Abstract
Data-driven decision making is central to improving success of children. Actualizing the use of data is challenging when addressing the social, emotional, and behavioral needs of children across different types of early childhood programs (i.e., early childhood special education, early childhood family education, Head Start, and childcare). Exploring tools that are available through cloud computing, within systems that are being increasingly adopted by education agencies, may be advantageous. This article will describe a research-based approach to enhancing the capacity of early childhood programs to gather and use a data-driven, team-based problem-solving model that relies on cloud computing tools. A description of a cloud-based knowledge management system for behavioral incident data will be provided along with descriptions of its use by a group of early childhood programs seeking to enhance their efficiency and effectiveness when addressing challenging behavior.
Multi-tiered systems of support (MTSS) provide frameworks to guide the instructional decisions and problem solving of educators (i.e., Sugai & Horner, 2009). Within MTSS, the availability of ongoing child assessment and progress monitoring data is described as a critical feature (Buzhardt et al., 2010; Sugai & Horner, 2009). For some time, there was an assumption that if tools were created to gather information about children’s proficiency with and progress in developing essential learning targets, educators would be equipped to effectively engage in two essential components of instruction, data-driven decision making and ongoing problem solving (Hoogland et al., 2016; Marsh & Farrell, 2015; Todd et al., 2011). Unfortunately, multiple researchers have demonstrated that despite the availability of tools for gathering data, those data do not always lead to decision making about instructional practices that are in service of improving child outcomes (Heritage, Kim, Vendlinski, & Herman, 2009; Olah, Lawrence, & Riggan, 2010).
The purpose of this article is to explore one approach to trying to address the problems associated with data-driven decision making in early childhood through use of cloud-based tools for gathering and summarizing data. To begin that exploration, a brief description will be provided that grounds the work in critical efforts to address the social, emotional, and behavioral development of young children. The complexities of the work in early childhood will then be described as a means for considering why existing tools may not be sufficient when promoting data-based problem solving. Then, a model for team-based problem solving will be described as a necessary approach to building knowledge about the use of data that meaningfully supports instructional decisions. Finally, a cloud-based knowledge management system will be described, along with the experience of a group of early childhood programs using the system, to demonstrate the potential benefits that that type of system may offer when the goal is to promote actual use of data, not just the gathering of data.
Introduction to the Problem
Data-Based Problem Solving Within Frameworks for Positive Behavior Support
Actualizing the use of data for effective decision making is a multifaceted issue. In addition to the barriers that are often expected during the installation of new practices, such as time and expertise (Anderson, Leithwood, & Strauss, 2010; Copland, Knapp, & Swinnerton, 2009; Cosner, 2012; Park & Datnow, 2009; Young, 2006), use of data for systematic decision making appears to present some unique challenges. The unique challenges are particularly salient when addressing the social, emotional, and behavioral needs of young children across the many different types of early childhood programs that exist (i.e., school readiness, early childhood special education, early childhood family education, Head Start, and center-based childcare). Program-Wide Positive Behavior Intervention and Support (PW-PBIS; Frey, Park, Browne-Ferrigno, & Korfhage, 2010) and the Pyramid Model (Fox, Dunlap, Hemmeter, Joseph, & Strain, 2003) offer important frameworks to support decision making about implementation of practices that effectively address young children’s social emotional learning. Broadly, these frameworks represent MTSS with the specification of practices that are organized along a continuum (Fox et al., 2003; Sugai & Horner, 2009). The continuum ranges from practices that are implemented universally as a foundation of success for all children, to practices that augment the universal supports to promote skill development and success needed by some children, to focused interventions that systematically target needs of specific children (Fox et al., 2003; Frey et al., 2010). As frameworks within MTSS, the necessity of routinely gathered data that inform a protocol for problem solving and action planning is often discussed (Sugai & Horner, 2009), but not yet meaningfully addressed. In early childhood, gathering, summarizing, interpreting, sharing, and collaboratively action planning around data about children’s social, emotional, and behavioral performance within real-world programs is complicated by both the systemic structures and the still emerging data tools.
