Back to Top

 Skip navigation

What is a frame - why and how are they used in the collection of official statistics? 

One of the main goals of statistics is to determine the characteristics of a population and describe it in a numerical way. These characteristics are very often investigated by surveying units belonging to the population. In official statistics the main focus is on special populations, so-called target populations. To do this efficiently, units of a set, ideally covering the target population as best as possible, are selected. Having identified a set, we talk about a frame. (Eurostat, 2019)

How did the Central Statistics Office (CSO) create its Safety of the Person frame?

The CSO collects data for statistical purposes from a wide range of sources, including directly from survey respondents and from a wide range of Public Sector Bodies (PSBs). Data collected from PSBs is known as administrative data.

As a Census of Population typically happens only every five years, for this frame, the existing Census data (2016) was supplemented with more up-to-date administrative data. This work is enabled by linking administrative data to create a frame of persons in Ireland from which to draw the sample for the Safety of the Person Survey.

What data sets were used to create this frame?

In creating the Safety of the Person frame, a wide number of sources were used, both from survey data collected directly by the CSO and from other statistical datasets received from PSBs.

A full list of agreements and Memorandums of Understandings (MOUs) that the CSO has with PSBs can be found at the following link: Memorandums of Understanding

How was data from public sector bodies used in creating the frame?

In order to create a complete, accurate and up-to-date frame, there are three essential elements:  (Eurostat, 2019)

  1. “Key Variables (IDs)”, (e.g. VAT number, social security number, personal ID);
  2. “Communication Variables” (e.g. Address, telephone number, e-mail); 
  3. "Classification Variables” (e.g. gender, age group, region).

For the Safety of the Person frame, the following variables were used:

  1. PPSN – only used to ensure there was no duplication in the sample and to confirm that the person is likely to be a resident in Ireland – not used in the final sample and not linked to responses. 
  2. Name & Address - used to contact the person to invite them for participation in the survey and also to stratify by Local Authority ensuring the sample is spread correctly around the country in line with the overall population.
  3. Gender and Age Group – used as part of the sampling process to ensure that the overall sample contains the correct breakdown of gender and age that is representative of the general population.

What is the legal basis for the creation of a frame?

The CSO is Ireland's national statistical office, and our purpose is to impartially collect, analyse and make available statistics about Ireland’s people, society, and economy.

Specifically, our mandate under the Statistics Act, 1993 is: "The collection, compilation, extraction and dissemination for statistical purposes of information relating to economic, social and general activities and conditions in the State".

Section 30 of the Statistics Act, 1992 provides the legal basis for the CSO to access the records of public authorities for statistical purposes.

Who in the CSO has access to this frame?

Statisticians who require access to identifiable data for statistical processing purposes must comply with extensive governance procedures and have their business case signed off at management board level. One area within CSO, the Administrative Data Centre (ADC), is responsible for the frame. The ADC is a dedicated business unit, which is a specialist team responsible for decrypting, processing, pseudonymising and storing records in a format accessible for statistical analysis. ADC statisticians working on the frame then provide the survey statisticians with a sample containing only the necessary variables for administrating the survey, with data minimisation being at the core of the decision in relation to the variables made available to these statisticians. Survey area statisticians can and will never have access to the full frame.

Names and addresses for the sampled respondents are only used to contact them for surveying purposes. Once survey data is received, names and detailed addresses are removed from the file so that the link between a named individual and their survey response is irrevocably broken. In addition, access to survey sample containing personal data is restricted to named CSO staff working directly on the survey sample, and only where strictly necessary.

Staff working on the sample will not have access to any responses. Analysis of the responses will be done on anonymised data meaning all direct identifiers such as names and addresses are removed prior to analysis.

What other information does this frame hold about me?

The survey frame only contains the variables used for selecting a representative sample. To obtain the sample PPSN, Name, Address, Age group and Gender were used. These variables were used to ensure that the random sample selected is representative of the general population in terms of age groups, gender, and county of residence.

After participants are contacted and provide a response to the survey, all further analysis will be done on pseudonymised data meaning all direct identifiers such as names and addresses are removed prior to analysis.

Why has the CSO chosen to address a specific person in the household rather than use the general household-based approach?

For this survey, in the best interests of data quality a methodological decision was taken to use a person-based sample, derived from primary and administrative sources accessible to the CSO. Certain groups or subsections of the population can sometimes be difficult to contact. Choosing to address specific persons in a household minimises the risk that certain groups/cohorts of the general population are underrepresented in the results of the survey. For this Safety of the Person Survey (which covers the broad range of unwanted sexual violence experiences a person may or may not have had), given the very sensitive nature of the survey, it is important to directly contact the sampled respondent rather than attempt to sample at the household doorstep.  There are significant ethical considerations underpinning why a direct contact approach such as that a respondent is safer than the traditional household sampling approach.

How did the CSO select the sample from the frame?

For this Safety of the Person survey, a technique known as Stratified Random Sampling was used. Stratification ensures that each population group is represented accurately in the sample. The stratification was done by age, gender, and Local Authority area, meaning survey participants are correctly spread across all ages, genders and around the country.  

Has the CSO shared data with, or sold this data to, any third parties including government departments, commercial enterprises, or bodies?

The CSO will never share any personal data with any third parties – State, commercial or otherwise. In certain specific cases, pseudonymised versions of data may be made available to approved researchers under strict researcher protocols and detailed governance procedures.

One of the core principles governing the operation of National Statistical Institutes is the protection of the confidentiality of all information supplied by data providers. The CSO’s ability to compile Official Statistics is based on the extent to which individuals and companies trust the CSO with sensitive information and the CSO’s guarantee of confidentiality for all data providers is built on the fundamental requirement of non-disclosure of confidential data as set out in national and EU statistical legislation.

How is my data protected?

All CSO staff sign up to, and abide by, the CSO Data Management Policy as approved by the Office’s Confidentiality and Data Security Committee (CDSC). Stringent confidentiality rules apply under the Statistics Act to all staff of the CSO, who are designated as Officers of Statistics under Section 20(a) of the Statistics Act, 1993.

As well as the strict legal protections set out in the Statistics Act, 1993, and other existing regulations, we are committed to ensuring compliance with all data protection requirements. including the General Data Protection Regulation (GDPR).

What about GDPR legislation?

Data Protection Transparency Notice for the Safety of the Person Survey can be accessed here Safety of the Person Survey.