Frequently asked questions (FAQ)

The FDZ at IQB is also involved in Forum4MICA. This is a platform for public exchange on topics related to the availability and use of research data in the social, educational, behavioral and economic sciences and serves the dialogue between researchers as well as research data centers. It is available to anyone who is looking for topic-related information and/or wants to actively participate in the dialogue. The forum adds an interactive component to existing user services and documentation materials. With each new question, answer and comment, the generally accessible knowledge base of Forum4MICA grows and offers all relevant functions of an online discussion platform.

Our presence there is linked here.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++

General questions on the application for data access, data usage and data submission to the FDZ at IQB

back to overview

General questions on the application for data access, data usage and data submission to the FDZ at IQB

Accreditation and certification

What does it mean that the FDZ at the IQB is accredited (RatSWD) and certified (CoreTrustSeal)?

Accreditation by the RatSWD

The FDZ at IQB has been accredited by the German Data Forum (RatSWD) and bases its work on its criteria.

The RatSWD advises the Federal Government and the governments of the federal states on issues relating to the expansion and improvement of the research data infrastructure for the empirical social, behavioural and economic sciences. It is an institutionalised forum for exchange and dialogue between science and data production on improving access to high-quality and scientifically interesting data. Its tasks include

  • Strategic development of the research data infrastructure
  • Representation of interests of data producers and users
  • Accreditation and evaluation of the work of research data centres
  • European and international networking of research data infrastructures
  • Scientific and research-strategic exchange through dialogue events, symposia and the Conference for Social and Economic Data (KSWD)

Certification with the CoreTrustSeal (CTS)

The FDZ at IQB is certified with the CoreTrustSeal for trustworthy repositories. Here you can download the self-disclosure of the FDZ at IQB with comments of the reviewers.

Funding agencies are increasingly demanding that data resulting from projects they support be archived in the long term in accordance with an open data and data management policy and made available for subsequent use. Researchers must therefore ensure that their data can continue to be accessed reliably and unambiguously in the future and are therefore increasingly looking for trustworthy repositories. The CoreTrustSeal Foundation (CTS) - a legal entity under Dutch law with headquarters in The Hague, the Netherlands - is a non-profit organisation that promotes sustainable and trustworthy data infrastructures. It has been in existence since January 2018 and is a merger of the predecessor certificates, the Data Seal of Approval (DAS) and the ICSU World Data System (ICSU-WDS). In practice, it is a kind of self-governing body of international experts from the field of research data archiving and data security who work on a voluntary basis (CoreTrustSeal). The so-called "CoreTrustSeal" is a certificate for data archives that reflects trustworthiness, for example with regard to data protection and data security, but also to quality assurance of the archiving processes.

Such a certification helps to improve the quality and transparency of processes and to increase awareness of standards and compliance with these standards. In this way, processes could be optimised and workflows standardised and harmonised.

The certification of the FDZ at IQB with the "CoreTrustSeal" takes into account the efforts to permanently guarantee adequate quality assurance as well as a constant further development of the research data infrastructure and to be perceived as a trustworthy archiving partner also in the future. The certification proves to users and sponsors that the FDZ at IQB works seriously, according to established standards and internationally compatible.

back to overview

Search

Where do I look for data sets?

Click here to search.

back to overview

How do I get an overview of the data sets that the FDZ at IQB offers?

Here is a first overview of the studies available at the FDZ at the IQB.

back to overview

How do I check which studies fit my topic?

When searching for suitable data for your research project, you should search for both sample characteristics (e.g. year of birth or survey period) and constructs (e.g. self-concept or special educational needs). You can use reports, scale manuals, blank data sets and the study search for this purpose. If you have specific questions, it is worth taking a look at the notes on the respective study pages on our homepage. If you have further questions, we will be happy to advise you by phone or e-mail.

back to overview

In which (IQB) surveys are there information on the textbooks used?

For example, in addition to the IQB National Assessment Study 2011 and 2012, TIMSS 2015 also provides information on the books used. In addition, the QuaSum study records information on the textbook and on the use of the textbook in class, but the textbook information has been anonymized (i.e. the data set contains only consecutive numbers instead of textbook names).

You are welcome to use the search function on the FDZ website and search for constructs. If you have further questions, we will be happy to advise you by phone or e-mail.

back to overview

Can I search for constructs?

