Skip to content

Latest commit

 

History

History
67 lines (45 loc) · 5.46 KB

align-data.md

File metadata and controls

67 lines (45 loc) · 5.46 KB
Raw

#Define Goals and Measure Progress

Align Data to Goals

Once the principal and the department leads have settled on the strategic framework and have set goals, it is time to align data with the city’s goals. Throughout the data uploading process when historical data is transferred to performance management software or a dashboard, the performance management team will have the ability to map the data to the goals the data supports, and the departments that are responsible for the activities that will advance those goals. The first data uploaded to the performance management dashboard will likely be what was gathered at the inception of this process. It should be data that already exists and is reasonably well organized.

Selecting First Datasets

Once a city has identified its goals and the corresponding metrics, the city must next prioritize its datasets for release.

A city will upload a broad variety of data that aligns to the outcome measures and goals. When possible, the first datasets should be organized by data type, typically in the following categories:

  • Operational data, which is typically collected across multiple organizations, such as financial or human resources data.
  • Subject matter data, which tells leadership about the performance of various programs, such as average length of stay in juvenile detention or third grade reading test scores.
  • Validating (external) data, which are common datasets that are universally understood as standard. Examples include the Uniform Crime Reports (UCR) published by the FBI, the Bureau of Labor Statistics Unemployment Rate, and the Big Three credit agencies’ bond ratings. The closer the data aligns to validated data, the more credible the goals become.

Operational Data for Measuring Outcomes

There will always be some standard operational data that will be collected consistently across a city government. This operational data will most likely come from a centralized personnel system or financial management system.

These are high value datasets because they are usually collected across departments and are typically standardized, which makes them ideal for pointing out cross-departmental trends. Because they will be used frequently and relied upon when conducting analyses, these datasets should be prioritized when beginning to embark on performance management work. The following are examples of data to include:

  • Dollars spent on overtime
  • Overtime hours paid
  • Hours not worked
  • Sick leave
  • Open and filled positions

Please see our [Data Set Starter Kit] (https://www.gitbook.com/book/centerforgov/open-data-getting-started/details) for additional data sets to consider.

Subject Matter Data for Measuring Outcomes

Cities that are successful in achieving results for their residents use subject matter data to measure outcomes. Subject matter data gives city leadership an indication of how multiple departments are working together to produce outcomes. Some subject matter metrics for public safety goals are found in:

  • Offender case management systems
  • Localized crime data
  • Court processing data
  • Warrants

Validating Data for Measuring Outcomes

Validating data is data that is recognized as the standard, or that is endorsed publicly as the key metric for specific measures. The following are common examples of validating data:

  • Unemployment Rate: In this changing economy, most governments are focused on employment and use the unemployment rate as a bellwether for economic recovery. This information is provided in a dataset from the Bureau of Labor Statistics, and is universally used as benchmark employment data.
  • Crime Rate: The FBI's Uniform Crime Reports (UCR) are used domestically as the standard for measuring crime from the national level to the local level.
  • Educational Test Scores: Scores from standardized tests are frequently used to measure average student achievement and readiness. These can be local or national tests and provide prevailing data for a particular jurisdiction’s education systems.

Louisville, KY

In 2012, Louisville, Kentucky, created an Office of Performance Improvement to engage residents in strategic planning, develop and implement an open data portal, and administer LouieStat. LouieStat requires each department to identify data sources and track metrics in order to reach goals determined by the department, city management, and residents. The outcomes have been impressive. By using LouieStat to inform decision making, Louisville seen the following results:

  • Removed more than two hundred days from key administrative processes
  • Reduced unscheduled overtime and worker’s compensation expenditures by more than $2 million
  • Aligned the budgeting process with city and departmental strategic objectives.

Mapping Data to Outcomes

Once the data is identified and uploaded, it is time to map the data to outcomes. This requires creating a consistent taxonomy across all measured units or departments of the organization. A taxonomy organizes the process of measuring results in the following way:

  1. Priority
  2. Goal or Outcome
  3. Key Performance Indicator
  4. Metrics

Here is an example of an applied taxonomy:

  1. Public Safety
  2. Reduce Violent Crime by 20% by 2015
  3. UCR (FBI) Violent Crime Data
  4. Number of Open Warrants, Number of Inmates, Crimes Committed, Domestic Violence Crimes, Visits of Probation Officers for Serious Offenders, Technical Violation of Parole or Probation, etc.