Frequently Asked Questions (FAQ)

Scoop.it ShareThis

In this section, you will find frequently asked questions (FAQs) about:

Beta testing

What do you mean by beta testing?

A beta test is a large scale test during which you provide your content to real potential users and ask them for feedback. CLOSER Discovery entered its beta testing phase in December 2015, and will be seeking feedback from users until the end of 2016.

The beta version seems to work pretty well. Can I start using it for my research?

The beta version is limited in terms of both content and functionality. The content that is included is usable – you can search for questions and variables on a topic and add those you find to a List. However, you cannot rely only on this search as there may well be other variables available that are not yet in the platform. It is also important to note that during the beta, links to items and Lists will be wiped on a regular basis. This will not happen once the platform is live. You should download any lists you create for use.

What features and functions will be added in the future?

  • Biosample documentation
  • Documentation of commonly used scales e.g. Strengths and Difficulties Questionnaire
  • Variable correspondence (is a similar variable available in the same or different study)
  • The ability to look at the way a selected group of similar variables have changed over time
  • Documentation of derived variables (but not the methodology of derivation)

How can I give feedback?

We welcome your feedback! You can provide this in a number of ways:

Feel free to use any of these, and please do give feedback as soon as you find something you want to say. You can provide feedback as many times as you like.

I found a spelling mistake or error in a variable label. Shall I report that as well?

If you find misspellings/errors in the variable labels, or are confused by them, please contact the study directly for details.
CLOSER Discovery documents these, but they remain firmly under the control of the studies. The CLOSER team does not have the expertise to give you a robust answer on variable lables. If you do send this feedback to us, we will direct you to the study contact best able to help you.

What if I find something that isn’t working?

If you find an error, or notice something doesn’t seem to be working quite right, please email us at discovery@closer.ac.uk. There are all sorts of reasons why an error might not have been caught before by someone else. For instance, it could only appear with specific browsers or operating systems, or it could only appear after a certain sequence of actions. For that reason, please try to give us as much detail as possible. Try to include:

  • A description or screen shot of the bug
  • What you were doing before you found the bug
  • What operating system and browser you are using (please include versions, for example Internet Explorer 9)

What happens to my feedback?

We will read and listen to every piece of feedback we are sent. All feedback is logged and then assessed against the following before being given a priority rating:

  • Is this something within our capacity?
  • How many people have pointed out the same issue?
  • How long will this change take to implement?

Certain things might take some time to implement. Even items we rate as the highest priority might take time to change. Send us all your ideas, even if you think they are totally unachievable! There are also lots of clever features behind the portal that you will not see as a user which might mean something that seems very difficult is actually easy to change. If we are unable to act on your idea within our current funding, they could be added to future projects or bids. We will keep a record of every suggestion we receive. We won’t be able to personally answer every piece of feedback. If you want to know what changes are being made and see if your ideas have made it in, please sign up to the CLOSER Discovery mailing list. This will alert you to new content and functionality as it is published.

What is included in CLOSER Discovery and what’s not

Which studies are included in CLOSER Discovery?

The eight studies included in CLOSER Discovery are listed below, but there is not complete information on each study in the beta test version. We will be adding the remaining content throughout the beta testing period. For an up-to-date list of what data collection instruments from each study are currently covered by CLOSER Discovery, please visit our Technical Wiki.

Why doesn’t CLOSER Discovery hold study data?

Data access arrangements are managed by the individual studies or the UK Data Service. There are two main reasons why CLOSER Discovery does not hold study data:

  1. Protecting study members’ privacy: Data protection and licencing remain the responsibility of the individual studies or the UK Data Service, ensuring study members’ identities are kept anonymous. See How to access the data for more information.
  2. Consent from study members: The availability of longitudinal data is only possible with the consent of the generous study members. Consent arrangements vary from study to study, and in some cases this prevents us from making data available via a tool such as CLOSER Discovery.

Does CLOSER discovery include information on administrative records that have been linked to the studies?

Where possible, this has been included. Exact arrangements vary depending on what records are being linked to.

What does CLOSER Discovery offer that isn’t already provided by the UK Data Service?

The data sets accessible via UK Data Service are large and important resources for research. There are three key features that CLOSER Discovery can help with when working with these, listed below.

  1. Descriptions on the UK Data Service are limited to 80 characters. By enabling searches on the full question text and filtering by topic, you are far more likely to find items that could be useful to you.
  2. The character limit on descriptions also restricts the context of the variable as it appears in the data set. Knowing that a question described “Have they moved?” in the UK Data Service data set is actually “Have they moved in the last 5 years?” could make a significant difference. CLOSER Discovery will give you that level of detail.
  3. Before CLOSER Discovery, the only way to determine if a variable provided enough statistical ‘power’ for your research was to carry out an analysis. CLOSER Discovery will allow you to preview summary statistics on the variables before you begin work.