Constructs can be understood here as umbrella terms that divide certain properties, behaviors, item words, etc. into classes of similar meaning. Thus, constructs serve to summarize the contents of the studies in broader categories. This is convenient for a quicker overview of the studies contents. Furthermore, constructs are suitable for comparing the contents of different studies and for searching across studies.

The FDZ at IQB uses the construct scheme of the Database on Quality of School (DaQS) for the systematic collection of constructs. In addition, the scale names as listed in the scale manuals of the original study are stored.

In the search of the FDZ at IQB, constructs can be searched for across the individual studies.

back to overview

Data usage

Who is entitled to apply for scientific use files (SUFs) and campus files (CFs)?

Applications are open to researchers and students, although they must be institutionally linked to a university or a publicly funded research institute. It is assumed that the data will be used exclusively for non-commercial scientific purposes.

back to overview

What kind of data do I receive as a researcher?

By default, the FDZ at the IQB provides scientific use files (SUFs). The SUFs have already been anonymised so that no personal reference can be made. SUFs are made available to data users by the FDZ at the IQB after conclusion of data use agreement, which regulates further conditions of use.

Scientific analyses with data versions that are subject to special access restrictions for data protection reasons are accessible via a specially protected path (e.g. remote computing). In such analyses, the FDZ at the IQB checks the results of the analyses for compliance with data protection and contractual provisions before they are made available to researchers.

And then there are the Campus Files (CFs). CFs are anonymised data sets that are designed for use in university teaching. They are particularly suitable for teaching statistics and empirical research methods.

Further information can be found here.

back to overview

How do I request data for my research?

A short research outline with a description of the theoretical foundations, derivation of the questions or hypotheses and planned analyses must be submitted.

Please always fill in all fields of the application form. In the field "Data/Variables", specify your central dependent variables and independent variables as well as central covariates. We need this information, among other things, to be able to check whether your project conflicts with a blocked research question. If you are requesting access to education monitoring data, it is important to indicate if you would like to use the federal state identification variable, because the use of this variable is subject to separate approval. Please then also indicate what you need the state identification for (e.g. for comparisons between the federal states, as a covariate for control purposes, to feed in contextual characteristics or other third-party variables, to form and compare aggregated groups of states or to describe the sample). After application approval and conclusion of a data use agreement, you will always receive the complete data sets (with the exception of variables that are questionable from a data protection point of view). It is therefore not necessary to list all the variables you want to look at individually.

Your application should be at least 2-3 pages and no more than 10 pages long and follow the FDZ at the IQB's Guidelines for the submission of applications.

For further information on data applications, please see here.

back to overview

How are the Scientific Use Files made accessible?

In principle, the data from the FDZ at the IQB are available in anonymised form, i.e. a direct reference to a person cannot be established. Depending on the sensitivity of the data, there are data sets with varying degrees of anonymity that are made available via different access channels.

By default, the FDZ at the IQB provides the scientific use files (SUFs) in SPSS or STATA format. In these data sets, all information that would allow a potential re-identification of individual people was recoded or removed from the original data. This may be the case, for example, if certain variables have values that originate from five or fewer people, or if geographical information was collected. SUFs are made available to data users by the FDZ at the IQB via the Humboldt University file exchange portal after conclusion of a data use agreement.

Scientific evaluations with data versions that are subject to special access restrictions for data protection reasons (e.g. federal state information) are possible via a specially protected access path (via the controlled remote computer access JoSuA, provided by the IZA in Bonn). In such analyses, the FDZ at the IQB checks the results of the analyses for compliance with data protection regulations before they are made available to researchers.

To get a first impression of a data set i.e. what was collected (variable names, labels), so-called blank data sets can be downloaded from the respective study website.

back to overview

Which requirements do I have to meet to receive data from the FDZ at the IQB?

If the data offered at the FDZ at the IQB are suitable for your research (analysis and re-analysis) or teaching and you are institutionally linked to a university or a publicly funded research institute, you do not need to meet any further requirements to submit a application for data access.

As a student, you must specify a supervisor, because only this supervisor can stamp your data use agreement with the stamp of your university.

back to overview

What data on individual schools are available at the FDZ at the IQB?

Unfortunately, it is not possible to use external data at the level of individual schools. This would contradict data protection regulations as well as agreements with the participating schools. For this reason, the FDZ at the IQB does not have any information on the correlation between the school IDs and the school names or on the location of the schools.

back to overview

What do data users have to consider after the end of the project?