Complexities of Early Childhood System When Promoting Use of Data for Problem Solving
Service delivery models and educational settings that set the context for addressing young children’s social, emotional, and behavioral learning have become more complex. The Office of Special Education Program’s (OSEP) 2015 reporting showed that approximately 66% of preschoolers receiving Part B Early Childhood Special Education (ECSE) services do so in regular early childhood programs (up from 42.5% in 2012), while 23% of preschoolers receive services in separate classes, and the remaining 11% receive services through some combination of specialized settings and other types of environments (U.S. Department of Education, 2015). Discussions of PW-PBIS and the Pyramid Model include examinations of how the diversity of early childhood systems impact implementation of critical elements of each framework (Frey et al., 2010; Hemmeter, Fox, & Doubet, 2006; Johnson, 2017). Data-based problem solving is considered a critical element of both frameworks given the explicit and visible information that may be used to assess usefulness, effectiveness, and efficiency of a practice intended to address the social, emotional, or behavioral needs of children (Coffey & Horner, 2012; Mathews, McIntosh, Frank, & May, 2014; McIntosh et al., 2013). Explicit suggests that data being gathered are clear and leave nothing to be implied, and visible suggests those data are presented in ways that allow individuals to see and make meaning of the data. With more complex service delivery models and settings, gathering explicit data that are able to be made visible to all stakeholders may be more challenging for early educators than their K-12 counterparts.
To demonstrate the challenges, consider a large urban school district that has adopted the Pyramid Model for implementation throughout their early childhood services, though the ECSE program within the district is providing most of the leadership and fiscal resources. The school district provides ECSE services to approximately 3,000 children aged 5 years and younger. Of those children, 17% receive services in one of 15 separate special education classrooms. Just over 55% of children receive ECSE services in regular early childhood programs that are distributed to different locations throughout the school district. These services and programs represent a combination of eight classrooms that are co-taught by ECSE teachers and early childhood teachers supported through Early Childhood Family Education, as well as 15 School Readiness, 25 Head Start, and 30 childcare classrooms in which an ECSE teacher provides only itinerant services. Each type of classroom is subject to different administrative leadership and operating structures (Greenwood et al., 2011; Johnson, 2017). In this example, the gathering of explicit data to monitor children’s social, emotional, and behavioral performance on a regular basis across these varied early childhood programs within the district requires approximately 93 classrooms to adopt the same tool for gathering and reporting data. Though a tool exists to facilitate that process (Fox, Binder, Liso, & Duda, 2010, in Fox, Veguilla, & Perez Binder, 2014), the visualization, interpretation, and sharing of those data once gathered highlights the significant challenges associated with data-based problem solving across early childhood programs. Many classrooms use a paper form that must be entered into a spreadsheet in order to be summarized and interpreted. To meet the call for regular review of data, the school district must develop a system to obtain all paper copies of the data sheets from all classrooms spread throughout the district, enter data from each paper copy into an electronic database or spreadsheet, create summary tables and charts that present data in ways that are meaningful for problem solving, and share the data summaries with the problem-solving team and classroom teachers on a regular basis. The complexities, inefficiencies, ineffectiveness, and lack of usability of this approach within efforts to enhance team-based, data-driven problem solving should be clear.
Exploring tools that are available through cloud computing as part of systems that are being increasingly adopted by education agencies (i.e., Google and Microsoft 365; Gonzalez-Martinez, Bote-Lorenzo, Gomez-Sanchez, & Cano-Parra, 2015) may be advantageous, particularly given the need to work across different types of early childhood programs. That said, though the availability of tools that are cost-effective and efficient may facilitate gathering data across early childhood programs, alone, those tools will likely not facilitate actual use of data for decision making. Other challenges associated with use of data for decision making involve appropriate knowledge and skills needed to develop critical questions to be explored through a data system, identification of indicators within the data system that provide information related to those questions, knowledge of when the information provided is indicating that there is a problem, and ability to then identify solutions to address problems (Cosner, 2012; Marsh, Pane, & Hamilton, 2006; Means, Chen, DeBarger, & Padilla, 2011; Olah et al., 2010; Supovitz & Klein, 2003). In response to these challenges, a research-based approach will be described for enhancing the knowledge and skills of early educators to use data through team-based problem solving that relies on use of cloud computing tools. In describing this approach, an organizing model will be presented to offer an integrated view of how tools for gathering data play an essential role in contributing to usable knowledge in the context of team-based problem solving. With that, cloud computing tools will be explored as a means for gathering data and generating knowledge that is usable by teams when problem solving across varying types of early childhood programs.