Searching by topic

Please note that CLOSER Discovery is still in beta. Topics are a particularly challenging part of the process of documenting historical questionnaires. It is an area we would particularly like feedback on.

What topics does CLOSER Discovery use?

You can see the complete list of topics on the CLOSER Technical Wiki Controlled Vocabularies pages. CLOSER Discovery contains information from studies traditionally considered social and biomedical. Therefore the source of these is an amalgamation of HASSET and MESH.

How many topics are linked to each question/variable?

Each question/variable has only one topic assigned to it. The topics are also grouped, so each question/variable has the topic grouping applied.

Why can’t I find the topic I’m looking for?

The controlled vocabulary for topics is designed to provide a topic for every question and variable in the eight studies included. It is therefore a work in progress as long as content is being generated. Additionally, the topics will only appear if there is metadata currently linked to them. Beyond this, it may be that there is no content suitable for the topic you are looking for. In terms of browsing by topic, it is better to be inclusive rather than exclusive in the first instance. You can always refine your search as you go.

Why is an item tagged with one topic and not another?

Tagging historical metadata is an extremely challenging process. Almost every question and variable has had to be tagged manually.
We also had to ensure that there were no conflicts between the topics applied to linked variables and questions. This is being achieved at scale by applying a single topic to each item only. To do this, a bespoke piece of software has been built from the ground up.
There are also several ways of applying topics. A topic could be derived from the wording of the question, but this might differ from the intended research purpose of the question. This is further complicated by the fact that this is an historical project and the exact meaning of the question from either point of view may have changed over time.
We welcome feedback about the topic system and how it has been applied. It is one of the most challenging aspects of the project and understanding better how the users might interact with these is very important. Please visit our feedback pages for ways to contact us.

How to access study data

How can I access data from the studies?

The exact data access arrangements vary from study to study. In general they fall into two camps: accessible via the UK Data Service and accessible via direct contact with the studies. You can see further details on the page How to access the data.
Data available via the UK Data ServiceData available by contacting the studies directly
1958 National Child Development StudyAvon Longitudinal Study of Parents and Children
1970 British Cohort StudyHertfordshire Cohort Study
Millennium Cohort StudyMRC National Survey of Health and Development
Understanding SocietySouthampton Women’s Survey

Why can’t I download study data from CLOSER Discovery?

There are a number of complex reasons behind this that relate to legal, ethical and consent issues. None of the studies allow unrestricted access to the data. This is in order to protect study members. CLOSER Discovery was never envisaged as being able to manage these complex access processes and it is out of the scope of the work being undertaken. What CLOSER Discovery can do is help you better navigate those access arrangements, either by providing you with a list of variables you can request or by helping you identify what you need from large data sets with limited labelling.

Where can I find information on how variables were derived?

You will need to contact the studies individually for this information. This would require significant manual effort to make available electronically and doing so is outside of the current scope of CLOSER.

Are any of the variables weighted?

Only ALSPAC has any weighted variables. These are not currently marked and you will need to contact the study for further details.

Metadata

What is metadata?

Metadata is, simply put, data about data. It gives the data context. It describes, for example, when a given data collection wave took place and who conducted it. It describes the coverage of the data. It describes the exact questions asked and how they fit into an overall data collection instrument.

Why do researchers need metadata?

Using data without its metadata is like reading a long, complicated book with the index and glossary removed. You have most of the information you need, but not all of it. It’s harder to navigate. You have the terminology, but without the definitions.

Can I use the metadata available on CLOSER discovery?

CLOSER Discovery has its own Terms of Use. If you have further questions, please contact us at discovery@closer.ac.uk. Please also see further information on our Data security and intellectual property page.
Please be aware that CLOSER discovery is currently in beta and the content is incomplete. Use of the metadata should therefore be supplemented with other methods of identifying data.

What metadata standards are being used?

The metadata in CLOSER discovery are a sub-set of the DDI Lifecycle metadata standard. This is an internationally agreed standard used by data archives and other studies across the world. You can find further information about this on the DDI Alliance website.

Why use a metadata standard?

One of the challenges of data sharing is in the formats of data and metadata. A metadata standard overcomes these by providing a single set of definitions that everyone adheres to. The more people who adopt the standard, the more useful it becomes. Standards also promote an efficient ‘enter once, use multiple times’ approach to data management. If information is entered in an agreed way at the start of a process, it can then be used to feed other parts of the process automatically.In longitudinal studies, for example, if the DDI Lifecycle metadata standard is adopted by studies and survey agencies, this can be used to make the transfer of questionnaires and data between these partners much more efficient and robust.
For more information about metadata standards, please see the CLOSER Technical Wiki.