Xtraktr allows anyone to easily upload data files (think SPSS, STATA, and CSV) and make the data explorable using visualizations that are shareable and embeddable.

Too many organizations miss opportunities to extend the life and value of their raw data by shelving it after they have added the juicy bits to their reports. Why not extend the life of your dataset beyond your project and get the information in hands that can use and reuse it to make a difference? With Xtraktr, this is easy. Xtraktr exposes your datasets so people can find them, explore them, use them, and make a difference long after your project has ended.

Have you conducted a survey for your project? Does your project have a website? Then why not use Xtraktr to enable your stakeholders to explore the raw data right there instead of having to download the data and learn SPSS, STATA, R, or other complicated statistical computing environments. With Xtraktr, your audience can explore your data easily.

Open your data

Don’t wait for people to ask for your data, put it out there. Make it searchable. Let Xtraktr be the face of your datasets. Enable your audience to explore your data’s variables and see the relationship between two variables in a variety of charts and thematic (choropleth) maps. Would you like to add those visualizations to your blog or another web page? Then use Xtraktr’s embed feature. Does your audience need more advanced features? Xtraktr exposes pulic data via an API and also enables your audience to easily download the raw data in a variety of formats and explore it further.

Survey data

Xtraktr excels at making those tricky SPSS and STATA files easy to work with. You no longer have to format for a spreadsheet first. Just choose the file and upload it. Xtraktr’s sensible defaults make a normal upload easy.

Not happy with what your statistical software gives you? We know! That’s why Xtraktr allows you to edit the questions and answers to move beyond the text limitations that SPSS and other software imposes and makes it easy to give audiences meaningful text.

Want to configure what the public sees a bit more, go ahead. You can pick and choose what the public can explore online and what the public can download. Xtraktr was designed first and foremost for statistical data, and it handles those well.

Do you have weighted data? Select the variable that contains your weights and you are done. This works for individual datasets as well as time series (panel or longitudinal) data.

Localize your data

Putting your data online makes it more accessible, but often audiences are limited by the language the dataset is in. Xtraktr makes it easy to localize your dataset (questions and answers) into as many languages as you want.

Translate directly through the application itself or via a spreadsheet for bulk translations. Make your data truly accessible!

Are you conducting a survey in multiple languages, Xtraktr makes it easy to collaborate on questions and answers and then translate them so they are all in sync. How often have you asked for questions and answers for two languages and received instances where there are more questions for one than the other? Or the answers for one question are different than in another? Or the order for each version is messed up? No more!

Catalog your data

Are you having a difficult time keeping track of your or your organization’s datasets? Need a way to catalog them so you don’t lose them? Use Xtraktr. Make your datasets private or public as you need them to be.

Xtraktr allows you to create an organization account and enable administrators to manage your datasets. Want to maintain a private list of datasets? No problem. Want some of them open and accessible? Also, no problem. Use Xtraktr to not just manage your organization’s datasets, but as a part of your organization’s open data policy as well!

Too often we hear that older datasets are lost due to personnel turnover or when a laptop was reformatted without being backed up. Your organization is often the accrual of the data and information it has gathered and created over the years. Why risk losing an asset as important as a major survey? Use Xtraktr to catalog and maintain your data for years to come.

Central data repository

When editing survey data, do you often stumble over different versions, not knowing if the members of your team are each making edits in their own version? Use Xtraktr to coordinate team efforts when editing questions and answers and for translations and know you are working with the up-to-date and definitive version.

Open Source and Free

Xtraktr is built using open source software and has, itself, an open data license. You can use JumpStart's hosted version, or setup your own instance of Xtraktr for your or your organization’s needs. Xtraktr is built using Ruby on Rails, MongoDB, and R (for importing data). Check out the code repository on GitHub.