Enhancing Knowledge and Skills Through Team-Based Problem Solving
Extending recommendations made by Marsh and Farrell (2015), efforts to enhance the capacity of education systems to meaningfully use data for decision making may benefit from (a) coaches with content area expertise who are able to guide problem solving and enactment of solutions at a local classroom or program level, (b) data coaches or analysts who have a more singular focus on interpreting and using data, and (c) teams that are comprised of content and data coaches as well as others who seek to make systemic improvements based on knowledge gained through data. A substantial set of literature exists for how to prepare teams with the skills needed to use data within a data-based problem-solving process (Alonzo, Ketterlin-Geller, & Tindal, 2007; Deno, 2005; Johnson & Reichle Monn, 2009; Newton, Horner, Algozzine, Todd, & Algozzine, 2012). To enhance the efficacy of that preparation, a systematic Team-Initiated Problem Solving (TIPS) model with accompanying professional development was created to provide teams with a concrete structure for engaging in each step of a data-based problem-solving process (Newton et al., 2012; Todd et al., 2011). Todd et al. (2011) provides a description of the conceptual foundations for the TIPS model while Newton et al. (2012) and Algozzine, Horner, Todd, and Newton (2016) describe the operationalized components of TIPS to support implementation and monitoring by teams. Figure 1 displays an organizational model that includes the steps associated with TIPS when it is incorporated into a model for promoting data use (Bertrand & Marsh, 2015; Marsh & Farrell, 2015). As portrayed in Figure 1, data remain an essential driver of the TIPS problem-solving process. However, to more fully represent the nuances of what is necessary for data to effectively contribute to a decision support data system (National Implementation Research Network [NIRN], 2016), there is recognition that though the gathering of accurate data that are linked to key data elements creates information for teams, it is not until those data are summarized and interpreted that usable knowledge is generated and actionable items may be identified (Brawley & Stormont, 2014; NIRN, 2016).
Organizational model incorporating TIPS into a model for promoting data use.
The work of several researchers highlight barriers to data use when educators must summarize and interpret data without support (Gummer & Mandinach, 2015; Means et al., 2011). There is some evidence that these barriers may not be entirely attributable to a lack of value or perceived importance of data. When examining the perceptions of early educators regarding general data practices, most rated use of data as very important for programmatic decisions, though using academic and behavioral data to monitor the class as a whole was rated as less important (Brawley & Stormont, 2014). Brawley and Stormont (2014) also found that despite early educators reporting value for having data, there is infrequent data collection and even less frequent, and less valued, creation of visuals that graph data. In combination, infrequent data collection and visualization of data through graphs diagnosticate an ineffective system for data-based decision making (Hojnoski, Caskie, et al., 2009; Hojnoski, Gischlar, & Missal, 2009). In addition, though early educators recognized the importance of having data that were readily available to support communication with administrators and families, their systems for gathering data did not facilitate efficient sharing of those data in ways that are meaningful to others (Brawley & Stormont, 2014).
There is a concerning disconnect between the importance early educators give to having data that may be meaningfully shared with others and the practices they use for gathering and summarizing data. With recommendations that collected data should be made regularly available in summary format for interpretation and use (Brawley & Stormont, 2014; Ingram, Louis, & Schroeder, 2004; Sandall, Schwartz, & Lacroix, 2004; Schwartz & Olswang, 1996), there are aspects of cloud computing that should be explored as a means for enabling more effective use of data. Cloud computing is defined by groups of computers that provide on-demand resources and services (i.e., Google, Microsoft OneDrive and 365) through a network, like the Internet, such that software does not need to be installed on devices in order to access those resources and services (Sultan, 2010). When used in education, cloud computing is thought to improve efficiency, reduce costs, and enhance convenience (Gonzalez-Martinez et al., 2015; Sultan, 2010). The conceptualization and application of cloud computing in education support an evolution from information and communication technology that pushes out information, to a knowledge management system that facilitates the creation and sharing of knowledge (Hoong & Lim, 2012). As described by Anupan, Nilsook, and Wannapiroon (2015), knowledge management systems are supported by (a) Internet access, (b) tools for gathering information, (c) database management systems that are accessible through the network, (d) expert systems that are continuously available to create knowledge, and (e) tools for communicating and collaborating within an organization. Despite many arguing that cloud computing offers important opportunities for educators to access and share data in time and cost efficient ways, much of the potential that exists for use of cloud computing tools as knowledge management systems has yet to be realized in school systems (Sultan, 2013). Sabi, Uzoka, Langmia, and Njeh (2016) describe one important factor in educators’ decisions to adopt and use cloud computing tools is the perceived usefulness of the system as educators balance the benefits that may exist with effort required to use the system.