The data must be deleted after the end of the project.

Please also send us a copy (preferably electronically) as soon as possible after the project is completed. This also refers to student theses that are only published within the university (e.g. bachelor's, Master's or diploma theses). It should be noted that both data depositors (see the proposal for citation on the study pages and in the data use agreement) as well as the FDZ at the IQB (e.g. in the acknowledgements, in data or in a footnote) must be cited according to our specifications.

References to citation can be found here ("Citation/Research results").

back to overview

Campus Files

What are Campus Files?

Campus Files (CFs) are anonymised data sets designed for use in academic teaching. They are particularly suitable for teaching statistics and empirical research methods. Compared to SUFs, CFs have significantly smaller sample sizes and a reduced selection of variables. Campus Files can be applied for without the conclusion of a data use agreement. Please refer to our guidelines for the CF application procedure (www.iqb.hu-berlin.de/fdz/Datenzugang/CF-Antrag/NutzungsordnungCFs).

Further information about Campus Files can be found here.

back to overview

Can I use the data from the FDZ at the IQB in teaching?

Yes, definitely! That's what the Campus Files (see "What are Campus Files?") are for. In addition to providing a valid university e-mail address and a purpose of use in teaching, applicants must agree to the Regulations for the Use of CFs in order to gain access to the CFs.

back to overview

How do I request data for academic teaching?

You can find information on how to apply for data here.

back to overview

Online application procedure

Where can I find information on the online application procedure? How does it work?

Information about the scientific use files application can be found here.

Information about the Campus Files application can be found here.

back to overview

Information for students

Can I get data for my seminar papers and theses from the FDZ at the IQB?

Standard case: Campus Files (CFs)

Campus Files are anonymized data sets that have been specially designed for use in university teaching. They are particularly suitable for working on statistical or methodological questions.

If you want to use Campus Files for your seminar paper or thesis, you only need to enter your university e-mail address and the purpose of use of the desired data in the corresponding application form and select the desired study(s). We will only check the institutional affiliation (based on the e-mail address), we will not check the content of your project.

After a positive check of the e-mail address, you will immediately receive an e-mail with a download link to the desired campus files. Please note that this link is valid for a limited period of time. If you do not retrieve your Campus Files in time, you will have to request them again.


Suitable for larger projects: Scientific Use Files (SUFs)

If you do not want to use Campus Files for your seminar paper or thesis (e.g. Bachelor, Master), but Scientific Use Files, a more extensive application is required (see Applying for SUFs), which will be checked for content at the FDZ at IQB.

Students (also doctoral students without employment at a university) please indicate in the application form their private address as well as the contact data (name, institution, office address and e-mail address) of the person in charge of their project, even if this person does not need his/her own data access. Please remember to obtain the consent of the persons concerned to pass on their data.

In case of approval, you will be sent a data usage agreement for the requested data. Please note that the person supervising your work must also sign the contract, and allow sufficient time for the contract to be processed. Once the contract is signed, you will receive access to the requested data.
Processing time for SUF applications

When preparing your thesis timeline, please keep in mind that it may take 4-6 weeks from receipt of the request to data submission. In exceptional cases, e.g. if the contract documents are in the mail for a very long time (abroad), processing may take even longer.
Software requirements

By default, we provide the data as R, SPSS or STATA files. You can select this when submitting your application. If you need more than one or other data formats, please contact us.
Consulting

If you have any further questions about working with FDZ data, we will be happy to advise you:

    Dr. Aleksander Kocaj
    Christin Rüdiger
    Annegret Rucker

It is best to send your request by e-mail to our central address fdz@iqb.hu-berlin.de; we will then contact you by e-mail or, if you wish, by telephone.

back to overview

Information on the application period

How long does it take to receive the data I have requested?

The following applies to scientific use files: Upon receipt of an application for data, we first check whether the formal criteria for application approval are fulfilled. This usually takes 2 weeks.

In exceptional cases, for example if it is unclear whether the application contradicts contractual agreements with the data owner of the study, the FDZ at the IQB reserves the right to involve external experts, the respective data producers or the data owners in the review process. In such cases, an extension of the approval process by several weeks can be expected.

