Step 5: Knowledge Management

Before documenting Indigenous Knowledge or collecting scientific data, it’s important to assess your anticipated needs for data management, analysis, and sharing and make informed choices about which data management system(s) to use. There are many data management systems available. Deciding which system is right for your needs will depend several factors including but not limited to:

• Ease of use
• Where the data is hosted and data ownership
• Flexibility
• Features
• Security
• The format of your data
• Cost

If you have mapping data, you will need hardware (e.g., server) and software (e.g., QGIS, ArcGIS, etc.) capable of managing spatial data. You may need unique systems for storage and managing various types of data depending on the type of data, security, and whether you want the data to be public, private to a select group, or completely private.

An important consideration is whether the system will provide security for storing and managing sensitive information such as Indigenous Knowledge. Some knowledge is a sacred gift that is shared between a Knowledge holder and an individual who is in a position of trust. There’s a need to determine protocols to ensure this knowledge is managed in a manner trusted by the Knowledge holders. Ensure your data management system will provide adequate security for storing and managing sensitive information such as Indigenous Knowledge. Another important consideration is whether the system will provide you with adequate ownership and control over your data.

There are many different types of data management applications including:

• Web-based data management platforms: There are several on-line platforms tailored to support management of monitoring data. These systems often offer the user choices about who can manage and view the data.
• Desktop data applications: Some data management systems are designed to run off your computer or local network and are not shared with other users. For example, Microsoft Excel, statistical computing software such as R-Project, and Geographical Information Systems (GIS) are often desktop-based but may have an on-line component.
• Protocol specific on-line portals: Many citizen science programs have web-based platforms where you can upload data related to specific indicators or more general observations using on-line forms e.g., iNaturalist and eBird.
• Metadata catalogues and data repositories: Metadata catalogues are collections of information about data. Some metadata catalogues can also store data. For example, the Polar Data Catalogue allows you to attach files to your metadata records. Similarly, data repositories are searchable, on-line systems designed specifically to hold data, like a library holding books e.g., DataStream. Often these repositories are “open” meaning anyone can find and view your metadata or entire datasets hosted there.

Some monitoring projects choose to build custom applications specific to their needs using either web-based or desktop applications. Refer to the resources section below for tools to identify some of the commonly used community-based monitoring data management systems to help you select which one(s) are right for you.

Create clear systems for organizing the data you have collected. Before going into the field prepare and test your field data collection process, which can include creating field data sheets, either on paper or using existing digital forms or creating your own digital forms. Some data management platforms have app-based tools or forms to support field collection that can be synched with the data management systems (refer to this list of apps for examples). Using digital apps or on-line forms for data entry can mean less time transcribing data and fewer opportunities for errors. You should always record who monitored, where, and when on your data sheets along with other basic information.

Make time to ensure your field data is transcribed (i.e., digitized) or uploaded into your data management system once you’re back from the field. Be sure to include the units (e.g., cm or °C) for your measurements, and explain any codes, acronyms, or missing values.

Watch this video produced for Ebb and Flow First Nation for tips on recording, storing, and managing your climate data.

Assuring data quality refers to avoiding errors and finding and correcting errors in your data (quality control) and having a system of external review of your data (quality assurance). Data assurance should be built into all phases of your project – preparing to collect data, field data collection, entering data into a database, and combining and assessing data.

For automated instrumentation, data quality should be assessed by scheduled calibrations (consult the manufacturer). Calibration of instrumentation may involve:

• Operating two instruments (project and lab-standard instrument) simultaneously in the same location for several days.
• Exposing the instrument to a known physical standard (i.e., gas, liquid, solid).

As mentioned in Step 4: Approach and Methods, it’s important to build in a process to share results back to community members and validate findings and ensure errors of fact, omission, and interpretation are avoided.

It’s essential that you document information about your monitoring events and store that information with your monitoring data. This “data about data” is called “metadata”. Your metadata should include information on what was monitored, where, by whom, why and how (following what protocol), as well as contact information related to the data or project. Remember that your metadata should always be linked to your data so users can clearly understand your data. This metadata is necessary so that users can understand the data and whether and how it’s appropriate to use. Sharing your metadata will help users discover information about your monitoring initiative and contact you to find out more.

An important consideration is to determine how the knowledge and data you have collected will be stored over the long-term. Develop a system to routinely back up your data which follows the 3-2-1 data protection strategy. This will lower the probability of accidental loss of data. Ensure your metadata is stored with your data. It’s helpful to have a designated trusted steward in charge of your data. The steward is often knowledgeable about the data and acts as a point of contact for questions about data access.

Managing your data is a long-term commitment. It’s important to invest resources into managing your data and to include data management as part of annual work plans. Training on data entry, data management, and analysis are also key to managing your data well.

Once data has been collected or Indigenous Knowledge has been documented, a key question becomes, “Has a change occurred?”. A secondary question is, “Why?”. In some cases, answering this question simply involves asking an Elder or Knowledge keeper. Because many Indigenous Peoples have been watching, listening, and learning about changes in their territory for decades (and generations), they can provide very clear evidence and explanation about the impacts of climate change.

In other cases, further analysis is needed. To understand changes observed, you may need to combine or integrate your data from various time periods or with data from other providers to help you better understand why changes you have observed have occurred. For example, you may wish to find historical weather data near your community or information on future projections because of climate changes. Seeking existing data is called “discovering” data. There are tools [LINK: Resources – Discovering data] to help you identify relevant climate data. Bringing together data from various sources often involves organizing data in a consistent way. This may involve integrating data into a common spreadsheet or database, ensuring the data has consistent units, etc.

Once the data is organized, you’re ready for data analysis. It’s important to “sit with the data” and gain a deep relationship and understanding of the novel dataset being analyzed. By sitting with the data, your project team will inform future data management, data collection campaigns, and increase the ability of the team to defend raw data and knowledge created data analysis efforts. The first step in data analysis is often exploratory. Describing, illustrating, or plotting your data as either a story, drawing or creation of summary statistics (e.g., mean, median, quantiles, standard error, minimums, and maximums) allows you to understand the big picture of the changes being documented. You can summarize data you have in a bar graph or histogram to show trends or unique events. Creating a graphical image tells a visual story to explain what you have learned from the data you have collected. You may need to create maps or compare your data against a reference point.

There are many kinds of variations in the environment that are common. Sometimes small changes can be very important. In other cases, larger shifts are normal or naturally occurring (e.g., cycles in barren ground caribou populations). Finding out what kinds of changes are outside the range of natural variability is important for determining what changes are climate driven. The ranges can be represented in a “control chart”. Your study design should help you to understand if you have enough data to conclude there’s a significant change or if the trend has crossed a threshold.

In other cases, the insights can be very immediate and urgent and must be communicated back to community members. For example, monitoring and communicating about ice conditions on a daily basis through the winter is important for travel safety. Use the precautionary principle to ensure monitoring outcomes are communicated right away.

Data analysis can be complex and expert support may be required. If you need support, you may wish to talk to your research partners or seek training in data analysis.