A Description and Demonstration of a Cloud Computing System for Problem Solving
The purpose of the remainder of this article is to describe a knowledge management system that is currently being pilot tested in a variety of early childhood programs associated with 15 school districts in one Midwestern state. The pilot test begins the process of exploring the feasibility of a cloud-based system with a focus on how acceptable it is to early educators and their teams (Bowen et al., 2009). According to Bowen et al. (2009), the following are key aspects on which to consider the acceptability of a new innovation, (a) general satisfaction, (b) intent to continue use, (c) perceived appropriateness, (d) fit within organizational culture, and (e) perceived positive or negative effects on the organization. As the knowledge management system is described, the experience of one particular school district will be discussed relative to these key aspects of acceptability. Though the system is being pilot tested in 15 school districts, many are still very early in the process of installing use of the knowledge management system into their programs. In addition, the development process for the knowledge management system represents a community-based participatory research approach. As such, the descriptive feedback and shared experiences between the educators of one school district and the researcher offer many salient demonstrations of the potential benefits when cloud computing tools are leveraged to promote team-based problem solving.
Overview of the Knowledge Management System for Behavioral Incidents (KMS-BI)
The KMS-BI was designed in direct response to an expressed need of educators in early childhood programs that had been implementing a PBIS framework. The need was to have data about the occurrence of behavioral incidents that could be easily summarized, interpreted, and shared across their complex systems, while not requiring additional fiscal resources. The need for the system to not require additional fiscal resources was an essential ask of school districts whose collaborative partners for implementation of the Pyramid Model were often child care centers and Head Start programs. Despite shared goals, requiring funds for ongoing access to any type of online tool was considered to be a barrier given distinctly separate funding streams, limited resources, and different priorities for resource allocations across different types of programs (ECSE, Head Start, Early Childhood Family Education, and Childcare). By attending to this issue in development of the KMS-BI, the goal was to create a system that might be more acceptable for use given a better fit within the organizational culture and means by which early educators work. To that end, information was gathered from school districts that had been implementing the Pyramid Model for at least 1 year to understand their access to different cloud computing tools that may be useful in constructing a knowledge management system to support their work.
Three cloud computing tools were selected to construct the KMS-BI for use in early childhood programs. To again maintain a goal of building a system that would be acceptable for use in early childhood programs, tools were selected that programs already expressed satisfaction with and perceived to be appropriate given their prior experiences with the tools. First, Google Forms, a survey tool within the Google Drive suite, is used to create an online form that is filled out to report data about a behavioral incident. The content of the form was adapted from the Behavior Incident Reporting System (BIRS) originally developed by Fox et al. (2010). Table 1 provides a summary of the items and response options included on the Google Form. Second, Google Sheets, a spreadsheet tool within the Google Drive suite, is used to automatically gather and store any behavioral incident data reported using the Google Form. Google Sheets receives data from Google Forms in real time, securely stores the data, and is easily shared with team members or administrators who may need access to the original data. A Google template, which includes the Google Sheet with the linked Google Form, was created for teams to personalize for use across their programs. That template is available for use by others at: https://docs.google.com/spreadsheets/d/1GHwJlEOc3ECO1qNbsIJSMiFotyU3YiW_4vo6Ha8jbZ8/edit?usp=sharing. To simplify data entry and facilitate data summarizing, teams personalized the form to their program by entering location, teacher, and child numbers or initials on the form. Despite enhanced security and privacy measures being used by Google and Microsoft to protect information gathered using their tools (Sultan, 2013), teams were encouraged to limit the use of information that could make the data readily identifiable outside of their programs. Finally, the third component of the KMS-BI is an Excel Online workbook that is used on Microsoft OneDrive. Similar to Google, OneDrive facilitates secure sharing of information that is stored in the cloud. The Excel Online workbook allows teams to copy and paste data that are received in their Google Sheet into the workbook that will then create automatic summaries and visual displays for the team to use during problem solving. The use of Excel Online through OneDrive allows problem-solving teams and program administrators to access the information from any type of device without needing specific software or a specific type of operating system. For the KMS-BI, the original Excel Online workbook was created by the lead author with embedded formulas and visual displays to support efficient access to data that are summarized and visualized for use by problem-solving teams. Given the community-engaged approach in development of the KMS-BI, as teams use the system, summaries are changed and adapted by the lead researcher to support the ways in which teams are trying to meaningfully use the data. The Excel Online workbook with example data is available for downloading and use by others at: https://1drv.ms/x/s!Av4jJ8fh9Jqdhl-8E9mEvDiwHI3Q. To again maintain a goal of building a system that would be acceptable for use, all early childhood problem-solving teams that are currently piloting the KMS-BI system worked with their respective school districts to ensure that the district administrators perceived the system to be appropriate given that they had already adopted, or would support adoption of, these cloud computing tools for the purpose described here.
