Quarterly National Household Survey (QNHS) Anonymised Microdata File – Frequently Asked Questions

If you have any queries on the QNHS Anonymised Microdata File (AMF) please read through these frequently asked questions (FAQs) which cover the most common queries received. It is important that you familiarise yourself with the codebook (available here ) and understand the data you have requested to use before referring questions to the ISSDA or CSO. We will endeavour to help users where possible and where resources permit.

FAQs are grouped into two categories:

Information contained in the data

Q. I understand that age is not given for anonymity reasons but our study aims precisely at comparing young workers aged 18-21 to those aged 22-25. Would it be possible to receive a variable identifying these two categories?

Q. There are many questions on QNHS questionnaires, but most of these variables are missing in the SAS datasets. I'm looking for particularly important variables such as: Month the participant moved to Ireland (YRRESIDEN), Age completing highest level of education (EDAGE), and all income variables (INCPERM,WEEKPAY, FORTNPAY, MONTHPAY). Do you have these information that I am looking for?

A. The Anonymised Microdata File (AMF) is the only data lodged with ISSDA. There is the Researcher Microdata File (RMF), which is more detailed:

"RMFs are unit record files that do not contain direct identifiers but where the risk of disclosure through indirect identification is considered to be significant. RMFs are not statistical products, as CSO products relate to aggregated statistical analysis, but are research files that are made available to persons authorised to access such files under the Statistics Act, 1993 subject to strict criteria. Currently, RMFs are the only means through which access can be granted to business microdata. In addition, RMFs have also been prepared to allow access to household based data at a more detailed level than that possible through AMFs.

Access to RMFs is strictly controlled and can only be granted within the framework of the Statistics Act, 1993 and must be approved by the Director General of the CSO. Any persons accessing such files must be appointed as Officers of Statistics, sign a Declaration of Secrecy and agree to abide by the Standard Agreement which outlines the specific conditions under which the microdata will be made available."

You have the option to apply to the CSO directly to gain access to the RMF (Research Microdata File) but note that each application is considered on its individual merits as the CSO will at all times consider the validity of the proposal you submit and making an application does not imply that access will be granted as the CSO is extremely protective of the data it is mandated to collect.

Information on RMF assessment criteria: http://www.cso.ie/en/aboutus/dissemination/accesstomicrodatarulespoliciesandprocedures/assessmentcriteria/

Application procedure:

http://www.cso.ie/en/aboutus/dissemination/accesstomicrodatarulespoliciesandprocedures/applicationprocedure/

Contact details for QNHS: http://www.cso.ie/en/qnhs/contactqnhs/

NB: Access to the RMF data is by on-site appointment only.

Q.  I was just reviewing the data and I was wondering is there any wage data in this dataset? If not is there anyway I could use as a supplementary set?

A. We don't have any income data on the QNHS Annonymised Microdata file (AMF) that we lodge with ISSDA. We do have a variable on the Research Microdata file (RMF) for the QNHS. That variable only indicates the income decile range that a persons take home pay falls into and not their exact income. Exact income is not collected by the QNHS. Researchers apply directly to the CSO for access to the QNHS RMF datafile but as the data is less anonymised than the AMF the application procedures and controls are a lot more stringent than for the AMF. When applying for access to the RMF researchers have to specify their research question and why they require access to the RMF. They also need to be able to provide a secure desktop PC in order to be able to access the RMF securely and need to meet with the Data Office from the CSO prior to accessing the RMF in order to be briefed on data security and confidentiality issues. More information in relation to RMFs including the CSO policy in relation to access, the application procedure, assesment criteria and application form are available at: http://www.cso.ie/en/aboutus/dissemination/accesstomicrodatarulespoliciesandprocedures/researchaccesstomicrodatafiles/.

Q. It would be very helpful if I could identify people (children in particular) according to their age. However,the age of surveyed people is disseminated in 5-year bands. Is there any way I can access a file reporting a disaggregated age variable?

A. The use of age bands is an anonymity measure. You can make an application to access the RMF data (see previous FAQ).

Q. How long is the RMF application process? I am on a tight deadline.

A. There is no set time for the RMF application process. If you do intend to apply for RMF access you can note that you should apply, at the very least, 4 weeks in advance.

Q. What kind of information is available on the Researcher Microdata File (RMF) which is not available on the Anonymised Microdata File (AMF)?

A. Additional variables which are part of the QNHS survey are contained in the RMF files (e.g. type of temporary work, the type of second job, hours worked in the second job etc.); some variables are offered in a more disaggregated form (region of residence is supplied at NUTS3 rather than NUTS2 level, industry of employment is supplied at 2 digit NACE level rather than NACE sectoral level etc.); RMF files are made available for all quarters of each year (Q1, Q2, Q3 and Q4 are all supplied).

Q. The data I have received is only labelled by year so I am unclear as to what quarter it is you normally issue? I notice you hold data for Q2 and Q4 so I'm not sure which I was issued with.

A. Q2 and Q4 are combined as a single file. You should ensure that you are familiar with the codebook supplied so that you understand how to use the data – in this instance individual quarters within the year can be distinguished by using the SURVEYQTR variable.

Q. It seems to me that survey data are only available for two quarters each year (Q2 and Q4), is this correct?

A. The CSO has a policy of only allowing access to Q2 and Q4 in order to limit the ability of researchers to link data across quarters, which could potentially jeopardise the anonymity of the respondents.

Using the data

Q. There is so much information in the Anonymised Microdata File - how do I get started using the data?

A. The first step is to check the various codebooks that are available from the ISSDA website above. 

Q. In what formats is QNHS data available? My preference is for SPSS.

A. QNHS is available in SAS but not SPSS. It is possible to open the files in SPSS. Please read the help files available with your statistical programme or search for data format change guidelines online. We will have a dedicated help section for this shortly.

Q. Although I am familiar with using SPSS and the idea of weighting in order to match the sample answers to the population estimates, I am getting a bit confused with what the frequencies actually reflect and how to look at for instance the household level of various variables.

A. You should ensure that you are familiar with the codebook supplied so that you understand how to use the data – the detail of which weights to use are contained within the codebook and the appropriate weights to use then depend on the particular usage you have for the data. For example, results at an individual level should be weighted using the variable GF; if you wish to consider results at a household level then the variable HH_GF should be used for weighting.

Q. I am trying to calculate the mean hours worked by sector and need to classify hours worked as a ratio variable. However, the dataset only allows me to classify it as ordinal or nominal. Advice welcome!

A. You should ensure that you are familiar with the codebook supplied so that you understand how to use the data – the hours worked variable can be converted to a numeric variable by you if you wish to produce mean hours worked.

Tools