In order to get the maximum possible value from your data in Open Performance, we have provided some tips below:
Basic Data Structure: The data you use in Open Performance to set goals and build out dashboards should be structured so that each row of data represents a ‘record’ or ‘event’, corresponding with a specific time, while each column represents a ‘field’ that may be used to track a specific measure.
Here is an example to help illustrate this structure
Working with Categorical Data including Date/Time Categories: In addition to the basic structure, you may wish to combine multiple data points into a single dataset. For example, a 311 call center dataset may contain a record of each incoming call received in a given month, but you may want to build a goal on one specific subset of that data. By including a column to indicate the ‘type’ of incident being reported, you can use Open Performance’s in-line filtering to structure goals on one particular type of service.
Below we've compared survey results taken in both Sky City and nearby Lincoln for benchmarking
When building out the reported survey data for Sky City, we've choosen to filter out the Lincoln results when choosing the data source during goal page creation:
Similarly, you may want to use text-formatted columns to represent time elements beyond than the basic time- and date-stamp - such as fiscal years (eg. FY2016), quarters (eg. Q3 2014), or reporting periods (eg. Spring 2015). This is particularly useful if you want to create supporting column, bar, or pie chart style visualizations to feature in the narrative as they require a ‘text-based’ field to represent labels.
Working with Aggregated Data
At times, you may also want to build a goal based on a simple sum (or average) of events that take place during a certain time frame. In this case, it’s helpful to structure your data so that you have a field available to apply a count/sum function. This could be as simple as adding column (set to datatype: Number) where you record a 1 or 0 indicating whether that field should be counted when calculating a particular goal.
For example, if you have a public safety goal of having officers attend 350 community meetings over the course of the year, you could quickly tally your progress towards that goal by applying the ‘sum’ function on a ‘count’ field associated with each recorded visit.
Open Performance allows you to create measures and goals on data exactly as it is imported into the system. One consideration to keep in mind with regard to tracking aggregated measures (such as monthly transit riders) is the need to import your data in the format you wish to visualize it. For example, if you would like to track monthly incidents of crime, you will need to import a dataset for that goal that contains your crime data with monthly counts.