Items and Response Options for Documenting Behavioral Incidents Using Google Forms.
Using KMS-BI to Turn Information Into Knowledge for Problem Solving
Within the TIPS model (Todd et al., 2011) that is included as part of Figure 1, problem solving begins with a team’s ability to identify that there is a problem. The ability to identify that there is a problem that needs to be solved engages an interpretive process that begins with first noticing data (Coburn & Turner, 2011). The intent of a cloud-based data system, such as the KMS-BI, is to help the data that are gathered through the system become more “noticeable” because the data are automatically summarized and key aspects are highlighted. Furthermore, data summaries and visualizations are intended to align with information needed to precisely identify problems and explore the critical questions team members and administrators should be seeking to examine in relation to implementation of the Pyramid Model (Fox et al., 2014). Precisely defined problems are those that include reference to the what, where, who, when, and why of the problem, based on explicit data, so that problem solving is appropriately targeted to the identified need (Algozzine et al., 2016). In addition to identifying needs that are revealed through data, critical questions may guide teams through explorations of their data such that they are able to notice particular aspects of the data and more efficiently identify problems. The alignment between the explicit data being gathered and the automatic data summaries and visualizations for examining critical questions is where information may be transformed into usable and actionable knowledge. What follows is a description of how the KMS-BI may support this transformation of data to support more effective problem solving at the program, classroom, and child level.
Usable and actionable knowledge for program-wide problem solving
Though the reporting of behavioral incidents for individual children naturally draw educators’ attention to problem solving for individual children, if a proper system exists to summarize those data and link those data with other key pieces of information about the program, a more robust data system will exist for teams to use for programmatic decisions. To facilitate more robust thinking about use of behavioral incident data, the data summaries and visualizations within the KMS-BI are informed by several critical questions that problem-solving teams should consider: (a) Are we effectively implementing good universal supports across the program such that certain groups of children are not disproportionally receiving incident reports? (b) Are there certain program locations, classrooms, or times of day when incidents happen more often? and (c) Are there certain classrooms in which it is known that incidents are occurring, yet there are no incident reports?
To explore data that are associated with disproportional receipt of incident reports, the key link in the data system is between not only which classrooms and children are in the system because a behavioral incident has been reported, but also to knowing how many classrooms and children are part of the program. Extending the recommendations of Boneshefski and Runge (2013), enrollment numbers that are also linked to demographic information, such as gender, race, English language learner status, and special education services status, creates important opportunities for programs to explore the presence of disproportionate practices within their program. Figure 2 displays an example of how this information is summarized in the KMS-BI to provide programs with knowledge they may be able to use and act on if a problem is identified. In the example displayed in the top panel of Figure 2, the embedded formulas and visual displays that are included with the Excel Online workbook create visual color coding of the values to help the problem-solving team to notice if there is a problem. The use of this work within a cloud-based system allows teams and administrators to have immediate access to actionable knowledge that is shared in a transparent way to promote continuous improvements and, hopefully, opportunities to highlight successes over time.
Example KMS-BI data summaries for program problem solving.
Similar opportunities exist when other data, such as locations or activity times associated with incidents within a building, are summarized for a program on a monthly basis (see Figure 2, bottom panel). In this case, the manner in which the Excel Online workbook is organized allows teams to sort the data by the building or type of program that submitted the data, such as all classrooms clustered in a particular childcare center or elementary school. By having these types of summaries available, the problem-solving team is able to efficiently review incidents across all of the centers, buildings, or program types that are intended to receive their support. Furthermore, by leveraging cloud-based tools, the team is able to easily share the data summaries and engage in collaborative problem solving with administrators and educators in various settings.
The experience of one school district
The KMS-BI was adopted by an urban school district for use by all early childhood programs that were associated with the school district. This includes 36 educators representing 64 different class sessions (i.e., one, two, or three morning/afternoon sessions per week, full-day sessions) across five different types of programs (self-contained early childhood special education, inclusive co-taught classroom, school readiness/pre-K, early childhood family education, and center-based childcare), located in seven different buildings, with an average total monthly enrollment of 728 children across those programs. A single, district-wide problem-solving team exists to support the needs of educators and children.