If the research project includes state comparisons that are new and have not yet been published, a longer processing time is also to be expected, since the application will be subject to a scientific review due to the new regulation of 20.09.2012 (in the version of 31.01.2019) (details of this procedure are described here).

After your application has been reviewed, a data usage agreement will be drawn up and signed by both parties. You will then receive the requested data. Please note that the total duration of the procedure can be about 8 weeks in standard cases.

For Campus Files: Immediately! An application for the use of CFs can be submitted by researchers at colleges, universities or non-university scientific research institutions for research, teaching and qualification purposes as well as by students. The only prerequisite is a valid e-mail address of a university institution as well as agreement to the usage regulations of the FDZ at the IQB.

back to overview

Blocking notices

What are blocking notices? How do they influence my research questions and application for data access?

For some studies, we have blocking notices. This means that certain questions have been blocked by the data depositors,  because they are still writing manuscripts or completing qualification papers. For this reason, we check at the time of application whether the question applied for violates a blocking notice. If this is the case, the application cannot be approved until the blocking notice has expired. However, this happens very rarely. You can find out whether a study is subject to a blocking notice on our study pages.

back to overview

There are research questions concerned by blocking notices in the data set I want to apply for. What does that mean? Can I no longer use this data set for my research?

Some of our studies have blocking notices. This means that certain questions have been blocked by the data depositors. No variables or scales are deleted from the data sets. Instead, when a proposal is submitted, we check whether the proposed research question contravenes an ongoing research project of the data depositors. In the case of queries regarding blocking notices, the FDZ at the IQB contacts the project manager/scientific director of the project. You will therefore not receive any data if you apply for the same or comparable research questions. Applicants will be informed of the existing blocking notice and will not receive any further information on the background of the blocking notice. As a standard feature, our website contains a general notice that blocking notices exist for the chosen study and that these will only be issued upon request.

back to overview

Handling of the federal state identifiers

I am doing research on a topic for which it is necessary to compare different German federal states. What do I have to bear in mind?

1.)    Not all studies contain information on the federal states. If this is the case, it is noted on the study page under "Access information".

2.)    Access to the federal state variable, if available, is regulated in a special way to protect scientific and ethical quality standards. This means:

3.)    The scientific use files of the FDZ at the IQB do not contain federal state variables by default, unless agreed upon in the application.

4.)    The comparison of two or more groups of countries (e.g. countries with vs. without early tracking or countries in which certain reforms have taken place vs. have not taken place) is possible and must be described in the application for data access.

5.)    The use of the federal state variable as a control variable (for example in a regression model) is possible and must be described in the application. In that case data access is provided via the secure remote access application JoSuA hosted by the IZA (https://iqb.josua.iza.org).

6.)    In the cases described above (4. and 5.) single federal states must not be identifiable in published analyses or results.

7.)    In the case of applications in which federal states are explicitly identified and compared to investigate novel research questions (i.e. questions that do not replicate previously published comparisons of federal states), a comprehensive application procedure including a scientific peer review is required. This is based on the DFG's review procedure (for further details see the PDF Rules of Procedure for Innovative State Comparisons).

8.)    Publishing analyses and results based on the analysis of a single federal state requires the consent of the respective federal state (for further details, see the PDF Rules of Procedure for Innovative State Comparisons).

back to overview

How do I get the  anonymised federal state variable as a covariate for multi-level analyses carried out?

You can use the federal state variable for your multi-level analyses via the secure remote access application JoSuA hosted by the IZA. The statistics programs R and Stata are accessible via the JoSua user interface. If you include the federal state variable as a covariate in your multi-level models, you must bear in mind that publishing results or any other content which makes a single federal state indirectly identifiable is not allowed. Unfortunately, we cannot provide you with an anonymised federal states variable in SUF data sets, since it would be too easy to identify single federal states through their unique features.

back to overview

Data linkage

I would like to link data from other sources with data offered at the FDZ at the IQB. What do I have to consider?

1) It is possible to link data offered by the FDZ at the IQB with data from other sources (at the level of federal states), if this does not cause any problems in terms of data protection law. Therefore, a planned linking of the data sets should already be described in the application for data access. Links can only be established for purposes described in the proposal.

2) Unfortunately, regional data (e.g.the percentage of employed people) can currently only be linked to the data sets available at the FDZ at the IQB at the level of the German federal states, but not at the level of cities, municipalities or districts.The FDZ at the IQB does not have any studies with  regional identifiers - except federal state identifiers - in its database.

