Reference datasets: everything you need to know

The Reference dataset gives you the flexibility to track your reference data in one of three ways:

  • You can connect to your LibAnswers system to automatically keep track of the number of tickets, SMS messages, and LibChat sessions with your patrons, as well as the number of FAQs views in your system.
  • You can keep track of the aggregate number of tickets/questions, SMS messages, chats, and/or FAQs views from another reference system.
  • You can keep track of individual reference questions, with data such as the question asked, the answer provided, who answered the question, and more. (This works great for analyzing Reference Analytics data exported form LibAnswers!)

No matter how you choose to track your data, this dataset provides reports showing the overall volume, distribution, and trends in your reference transactions. This makes it easy to analyze your reference services over time.

Example of a reference Overview report

In this Springboard, you'll learn how to create and configure a Reference dataset, as well as record, upload, and analyze data.

Create a Reference dataset

Reference datasets are only available to LibInsight Full customers. If you're interested in upgrading to the Full version of LibInsight, contact our Springy Sales team.

To create a new dataset, Admin users can go to Admin > Manage Datasets and click the Add New Dataset button.

Navigating the the Manage Datasets page and adding a new dataset

Before you begin, you may find it helpful to do a little planning. This can make the process of creating your dataset go more quickly.

  • How are you going to use this dataset? A Reference dataset can be configured to:
    • Upload aggregate totals via a spreadsheet. Use this option if you only want to gather the aggregate numbers of tickets/questions, SMS messages, chats, and/or FAQs views, but do not have a LibAnswers system.
    • Automatically import aggregate activity data from your LibAnswers system. This will run a daily import of the total number of tickets, SMS messages, LibChat sessions, and FAQs views from LibAnswers. This does not import individual tickets or chat transcripts.
    • Record individual reference transactions (either manually or in bulk via upload). Use this option if you want to record data about each individual transaction, such as the date, time, question asked, answer, etc. If you choose this option, consider:
      • What type of timestamp works best for your data? When you record a new entry, it will include a timestamp. For datasets where you want to analyze durations (such as instruction sessions), select the Start & End Date/Time option. Otherwise, you can choose to record a single date & time; just the date; or the month & year (not the day). You can also choose to automatically calculate a single date & time stamp, which will happen at the moment a user submits the new record.
        • Important: once you start collecting data, you cannot change the timestamp type. This can only be changed while the dataset is empty, so it's important to think carefully about what option will work best ahead of time.
      • Do you want to use the READ scale to rate each transaction? This can help assess the overall effort required to answer the patron's question.
      • What data do you want to collect? It can help to sketch out a list of all of the data you want to collect and analyze. Once you have a list, consider what type of field is best suited for that data. See the Available field types for individual transactions panel below to learn more about each type.
        • In addition to the required timestamp and optional READ scale, all submissions will include a required Entered By field (who "owns" the record), as well as an optional Internal Notes field where you can leave a comment about the data.
        • You can also choose to enable or disable recording the submitter's IP address and submission source (where the record was entered, such as the Record Data page or a widget).
        • When adding your fields, you can arrange them in up to three columns.
  • Who should have access to the dataset? By default, only Admin users are allowed to manage, record data to, and analyze a dataset. However, you can choose to extended each of those permissions to selected Regular users, or all Regular users in your system. For example, if you only want a few people to add data to your dataset, you would extend them Record permissions. But, if you want everyone in your library to be able to view and analyze (but not edit) the data, you could choose to give Analyze permissions to all Regular users.

Learn more

[Return to top]

Record data to a Reference dataset

There are three primary ways you can add data to your reference dataset:

  • Aggregate data: if you configured your dataset to record aggregate data, you can manually enter daily or monthly totals, or upload the data from a CSV or Excel file.
    • LibInsight will provide you with the list of field names your file must include in order to be uploaded successfully.
  • Individual transactions: if you configured your dataset to record individual reference transactions, you can either enter each transaction manually, or upload your transactions from a CSV or Excel file.
    • When manually adding records, you can create and use pre-defined entries. Think of these as templates that can make recording commonly-added records quicker and easier: when a user selects a pre-defined entry, it will automatically populate the Record Data form's fields using the entry's preset values. Users can then make any changes they need before submitting the new record.
    • If uploading data, the column headings in your spreadsheet must match the field names in your dataset. All required fields (such as the Start Date) must be present, but optional fields (such as Internal Notes) may be omitted.
      • LibInsight will provide you with the list of field names your file must include in order to be uploaded successfully.
      • When uploading a file, make sure that you select the date format that matches the format used in your file's Start Date column. You can set the default date format for file uploads by customizing the System Date Format in your system settings.
    • You can also create widgets so you can record data without having to log into LibInsight.
  • Import from LibAnswers: if you connected your dataset to your LibAnswers system, LibInsight will automatically import the previous day's data each night.
    • For example, the data from January 1 will be available on January 2, and so on.
    • This follows the initial import of your data when setting up your dataset.

By default, users who have permission to add data can also edit or delete their own records. Users with Admin permissions, however, can edit or delete any user's records in the dataset.

Add and manage data

Create widgets and APIs

[Return to top]

Analyze a Reference dataset

Imported LibAnswers or aggregate data

When analyzing a Reference dataset that is importing aggregate data from LibAnswers or via spreadsheet, you'll find the following reports:

  • Overview: provides you with a breakdown of your total reference transactions over time. This includes a chart that can display transactions by day, month, or year, and a data table with the raw totals for each month.
  • Distribution: allows you to individually analyze your SMS, Tickets, Chats, or FAQ Views. For each, you can see the distribution of your data by day of the week, weekday vs. weekend, and by month.
  • Trends: allows to see how the totals of your SMS, Tickets, Chats, and FAQ Views have changed over time. You can create charts to visualize trends in your data for the past 2, 3, 5, or 10 years. In addition, you'll get a data table with each year's totals, with the ability to see year-over-year changes and the percentage change against the first year of the report.

Individual reference transactions

When analyzing a Reference dataset that is recording individual transactions, you'll find the following reports:

  • Data grid: provides a list of all records included in your report. From this view, records can also be edited and deleted. Admins have the ability to edit or delete any record. All other users can only edit or delete the records that they've entered.
    • You can filter the data included in your report by one of several date & time filters, as well as by field value(s).
    • Use the Export Data tab to download a copy of this data in CSV format.
  • Field Analysis: provides you with an in-depth look at the responses for individual fields.
  • Time-Based Analysis: allows you to track your data over time. This can include the total number of entries added to your dataset, as well as sum of Numeric fields and distribution of Select fields. You can analyze data by year, month, day, or hour.
  • Duration: If your dataset includes both a Start Timestamp and End Timestamp, this report will allow you to analyze the duration of each record.
  • Distributions: provides a breakdown of your data over time (i.e. year, month, day of the week, hour of the day) and by the Entered By user.
  • Cross Tab: allows you to compare the relationship between two fields.
  • Trends: allows to see how the totals of your numeric data has changed over time (i.e. the past 2 years, 3 years, 5  years, or even 10 years). In addition, you'll see year-over-year changes and the percentage change against the first year of the report.

In addition, LibInsight includes other tools to help you analyze, visualize, and share your data with others:

  • Dashboards: dashboards allow you to display charts of key data points from one or more datasets on a single page. Dashboard pages can be public or private, allowing you to create and share dashboards internally with staff or with the public.
  • Cross-Dataset Analysis: allows you to compare up to 4 variables from across one or more datasets. This can provide some really useful insight for seeing trends across services or resources over time.

Learn more

[Return to top]

Related Springboards