The district problem-solving team adopted the KMS-BI after several years of having teachers fill out paper copies of a behavioral incident reporting form. As a reflection of the acceptability of the system, there was immediate satisfaction with an electronic system that bypassed the need to gather paper forms and devote staff resources to entering data into a spreadsheet. To ensure that all educators had easy access to a means to complete the Google form that was used to record an incident, the team used a combination of links added to the home screens of iPads or tablets used in each classroom or posted Q-R codes that could be scanned within a classroom to access the Google form linked to that classroom. Though information provided by educators suggest this approach to be appropriate and a good fit, there was an unforeseen challenge in monitoring enrollments across the different types of programs. Entering accurate enrollment information was necessary so the team could examine data that were summarized in two ways that were a priority for this urban district, the percent of children with behavioral incidents and risk ratios. Obtaining this information required the district data manager to retrieve data from multiple sources each month. This not only identified an issue with the feasibility given the different data systems needed to monitor enrollment, but it also highlights a broader opportunity to enhance early childhood longitudinal data systems as different types of early childhood programs engage in collaborative efforts such as this.
Despite the challenges with obtaining enrollment information, several additional summaries are automatically generated by the Excel Online workbook to support program-wide decisions the problem-solving team may need to make. For example, the team is able to select a particular building and see data that are summarized across all educators within that building. This feature of the KMS-BI supports teams in more precisely identifying if there are problems that need to be addressed through coaching or other professional development opportunities given a preponderance of incidents in a particular location or within a particular type of activity. In the case of the district described here, these summaries allowed the team to identify that in one building, there was a need to offer supports that targeted circle time/large group activities given that most incidents were occurring during those types of activities (see Figure 2, bottom panel for example). In another building, transitions were most problematic, and in another building, outdoor play time was associated with the greatest number of incidents. The knowledge gained by this team by having access to these types of summaries allowed for more targeted and meaningful support to be offered efficiently and effectively.
Usable and actionable knowledge for classroom problem solving
When the KMS-BI is used to summarize and visualize data for classrooms, there are additional opportunities to engage in collaborative problem solving with staff for a specific classroom to address a specific problem. There are several critical questions that problem-solving teams should consider that were used to guide how summaries were created in the KMS-BI: (a) Are certain activity times particularly challenging? (b) Are there patterns for incidents involving specific adults or other people? (c) Is the frequency of incidents changing over time or is it stable? (d) Are there patterns for the form and function of behavioral incidents that may suggest a need for a universal or secondary group intervention? and (e) Are only a restricted set of strategies being used in response to incidents that occur? With a goal of again creating more efficient uses of resources, teams that are able to identify problems based on summaries, such as those associated with each of these questions, as well as share those summaries with others, are able to move more quickly to effective selection and implementation of interventions and supports given the precision with which problems may be identified.
Figure 3 displays part of the overall classroom summary page that may be used by teams for a summary of the most relevant elements of the behavioral incidents reported for the classroom. These types of visual displays allow teams, coaches, and classroom staff to efficiently explore data about not only the frequency and form of behavioral incidents, but also consider the environmental arrangements associated with those incidents to contribute to more precise identification of problems (Todd et al., 2011) and development of action plans. In addition, the monthly summaries allow teams, coaches, and classroom staff to explore changes over time relative to implementation of interventions or changes in environmental arrangements that may be associated with a new problem that requires problem solving.
Example KMS-BI data summaries for classroom problem solving.
The experience of one school district
In the district described earlier, there are two aspects of using the KMS-BI for classroom-based problem solving that were particularly noteworthy. First, in direct response to feedback from this team and several others, there needed to be an efficient way for an educator to request assistance following an incident such that the team would be alerted to a need to follow up and offer support in a timely manner. To address this need, the components of the KMS-BI were augmented. The Google Form was adapted so that the response screen that an educator sees after submitting a behavioral incident includes a hyperlink to an email address of a team member so the educator can quickly send an email requesting assistance. An item was also added to the Google Form so that during the process of reporting a behavioral incident, the educator was asked if assistance was needed. This item was then incorporated into the summaries generated in the Excel Online workbook so that the problem-solving team was able to see, at their weekly meetings, which teachers requested follow-up. The team used this information to not only develop plans for how to allocate their time each week, but also to provide district administrators with information about the degree to which educators across different programs were seeking additional supports that may require certain resources. In again aligning development of the system with aspects that promote acceptability, the ability for teachers to easily request assistance and the team to be more responsive to requests was perceived as having a very positive effect on the organization’s ability to support the needs of its workforce.
