Family counselling

• Increasing the relevance of the statistics (e.g. statistics on groups, children's participation)

• Increase the quality of statistics on man-years*

• More effective and less resource-draining data collection and processing of data

*Contracted man-years adjusted for long term leaves

9

Challenges – Child Welfare Services

• Old categories of measures

• The Norwegian letters «ÆØÅ»

• ISEE (editing system) had to cope with a hierarchical file structure

10

Results - Child Welfare Services

• A database summarizing monthly register status

• Eventually model a new and plausible event history database

• The services themselves will revise the errors

11

Better quality in Child Welfare Data?

• We will publish some new and more relevant variables (e.g.notifications, measures involving other services)

• Better opportunities for analysis through more dynamic data for researchers

• New and updated statistics

12

Better quality of statistics on family counselling services and mediations?

• A more updated statistical picture of the services

• Increased focus on children and group therapy

• More stable and thus more reliable data capture of man-years from register

13

Conclusions about quality

• Evaluation of the readjustments with

regard to the well-known

«European Statistics Code of Practice»

• Point 12.3: «Revisions are regularly

analyzed in order to improve

statistical process»

• Our paper is a first step towards analyzing the revision process