Skip to content

Before You Analyze…

Have you been in meetings where you were presented reports that later you wondered how they related to the task at hand? The purpose of a review is as important as interpretation of the data and often, the review fulfills the purpose for which the data was gathered. Stakeholders must understand the purpose of the data analysis before they can know how to gain useful insights or add value. Only then is there practical and actionable use of the information to be achieved.

Before you dive into interpreting, clarify what element(s) of potentially strategic actions might be being addressed. The responsibility of the preparer of the data is to assure reporting is both relevant and comprehensible enough for stakeholders to understand.  From there ask what part of the review requires action and what that action might be.  When in doubt ask!

Below we’ve outlined some key areas that users of data should clarify before beginning an analysis. They include Purpose, Context, Meta Data and Stakeholder Roles.

Purpose/Intent – There are a wide variety of reasons to analyze data, but the reviewer has to understand the stated purpose for which it was gathered.  Several reasons for data can include:

  • Background – why the data has been gathered and reported including historical perspectives that inform current situations
  • Status updates – what is the status according to target for projects, budgets, task assignments, growth or reductions in activities
  • Engagement – determining the effectiveness of fund raising campaigns, collaborative or outreach work
  • Advocacy or policies – are positions clarified and if so how?
  • Research and facts – creation or support for the hypothesis or ongoing analysis
  • Client level – the “who” of clients or beneficiaries demographic profiles, service needs
  • Operational – measurement of efficiencies, processes, infrastructure needs, financial health, employee productivity
  • Project level – iterative design changes, scaling or compliance
  • Predictions – indicators of success that support targets such as Key Performance Indicators, insights or trends

Context – Another important consideration is around external circumstances that may affect your overall understanding:

  • Universe – defined sets of the data bounded by attributes such as subpopulations, gender, ethnicity, geographic regions or service domains
  • Comparatives – activity periods or comparisons to other geographies or universes
  • Parameters – Filters that further inform the understanding of the data including date ranges, client groups, stand-alone project or grouped projects, inclusion of null or blank values
  • Influences – Collection methodologies, barriers to collection, governance models, policy directives
  • Transparency – Data source, quality, de-duplication/de-identification processes, importing or other record manipulation, quality, timeliness, consistency of presentation, shared domains and credibility of the the presenter

Meta Data – Meta data is defined as “data about data” and has been described as the “genetic makeup of the data”. Examples of client-level meta data can include the user who first created the client record, and the who, what and why for later changes made to the records. This provides an audit trail that protects client privacy and establishes the general reliability of the data set.

Another example of meta data is it’s use to catalog large data sets for more rapid retrieval. Elements for record cataloging that could aid in retrieval can include the author, date created, date modified, file size and type such as video, photo or document.

Users of data should feel comfortable asking for the basic meta data elements that establish the age of the data set and potential modifications that could have impacted the original integrity of the records and ultimately the summarized information.

Stakeholder Role – Perhaps the most important consideration in reviewing data is to be clear about your role as it relates to the outcome of the information being presented. Many times, data is presented as FYI only, but as a reader you should be clear about not only the purpose, intent and context but also how your role as a reader determines the value you gain.

Different levels of stakeholders benefit from a variety of presentation styles whether detailed or summarized, tabular or visual. A nonprofit board would see less value being shown a chart of client gender when they are grappling with shortening the length of stay in emergency programs unless a relationship between gender and days in shelter can be established.

When stakeholders receive data in a format best suited to their role, they can be confident their analysis will be of value to recommend solutions for our complex social problems.

Teddie Pierce has worked in custom database and stakeholder report design since 1996.  Most recently she served as the Sonoma County, CA Homeless Management Information Systems Administrator where she implemented data collection and custom homeless project designs for Coordinated Entry, HOST Street Outreach, Safe Parking and other homeless programs.  She currently serves on the National Human Services Data Consortium Board of Directors.