Date of this Version
When conducting workforce analyses, it can be valuable to capitalize on data from both human resources (HR) databases and child welfare (CW) information systems. Each can be useful on their own, but additional information can be learned when the two types of data are connected. There are two primary uses of combined HR and CW data: 1. CW data can serve as objective measures of employee job performance, which is one of the most important workforce outcomes to measure, particularly when examining the effectiveness of various workforce decisions and practices. 2. CW data can provide useful contextual information for studying workforce outcomes. More specifically, case characteristics and events (e.g., degree of complexity, unsuccessful reunifications) may influence workers’ job attitudes (e.g., engagement, job satisfaction), sense of well-being (e.g., burnout, stress), and retention. Child Welfare Data as Measures of Performance Child welfare information system data are often used to calculate a number of metrics that serve as indicators of agency performance (e.g., for Child & Family Services Reviews). Though supervisors and managers examine worker- or unit-level data to guide and monitor performance, there is no universal set of worker-level performance measures, especially for day-to-day performance (vs. more long-term outcomes). There are several potential reasons for this: • Performance measures depend on the employee’s role; different roles involve different behaviors and, therefore, warrant different performance measures. • Performance expectations vary across agencies. Different case practice models and policies result in different work behaviors and, therefore, warrant different performance measures. • Child welfare information systems also vary across agencies. Though standardization is increasing, there is still significant room for variation. Different fields and system functionality lead to different data and, therefore, different possible performance measures. To capitalize on child welfare information as a source of performance indicators, it is recommended that supervisors and SACWIS trainers be consulted for guidance on the best types of information; they will know best what data are useful for monitoring and managing performance. Here are some questions and tips for eliciting information from those subject-matter experts: • What kinds of system data are red flags for performance issues? Think of current or past poor performers—what evidence was there of that poor performance in the information system? • What kinds of system data best represent important performance differences in workers? Data that do not vary much are not likely to be the most useful indicators of individual differences in performance, which is what you are looking for. • Is there system functionality that alerts supervisors to overdue tasks or missing items, such as through tasks lists, action items, or automatically elevated overdue ticklers or alerts? Is the history retained after the issues are resolved? • What evidence is there of supervisors’ responses to things submitted by workers (e.g., requests, forms, plans, recommendations, reports)? Potential responses include approving, requesting revisions, not approving, or overriding. • Note that some of the answers will be things that require supervisor judgment of data entered in the system (e.g., narrative quality). Such things can offer very useful information, but they will require much more input and effort from supervisors. If you are looking for data that can be pulled from the system and used as is, ask supervisors to only suggest data that do not require additional interpretation or judgments about quality. • Be careful about case dependencies—don’t measure something that isn’t always necessary for a given case or type of case, unless you plan to account for that (e.g., performance measures related to parentchild visitation only apply to cases where children are in out-of-home care). • Don’t count tasks that are automatically accomplished by the information system (e.g., auto-generated correspondence or notification) • Potential process measures fall into categories such as contacts (e.g., initial and ongoing), information gathering (e.g., relative searches, tribal membership status), collaboration, communication (e.g., sending required notifications or letters), planning (e.g., safety plans, case plans), decision making, service provision (e. g., child health examinations), reporting (e.g., court reports), financial, or timeliness (e.g., of task completion or documentation of task completion). • In general, it is always important to select performance measures that… o Matter—measures that are meaningful, important, and relevant, rather than measures for which data are merely readily available or easy to understand o Vary—if performance is stable across people, time, etc., there’s not much to explore o Are under employees’ control—focus on measures that are largely under a person’s control or over which they have at least a fair amount of influence; if outcomes are beyond their control, focus on process measures that are most related to outcomes o Can be trusted—data are reliable and valid • To ensure the data will be useful and can be provided for analysis, it is also important to work with business or systems analysts who know the data well and are able to do extractions. Using Child Welfare Data as Measures of Performance Performance measures that are derived from child welfare information system data can be combined with HR data and used in any metric or for any purpose that involves performance or performance improvement. Segmentation can also be used to explore differences between different groups of people. Examples metrics include: • Functional (poor performers) vs. dysfunctional turnover (good performers) • Quality of hire • Training evaluation • Evaluation of any other intervention, both HR and case practice changes Child Welfare Data as Job Context Information Though there are many individual and organizational characteristics that are thought to affect workforce outcomes, there may also be role or case-related characteristics that might affect workforce outcomes. • Child welfare data can sometimes be needed to examine more specialized characteristics of workers’ specific roles, such as function (e.g., intake/hotline, investigation/initial assessment, ongoing, voluntary services, alternative/differential response, adoption). • Within job function, case characteristics can also be considered as they relate to workforce outcomes. Examples of factors that increase case complexity include: o Child Characteristics Disabilities/special needs Juvenile delinquency Many children o Caregiver/Family Characteristics Mental and/or behavioral health issues Sexual abuse Domestic violence o Placement Needs Large sibling groups Siblings in different placements • Case events may also be important factors that affect workforce outcomes. Examples include: o Placement disruption o Repeat maltreatment o Maltreatment in foster care o Unsuccessful reunification or adoption o Child fatality Using Child Welfare Data as Job Context Information Role and case-related information can be combined with HR data and used in any metrics that involve segmenting by role (e.g., tenure or turnover by role) or in any analyses that examine or attempt to account for the role that job function or case factors play in workforce outcomes. These factors may also be connected to employee attitudes (e.g., engagement, job satisfaction, stress) gathered through survey data.