3) The videography data of the study German English Student Performance International (DESI, DOI: 10.7477/6:1:1) are archived at the Research Data Center Education at the DIPF, but can be linked to the questionnaire and competency data archived at the FDZ at the IQB. The same applies to the videography data of the study Competence Acquisition and Learning Prerequisites (KuL, DOI: 10.7477/287:1:0). A planned linkage of these data sets should be mentioned in the application for data access.

back to overview

Data use agreement

What does the data use agreement say?

In order to gain access to scientific sse files, data users must sign a data use agreement. A sample contract can be viewed PDF here.

The data use agreement sets out the terms of use, access to the data and information on citation proposals including doi. The data use agreement must be signed by the users and stamped by the institutionally affiliated institution. For applications by students, the supervising person must also sign.

back to overview

Novel federal state comparisons

What do I need to consider if my research project includes new and previously unpublished federal state comparisons?

Information on the topic of innovative federal state comparisons can be found here.

back to overview

Unpacking the transferred data

How do I unpack the data that I have received from the FDZ at the IQB?

After signing the data use agreement, you will receive an e-mail with a link to the Humboldt University file exchange portal, where your requested data is stored in zipped format. The folder is password protected - you will find the password in an additional the e-mail sent to you by the FDZ at the IQB.

Please use a PC-internal program to unzip the data.

back to overview

Remote computing

My research question demands analyzing protected data. What do I have to consider?

For handling protected data that cannot be made available in the scientific use files (SUFs) we offer to use the secure remote access application JoSuA hosted by the IZAData users do not have direct access to the data, but can only send command syntax via the online portal and receive the output.

JoSuA allows analyses in internal mode and in presentation/publication mode. Analysis results obtained in internal mode may not be used for publications. Outputs in the internal mode automatically get a grey watermark with the text "Do not print or copy these results or make them available to persons outside the use agreement for this project". This output is only to be used for intermediate steps or control calculations. This output cannot be downloaded for further use, but appears as a pop-up window. If an output is defined by the data users as  presentation/publication mode, the output will only be made available to the data users after it has been checked by the scientific staff of the FDZ at the IQB. Analysis results from the presentation/publication mode may be published by data users after approval by the FDZ at the IQB.

Remote access is set up for one person per project for 4 months (extension possible if required).

back to overview

Submission of data

Do you have data sets containing competence measurements that you would like to archive?

On our data submission page you will find all the important information about how the data should be transferred.

back to overview

Research data management

You have questions about research data management?

The joint project German Network of Educational Research Data (VerbundFDB) offers a very good overview of the topic research data management (RDM). The Humboldt-Universität zu Berlin also offers answers to all aspects of RDM.

back to overview

Can I get a letter of intent from the FDZ at IQB e.g. for a grant proposal?

In accordance with many of the funding conditions imposed by research funders, projects must make the collected research data available after the end of the project for the purposes of subsequent use, transparency and traceability of the results. If it is necessary to prove that contact has already been made with a data centre, data producers are welcome to contact the German Network of Educational Research Data (VerbundFDB), which is a partner of the FDZ at the IQB. The German Network of Educational Research Data agrees in that letter to support the project in questions of research data management (RDM) within the scope of its possibilities and to save the data produced in the project in accordance with its collection policy and to make it available for reuse. The FDZ at IQB especially aims to archive, document and make visible data and studies concerned with academic competencies and educational achievement.

back to overview

Data protection

How does the FDZ at the IQB handle your personal data?

The IQB's data protection declaration can be found here.

The FDZ at the IQB implements  data protection regulations when subscribing to the newsletter.

When applying for data, the FDZ at the IQB also implements the data protection regulations according to the DSGVO.

If you have unresolved problems with the use of the FDZ at the IQB's data that you were unable to resolve in exchange with the responsible contact person from the FDZ at the IQB, you can also contact the complaints office of the RatSWD, with which the FDZ at the IQB is accredited.

back to overview

Notes on general and methodological issues concerning the data sets archived by the FDZ at IQB

Data availability

Which data are available at the FDZ at the IQB? How can I stay informed when new data (versions) are available?

An overview of all studies that are currently available can be found here.