  • process data
    • can process statistical data files (SPSS and STATA)
    • can process spreadsheets (CSV and XLS)
  • notify you of warnings
    • while processing your data, Xtraktr looks for questions with no text, question with no pre-defined answers and answers that are in the data but not in the pre-defined list of answers
    • you can then view these warnings and fix the issues as needed
  • download data in many formats
    • if you uploaded the dataset, you can always download the original dataset file
    • can download data in SPSS, STATA, R or CSV format
    • all downloads include a codebook, a list of all questions and the possible answers
    • instructions, when relevant, are included to help load the data into statistics software
    • the download files are updated as you make changes to the text and configuration settings in Xtraktr
  • enable multiple languages for your data
    • can select which languages you want your data to be available in (no limit)
    • fields that can be translated are questions, answers, methodology, reports, etc
    • can download spreadsheet of all questions and answers that need to be translated, add translations to the spreadsheet, and then reload spreadsheet into Xtraktr to apply translations en mass
  • handle metadata
    • dates data collected
    • date data released to public
    • source
    • description
    • methodology
    • categories
    • country(s) involved
    • donor
  • host related reports
    • upload reports that have been written about or that use your data
  • highlight
    • allows you to draw attention to specific results of your data
    • highlight shows on the dataset dashboard page, highlights page and possibly on Xtraktr home page (4 random highlights are chosen)
    • can provide description of highlight to help put it into context
  • edit questions/answers
    • enter full text of question and answers which may have been abbreviated or cut off in the data file due to statistical program limitations
    • add special notes for a question that will appear in the codebook and on the details tab of the explore page
    • indicate which questions should be excluded from the the public (all questions are always available for analysis under the admin section for those who have access)
    • indicate which questions should be included in the data download files for the public (admin data downloads always include all data)
    • change the order the answers will appear in the codebook and during analysis
    • indicate which answers should be excluded from  the public (all answers are always available for analysis under the admin section for those who have access)
    • indicate which answers can be excluded from the analysis by the user (e.g., refuse to answer, don't know, etc.). This means that the answers will be included by default but the user can exclude them from the analysis if they want to.
  • manage the data type of each question
    • a question can be of a categorical or numerical data type
    • categorical questions are analyzed using bar and pie charts
    • numerical questions are analyzed using histograms
    • when your data file is loaded, Xtraktr attempts to set the appropriate data type for each question; if a question's data type cannot be determined, one is not assigned
    • you can change the data types at any time
    • when setting a numerical data type, you can set the number type (integer or decimal), indicate how wide the histogram bar should be, and what the min and max value of the histogram should be
  • accept mass changes
    • instead of editing each question one at a time, you can edit them all at once
    • for questions, can change question text, indicate which questions to exclude and indicate which questions to include in the data download
    • for answers, can change answer text, indicate which answers to exclude and indicate which answers can be excluded
  • enable question groups
    • you can group your questions however you desire (e.g., group questions to match the survey)
    • each group has a name and description
    • a group can also have a subgroup
    • groups appear in the codebook and during analysis
  • enable group/question sorting
    • you can customize the order in which your groups/questions appear
  • enable data weights
    • you can apply one or more weights to your data
    • you can indicate to which questions the weights apply
    • during analysis, a dataset that is weighted uses the weight by default, but the user has the option to turn the weight off to see unweighted data
    • during analysis, if you are comparing two questions and they do not have a common weight, then no weight will be applied to the analysis
  • map your data
    • if you have questions that are geographic in nature, you can have the analysis of these questions appear on a map
    • Xtraktr has a list of predefined map shapes. If the shapes you need are not available, contact us and we will add the shapes. Xtraktr uses the geojson standard for map shapes.
  • provide a dashboard of your data
    • dashboard shows all of the information about your dataset:
      • dataset title
      • list of dates for when the data was collected and when it was released to the public
      • source
      • description
      • country(s) involved
      • donor(s)
      • highlights
      • methodology
      • codebook
      • reports
      • license
      • metadata
      • share links
      • explore data link
      • download data links
  • explore your data
    • the explore page allows users to analyze each of the questions in the dataset
    • the analysis results are shown in charts, maps and tables
    • users can compare the results of two questions
    • users can filter the analysis by the answers to a question
    • the codebook for the selected questions will be available
    • users can share their findings, download a chart, or embed a chart on another website
  • handle time-series (panel) data
    • if you have longitudinal datasets, you can merge these datasets into a time series
    • can either pick and choose each question from each dataset by hand or allow Xtraktr to look for matches by question code and automatically add the questions
    • each dataset can be labeled how you want (i.e., 2015, 4th Quarter, etc) and these labels are what appears in the chart during analysis
    • time series can also have groups, sorting and weights
  • processing data
    • cannot load SAS files due to the proprietary file format, however if you want to donate money to us so we can purchase a SAS license, then we will be able to process SAS files
    • cannot yet reload a dataset file due the complexity of determining what should happen to the existing questions, answers and data on file, especially if you have already use one of the many settings in Xtraktr
  • search
    • cannot search for questions across all datasets
  • date/time data
    • Xtraktr currently has the ability to combine datasets into a time series (panel or longitudinal) study, however it does not currently have the ability to properly analyze one dataset that has date/time data in it (i.e., the daily exchange rates for currencies)
  • help section
    • a help section with text, images and screencasts is in the works