The second aspect of the KMS-BI system that was noteworthy for this district was the use of the classroom summary page (see Figure 3 which displays a portion of the page) that provided summaries of all of the incidents reported within a particular class session for the purpose of promoting reflection on overall class needs as opposed to focusing on only a specific child. This approach to summarizing the data and reflecting on it was new for this team and the educators they support. Though the team made it a priority to send each teacher a weekly summary of the data from each class session they were responsible for, they quickly realized a need to support educators’ ability to look at the summaries, make interpretive statements, and use their interpretations to action plan. To augment the reflective questions the team provided with the summaries each week, the team also planned a professional development opportunity in which small groups of educators spent time rotating through different classroom data summaries to discuss different interpretations and practice exploring action items based on each interpretation. This proved to be a valuable approach to not only enhancing educators’ ability to use data, but also to enhancing their reporting of incidents given a newly developed appreciation for the type of action planning that becomes possible when data are summarized in a way that supports an iterative problem-solving process.
Usable and actionable knowledge for child specific problem solving
The precise identification and definition of problems may facilitate more efficient and effective problem solving by the team (Todd et al., 2011). As described by Newton et al. (2012), precision in the problem-solving process includes five elements: (a) what specific behavior or skill is discrepant from expectations, (b) who this is a problem for in terms of an individual student or group of students, (c) where within the building or program this problem is occurring, (d) when during the day this problem is occurring, and (e) why staff believe this problem may be occurring based on environmental conditions that maybe setting the occasion or reinforcing the occurrence of the behavior. The availability of a data system that facilitates precision in the problem-solving process is a necessary attribute to effective design and implementation of solutions, particularly those needed to address the behavior of specific children (Newton et al., 2012; Todd et al., 2011). In the design of the KMS-BI for use in early childhood settings, in addition to considering how information would be summarized to promote precision problem solving, there were two additional critical questions that informed the design of the summaries in order to help teams to notice certain data: (a) Are certain children involved in behavioral incidents with greater frequency than others such that problem solving is necessary? and (b) Does the frequency of incidents change over time when supports or new intervention strategies are introduced?
The summaries and visual displays that are incorporated into the KMS-BI for use in child specific problem solving include an overall summary of data for a child, similar to that displayed in Figure 3. These summaries may be used directly by the team to identify and define problems with precision and collaboratively plan interventions with the relevant staff. Though these displays are arranged to provide monthly summaries to look broadly at changes over time, two additional pages within the Excel Online workbook support targeted planning of child-focused interventions as well as weekly progress monitoring through data that are graphed and include a trend line to facilitate visual analysis. An example of an automatically created progress monitoring graph is presented in Figure 4. The advantages of leveraging cloud-based tools to gather and summarize data to support child specific problem solving further reinforce how information may be transformed into usable knowledge by (a) overcoming barriers associated with staff not having the knowledge, skills, or time to create the visual displays that are helpful to examining performance over time (Brawley & Stormont, 2014; Coburn & Turner, 2011); (b) providing an efficient means for gathering data that may be made available in near real time to support active decision making relative to implementation of interventions; and (c) facilitating buy-in and encouraging team-based problem solving across different types of early childhood programs by offering an efficient means of regularly sharing meaningful data to the staff and families who are spending the most time directly with children.
Example KMS-BI data summaries for child problem solving and progress monitoring.
The experience of one school district
The experience of the one school district when using the KMS-BI for child specific problem solving offered several lessons related to each of the advantages just identified. The automatic creation of the visual displays within the Excel Online workbook was essential to promoting use of the data in a manner that is timely and appropriate when the focus is on addressing the needs of specific children. To ensure that the problem-solving team was working with updated visual displays of data each week, a small amount of time from an educational assistant was used to copy new data from the Google Sheet that received all of the data as it was submitted by educators and paste it into the data entry tab of the Excel Online workbook. Once the educational assistant finished pasting the data into the workbook, by selecting “Refresh All,” any new data was automatically incorporated into the summaries and visual displays for the team to use when identifying and defining problems that warranted action planning. The team received guidance through a 1-hr webinar with the lead author to support their ability to use the various “slicers” to sort the data into targeted summaries and interpret those summaries in a meaningful way. Slicers are a term used in Excel to describe the buttons that are made available on a spreadsheet to select and display only certain subsets of data (i.e., all incidents for a given building, teacher, child, etc.).