You can subscribe to our FDZ at the IQB newsletter. The newsletter announces regularly new data sets as well as new versions of data sets that are available at the FDZ at the IQB.

back to overview

There are empty variables in my data set. Why is that?

Some variables are emptied or recoded for data protection reasons. If you need these variables for your analyses, you can access them via the secure remote access application JoSuA hosted by the IZA.

back to overview

Methodological training and workshops

What training programme does the FDZ at IQB offer?

We offer semi-annual academies on methods of empirical educational research. The aim of these training courses is to impart knowledge of methods for analysing data from educational studies. For futher information please see here.

back to overview

Do you also offer workshops for my entire research group?

The scientific staff of the FDZ at IQB are available on request for workshops on topics from the areas of research data management, secondary analysis and open science.

back to overview

Studies on educational monitoring

What is meant by "studies on educational monitoring"?

One of the core tasks of the FDZ at IQB is the provision of  data sets of national and international school achievement studies, which are primarily used for educational monitoring in Germany within the framework of the strategy of the Standing Conference of the Ministers of Education and Cultural Affairs of the Länder in the Federal Republic of Germany. These studies are also often called large scale assessment studies. In particular, studies like the IQB National Assessment Studies / IQB Trends in Student Achievement Studies as well as PISA, IGLU and TIMSS (for an overview, see here).

These studies are based on samples that are representative for Germany and, depending on the study, also for the federal states and are therefore well suited for analyses at the level of the education system. An up-to-date list of the school achievement studies available at the FDZ at IQB with data representative at the federal state level is available at.

back to overview

What do I have to consider when I want to apply for studies on educational monitoring?

The above-mentioned studies on educational monitoring in Germany are subject to special terms of use, which are formulated in the FDZ at the IQB PDF Rules of Procedure or the PDF Procedure for Novel Interstate Comparisons and which particularly concern access to information on the federal states identifiers as well as measures of cognitive abilities.

back to overview

What are the IQB National Assessment Studies / IQB Trends in Student Achievement Studies about?

Information on objectives, study design, procedure and results of the IQB studies can be found here.

For methodological questions on the data, you will find study-specific information on the study-specific pages and via the menu item.

back to overview

What are the methodological challenges in analysing data from studies on educational monitoring / large scale assessment studies?

Special features of the sample: Data sets from Large-Scale Assessment studies do not represent simple random samples from the population. Instead, they were drawn in different strata (stratification). As a result, the (unweighted) distribution of students among the different federal states, school types, etc. (depending on the definition of the drawing strata) do not correspond to the corresponding shares in the population. The sampling units for the random drawing are schools and, within the schools drawn, usually complete classes. Since students within a school or class are more similar than students from different schools or classes, the sample is more homogeneous than a simple random sample of the same size (clumping or design effect).

Special features of the variables: Missing data are usually imputed by a complex procedure.The plausible values technique is usually used to measure competences (Lüdtke, O. & Robitzsch, A. An introduction to the plausible values technique for psychological research. Diagnostics, 63(3), 193-205).

The R-Package eatRep allows the determination of mean values, frequency distributions, percentiles and regressions, considering the nested and imputed sample. Trend analyses can also be calculated. The above-mentioned special features of data from large scale assessment studies are taken into account in the R-Package eatRep as follows:

  1. inclusion of individual person weights
  2. replication methods (bootstrap, jackknife or balanced repeated replicate methods)
  3. pooling of results according to the rules of Rubin (1987)

In the tutorial PDF replication analyses with eatRep (DOI: 10.5159/IQB_Tutorial_Replikationsanalyse_v2) (German version only) some typical analyses with the R-Package eatRep are demonstrated by way of example. You find an english introduction to eatRep here.

back to overview

Are the competence estimators of the PISA, IGLU and IQB studies comparable with each other?      

In principle, the tests from PISA and the IQB studies correlate highly, but the underlying competency models differ. The IQB tests are based on the educational standards of the The Standing Conference of the Ministers of Education and Cultural Affairs of the Länder in the Federal Republic of Germany and thus more closely aligned with the schools´ curriculum than the PISA tests. Comparability can be tested using IRT methods based on studies in which both PISA and IQB items were used. Some studies for comparison are, for example

Please note additionally:

1.) In contrast to the PISA surveys, reading and mathematical literacy are only tested together in the IQB study in primary school: Reading literacy was recorded in the IQB National Assessment Study 2009 (lower secondary level) and in the IQB National Assessment Study 2011 (primary school) as well as in the IQB Trends in Student Achievement 2015 (lower secondary level) and in the IQB Trends in Student Achievement 2016 (primary school). Mathematics competencies can be found in the IQB National Assessment Study 2012 (secondary level) and IQB National Assessment Study 2011 (primary level) as well as in the IQB Trends in Student Achievement 2016 (primary level) and the IQB Trends in Student Achievement 2018 (secondary level).

2.) We do not have a federal state identifier for the PISA surveys.

3.) If you wish to conduct analyses that include unpublished, novel comparisons between single federal states, our Rules of Procedure state that an extended application procedure with a review process applies.

back to overview

Citation &  research results

How do I cite data that I have received from the FDZ at IQB?

The FDZ at IQB assignsa digital object identifier (doi) for each study - with the help of the registration agency da|ra. Data users must agree to cite both the data depositor(s) and the FDZ at  IQB in an appropriate form by stating the doi of the data sets analysed (in whole or in part) as well as related materials.

Order of authors (year). Title (version [no.]) [Data set]. Berlin: IQB - Institute for Educational Quality Improvement. doi

A citation proposal for each package of studydata package incl. doi can be found in its most recent version on the study page within the FDZ website at IQB.

back to overview

How do I cite the FDZ at  IQB in my publications?

Please make sure that you also cite the FDZ at IQB in every publication (see § 2, paragraph (5) in your data use agreement: "For every publication based in whole or in part on the materials provided, the FDZ am IQB must be cited"). For example, you could include the following sentence in the Acknowledgements or in the Author Note:

“The data was provided by the Research Data Centre at the Institute for Educational Quality Improvement (FDZ at IQB).”

back to overview

What to do about specimen copies?

An electronic copy (PDF) or a paper copy of any publication resulting from the use of the data and materials provided must be made available to the FDZ at IQB without delay free of charge. This also includes so-called "grey literature" as well as student papers that are only published within the university (e.g. bachelor, master or diploma theses). Such unpublished research results are, of course, treated confidentially by the FDZ at IQB and only serve to ensure that no legal or other regulations, in particular usage guidelines, have been violated.

back to overview

Methodical notes

What is meant by cognitive abilities?

Cognitive abilities are interdisciplinary abilities or ability dimensions that are relevant to school learning processes, such as crystallised/verbal knowledge or logical reasoning and solving figure analogies.

back to overview

Are cognitive abilities the same as school competencies?

This is a difficult question on which researchers take different positions (Köller & Baumert, 2012). School competencies are assessed in school achievement studies using standardised achievement tests, which can relate to curricula, learning goals - such as educational standards - or to the transfer of school learning content to requirements close to everyday life. In many school achievement studies, cognitive abilities are assessed using scales from established intelligence test procedures (Berliner Test zur Erfassung fluider und kristalliner Intelligenz für die 8. bis 10. Jahrgangsstufe, BEFKI; Kognitiver Fähigkeitstest, KFT; Grundintelligenztest, CFT). They are mainly used in cross-sectional studies to control for individual differences in area-specific prior knowledge (Baumert et al., 2006). Since usually only short subscales from these test procedures are used, these scales are not suitable for forming an intelligence quotient. In studies on educational monitoring (and occasionally also in other studies, see the section "Studies on educational monitoring"), according to the FDZ at  IQB guidelines, measures of cognitive abilities must not be considered as dependent variables in analyses. These do not include subject-specific performance, school competencies or school grades.

Further reading:

Köller, O. & Baumert, J. (2018). Schulische Leistungen und ihre Messung. In W. Schneider, & U. Lindenberger (Hrsg.), Entwicklungspsychologie (S. 663-680). Beltz.

back to overview

What are Plausible Values (PVs)?

Plausible Values (PVs) are estimators of school competencies (and occasionally attitudes and other individual characteristics) based on item-response models. PVs are suitable for the analysis of group differences and for drawing conclusions at population level (e.g. do girls and boys differ in reading comprehension?).