As reported by the problem-solving team, the efficiencies created by use of the cloud-based tools that allowed the team to examine and share summarized data on a weekly basis made the system very acceptable to them through providing a significant benefit to their work in a district culture that called for data-based decision making. The team was able to immediately share knowledge with program administrators and educators across different types of early childhood programs to support identification of additional resources (i.e., coaching or behavior support planning) that may be needed based on the frequency of child specific incidents. To date, the team monitored and provided targeted supports for an average of 95 incidents per month involving an average of 18 children (meaning an average of 5.3 incidents per child for 2.5% of enrolled children). Each week, the team quickly viewed a full summary of reported behavioral incidents for each child (frequency, type of behavior, location, activity time, others involved, possible motivation, and strategies used) to support the team’s decisions about their follow-up actions. In many cases, there was some brief discussion about possible solutions and supports that the assigned coach referenced when sharing the data with a child’s educator. Another benefit of using the cloud-based tools was in allowing the coach to access the Excel Online workbook while meeting with an educator so that they were able to collaboratively discuss data for a specific child, examine environmental arrangements that may explain ongoing behavioral incidents, and monitor a child’s progress over time. On several occasions, the educators also shared the visual displays with families to promote communication, support, and shared problem solving in a way that facilitated consistency across all of the adults and environments that are influential to a child’s learning and development.
Moving From Exploration to Full Implementation of Data-Based Problem Solving
The work of several researchers highlights the importance of maintaining data-based decision making and problem solving as a goal (Hoogland et al., 2016; Marsh & Farrell, 2015), with a particular need to build capacity in early childhood programs if that goal is to be achieved (Brawley & Stormont, 2014). The complexities of the service delivery models and varying types of programs in which young children receive care prior to starting school contributes to a need to explore solutions that offer the least complex option that is compatible with the nuances of early childhood systems while providing direct benefit to educators, families, and children (Johnson, 2017; Sabi et al., 2016). Sabi et al. (2016) discuss this need specific to the adoption of cloud computing in educational settings. There is growing consensus that cloud computing tools offer more cost-effective options for educators who are able to gain expanded access to research and collaborative opportunities that leverage high-performing technology infrastructure, despite limited resources (Gonzalez-Martinez et al., 2015; Sabi et al., 2016). This consensus is reinforced by findings that the specific cloud computing tools that form the foundation of the KMS-BI (Google Forms, Google Sheets, and Microsoft OneDrive) are already used broadly in education settings (Gonzalez-Martinez et al., 2015) for a variety of applications (e.g., see Bennett & Pence, 2011; Bonham, 2011; Herrick, 2009; Nevin, 2009; Sultan, 2010).
Application of cloud computing tools, such as those described for the KMS-BI, for use in team-based problem solving that relies on the TIPS model may contribute to addressing several issues that are considered central to promotion of data-based problem solving. Based on an extensive review of the literature, Hoogland et al. (2016) identified nine themes that they deemed to be “prerequisites” for data-based decision making: (a) collaboration, (b) leadership for data-based decision making, (c) a culture of data-based decision making, (d) facilitation by means of time and resources, (e) teacher knowledge and skills, (f) external factors associated with accountability and policy, (g) professional development, (h) data use attitude, and (i) assessment instruments and processes. A description of the KMS-BI was provided in this article as a demonstration of how the capacity of early childhood programs may be enhanced through adoption of cloud computing tools. The tools that form the foundation of KMS-BI are intended to promote collaboration through easy and efficient sharing of information across devices, software platforms, and especially across types of early childhood programs. Easy sharing combines with tools that may be adapted by individuals with expert content knowledge so that the knowledge and skills that need to be developed in leaders and educators may focus on actual use of the data to generate knowledge rather than how to create a graph that depicts certain information. As personnel within early childhood programs are able to see visual data displays that reflect their efforts to implement effective strategies and improve outcomes for children, it becomes easier to build a culture of data-based decision making that values data. Demonstrating efficiency and offering opportunities to promote efficacy, cloud computing tools may be a viable solution to not only the resource limitations of early childhood programs, but also to the need to work collaboratively and openly across a variety of different types of early childhood programs striving to achieve similar goals.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.