The drawing of PVs is based on the assumption that school competencies and skills are latent variables that cannot be directly observed.  From the observable test scores of individuals (e.g. number of correctly solved tasks in a competency test), conclusions are drawn about their underlying abilities, whereby measurement errors in the observed test scores are taken into account. PVs are multiple drawings from a distribution (e.g. 15 PVs per person), which provide a plausible estimate of students' abilities. The multiple drawing makes it possible to take uncertainties into account when estimating latent abilities. In addition, correlations between test scores and individual characteristics (e.g. gender, immigration background, socioeconomic status) can also be taken into account. This procedure enables a measurement error correction when estimating correlations between latent abilities and observed characteristics (Lüdtke & Robitzsch, 2017).

Analyses based on PVs have to be performed separately for each PV and the resulting individual results have to be aggregated to one value. For example, if a regression is performed to determine the relationship between reading comprehension and socioeconomic status, a separate regression must be performed for each PV and then the different regression weights must be aggregated into one value (this aggregation is offered in many statistical programs, such as R and Mplus).

Further reading:

Von Davier, M., Gonzalez, E., & Mislevy, R. (2009). What are plausible values and why are they useful? IERI monograph series, 2(1), 9-36.

back to overview

How do I use sample weights?

Most school achievement studies for educational monitoring use a complex sampling procedure (e.g. multi-level sampling, stratification of the sample). For example, schools are drawn first (e.g. sorted by region and school size) and then students are selected. As a result, students have different drawing probabilities to be included in the sample. The use of weights makes it possible to take into account the complexities of sampling in analyses (e.g. estimation of average competences and associated standard errors). Weights indicate for each child in the sample how many students in the population that student represents. The use of sample weights is necessary to make statements at the population level.

Further reading:

Rust, K. (2014). Sampling, weighting, and variance estimation in international large-scale assessments. In L. Rutkowski, M. von Davier, & D. Rutkowski (Eds.) Handbook of international large-scale assessment: Background, technical issues, and methods of data analysis (pp. 117-153). Taylor & Francis. www.taylorfrancis.com/books/e/9780429111112/chapters/10.1201/b16061-11

Stapleton, L. M. (2014). Incorporating sampling weights into single- and multilevel analyses. In L. Rutkowski, M. von Davier, & D. Rutkowski (Eds.) Handbook of international large-scale assessment: Background, technical issues, and methods of data analysis (pp. 363-388). Taylor & Francis. www.taylorfrancis.com/books/e/9780429111112/chapters/10.1201/b16061-23

back to overview

What are missing values and how to deal with them?

Data sets collected in school achievement studies are usually incomplete and contain missing values for certain individuals. Missing values can result from various failure mechanisms. For example, many school achievement studies use test and questionnaire rotations, so that not every student completes all tasks of a test or all items of a questionnaire. Missing values can also result from people deliberately not answering a question, from the fact that a person's answer cannot be interpreted, or from the fact that a person could not reach and complete a certain task in the test due to time problems. Missing values can be systematically related to unobserved or observed background characteristics of individuals or can be completely random. There are different ways of dealing with missing values in analyses. In school achievement studies, procedures of (multiple) imputation are often used,Multiple imputation procedures are used to generate several complete data sets which results can be aggregated.

back to overview

How do I deal with missing data in hierarchical data sets?

You can use the Full Information Maximum Likelihood (FIML) approach to deal with missing values in multi-level models.

You can find further methodological information on dealing with missing values in multi-level models in the following publications:

Grund, S., Lüdtke, O., & Robitzsch, A. (2018). Multiple imputation of missing data for multilevel models: Simulations and recommendations. Organizational Research Methods. doi: 10.1177/1094428117703686

Lüdtke, O., Robitzsch, A., & Grund, p. (2017). Multiple imputation of missing data in multilevel designs: A comparison of different strategies. Psychological Methods, 22, 141-165. doi: 10,1037-metre-0-00096.

back to overview

print preview
Kontakt

Dr. Aleksander Kocaj
Leitung Forschungsdatenzentrum (FDZ)
(030) 2093.46507

a.kocaj@
iqb.hu-berlin.de

Postal Address

Humboldt-Universität zu Berlin
Institute for Educational Quality Improvement (IQB)
Unter den Linden 6
D-10099 Berlin

Offices

Luisenstr. 56
10117 Berlin
Tel: +49 (30) 2093-46500
Fax: +49 (30) 2093-46599
Email: iqboffice@iqb.hu-berlin.de

Secondary Offices

Hannoversche Straße 19
10115 Berlin
Tel: +49 (30) 2093-46552


Schönhauser Allee 10
10119 Berlin